From d65fbc62849022a23bae60f5c9e14d11b143f248 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Mon, 2 Nov 2026 09:13:13 -0800 Subject: [PATCH 01/77] add xarray export plan Co-Authored-By: Claude Opus 5.5 --- xarray-export-plan.md | 188 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 188 insertions(+) create mode 100644 xarray-export-plan.md diff --git a/xarray-export-plan.md b/xarray-export-plan.md new file mode 100644 index 00000000..39214d25 --- /dev/null +++ b/xarray-export-plan.md @@ -0,0 +1,188 @@ +# xarray export plan + +Follows up dead-code audit item 5 (`dead-code-audit.md`). Status: design +**decided** (2026-11-02), implementation not started. Nothing in `mixins.py`, +`protocols.py` or the reverse half of `attrs_xarray.py` gets deleted, and +no tests get dropped, except as the phases below say. + +Scratch scripts used for the comparison (rerun them to see current +behavior): +`/private/tmp/claude-502/-Users-wbonelli-dev-pyphoenix-project/3ec4ddc6-e42e-49eb-bc12-ce39a377ee2c/scratchpad/compare.py` and +`/private/tmp/claude-502/-Users-wbonelli-dev-pyphoenix-project/3ec4ddc6-e42e-49eb-bc12-ce39a377ee2c/scratchpad/compare2.py`. +They build a 1x4x5 DIS GWF model (NPF, IC, CHD, WEL, RCHA, OC, IMS, TDIS) +and print `pkg.to_xarray()` next to `attrs_to_dataset(pkg)`, plus both +round trips. + +## Decisions + +1. **`to_xarray()` is the complete, lossless export** on every component. + Packages give an `xr.Dataset` and contexts (models, the simulation) an + `xr.DataTree`. `from_xarray(to_xarray(x)) == x`. +2. **The analysis view moves to the `grid` xarray accessor** + (`flopy4/mf6/accessor.py`), computed from the complete export: + `pkg.to_xarray().grid.dense(sparse=, duplicates=, grid=)`. It is no + longer a second export path on the package. +3. **Grid dims and coordinates** in the complete export (`layer,row,col` / + `layer,cell` / `node`, with x/y/z), falling back to DFN dims (`nodes`, + `ncpl`, ...) when the grid doesn't resolve. A package looks the same + whether you ask it or pull it out of a model's tree. +4. **Period arrays as given**, along `_period`, as `attrs_xarray` + does now. Filling forward is the accessor's job. +5. **The complete export is writable to netCDF.** It is distinct from + MF6's NetCDF input format (`flopy4/mf6/netcdf.py`, `NetCDFModel`), and + the docs must say which is which. +6. **Keep `protocols.py`; put the methods on `Package`/`Context` directly; + delete `mixins.py`.** The protocols are cheap, the SDD describes them, + and `isinstance(x, DataTreeConvertible)` is useful. The mixins are + two-line forwarders that sit awkwardly in the hierarchy (`Context` + subclasses `Component`). The reverse functions in `attrs_xarray.py` + stay and get finished. + +## Where things stand + +### What calls what today + +| Class | `to_xarray()` today | +|---|---| +| `Package` with griddata, period arrays or a period list | the view (`package.py`, `Package.to_xarray`), with `sparse`/`duplicates`/`grid` options | +| `Package` with none of these (IMS, ...) | `Component.to_xarray` (`component.py`) = `attrs_to_dataset` + a child-griddata merge that never fires | +| `Tdis` | its own `kper` dataset (`tdis_methods.py`) | +| `Context` (`Model`, `Simulation`) | `attrs_to_datatree` (`context.py`), already the generic layer | +| `Simulation` | the above + Tdis's `kper` vars merged into the root (`simulation_methods.py`) | + +`Package.from_xarray(ds)` is an *instance* method that sets stress period +data only, via `period_lists.from_dataset`. It accepts the dense and the +sparse forms. + +### Gaps in the generic layer (`attrs_to_dataset` / `attrs_to_datatree`) + +- **List data** (CHD, WEL, OC, ...) isn't encoded. `_stress_period_data` + goes into attrs as a live `PeriodRows` object. `dataset_to_attrs` then + raises `TypeError: unexpected keyword argument '_stress_period_data'`. +- **Item tables** (`Tdis.perioddata` as `ItemTable`) go into attrs as objects too. +- **Attrs aren't netCDF-safe**: `PosixPath`, `bool`, `ChildList`, + `PeriodRows`, `ItemTable`. +- **Empty child lists** (`api`, `drn`, ... `[]`) leak into a model's attrs. +- **Tree nodes are named by field and index** (`models0`, + `solutiongroup0`, `chd0`), not by component name. +- **`datatree_to_attrs(Simulation, tree)` doesn't recover concrete + types.** It rebuilds `models` as the abstract `Model`, with no packages. + Fields annotated with a base class or a union are never resolved to a + concrete class. +- **No coordinates**, and arrays sit on flat DFN dims. +- **Tdis appears twice** in `sim.to_xarray()`: as root vars, and as `/tdis`. + +### Already fixed (uncommitted on `develop`) + +`Package.to_dataarray` (and so the view) raised on structured DIS +(reshaping `delr`/`delc` onto the grid) and on any scalar griddata field +given before the package knew its grid (NPF's default `icelltype`, +`Ic(strt=10.0)` attached later). The fix fills constants to the field's +shape (a 0-d array if the shape is unknown). Arrays not on the cells get +one dim per shape name, via `_AXIS_DIMS` (`ncol` → `col`, `nrow` → `row`). +Tests: `test_to_xarray_constants_attached_later` and +`test_dis_to_xarray_axis_arrays` in `test/mf6/test_mf6_component.py`. +Should go in as its own small PR before phase 1. + +## Phases + +Each phase is roughly one PR. Per `CLAUDE.local.md`, verify with +`pixi run -e dev smoke` plus the test files touched; CI runs the rest. + +### Phase 1: make the generic layer complete (`flopy4/attrs_xarray.py`) + +Keep `Package.to_xarray` as it is in this phase. Work on +`attrs_to_dataset` / `attrs_to_datatree` and their inverses, and test +them directly. + +- **List fields** encode as the sparse table from + `period_lists.to_dataset(pkg, sparse=True)`: a variable per column + along an entry dim, with `per` and the cellid as coordinates. Decode + through `period_lists.from_dataset`. Map the private field name to its + init name (`_stress_period_data` → `stress_period_data`) both ways. +- **Item tables and other row lists** (`Tdis.perioddata`, OC's rows, + `packagedata`/`connectiondata` in SFR/LAK/MAW/UZF): the same sparse + treatment, one variable per column along a `_row` dim. Check + whether `period_lists` already has a helper for non-period lists before + writing one. +- **NetCDF-safe attrs**: paths as str, bools as int8 with a flag in the + variable/attr metadata (or a documented convention, decide during + implementation), references as names (already the case), None + omitted, empty child lists omitted. The decoder reverses each. +- **Coordinates and grid dims**: reshape cell arrays to the grid dims and + add the cell-center/elevation coordinates. Share the code with + `Package.to_dataarray` / `_with_cell_centers` rather than duplicating + it. Keep DFN dims when the grid doesn't resolve. Record the grid sizes + (`nlay`, `nrow`, `ncol`, ...) and `nper` in attrs so the accessor and + the decoder have them. +- **Tree nodes**: named by the component's name. Every node carries a + type tag (e.g. `attrs["type"] = "gwf-npf"`, matching the namefile + ftype convention if that's simpler) so the decoder can pick the concrete + class for `list[Model]` and union fields. The tag replaces positional + `f"{field}{i}"` lookup in `datatree_to_attrs`. Check how dict-kind child + fields fit (the module docstring says they can't round-trip today). +- **Tests**: extend `test/test_attrs_xarray.py` with round trips on real + components: a list package (CHD with boundnames and aux), OC, TDIS, a + whole simulation (`datatree_to_attrs(Simulation, attrs_to_datatree(sim)) + == sim`, with concrete model and package types), and + `to_netcdf` → `open_dataset`/`open_datatree` → rebuild. The existing + synthetic `Leaf`/`Node` tests stay until phase 3, and the `Welg`/`Rcha` + round trips in `test/mf6/test_mf6_period_arrays.py` stay. + +### Phase 2: the `grid.dense()` accessor, and repoint `to_xarray` + +- Add `dense(sparse=False, duplicates="raise", grid=False)` to the `grid` + accessor (`flopy4/mf6/accessor.py`) on Dataset (and DataArray where it + makes sense). It: + - densifies list tables onto the grid with the `duplicates` policy + (`period_lists`' existing logic), + - fills period arrays forward over `nper`, + - is a no-op on arrays already on the grid. + + With `sparse=True` it returns the list table as is (the current + `to_xarray(sparse=True)`). It needs only what phase 1 puts in the + dataset. If it needs the package or the grid object, phase 1 is + missing something. +- Add the inverse, `grid.sparse()` (dense list variables back to a + table), or have `from_xarray` accept the dense form. Pick the one that + keeps `period_lists.from_dataset`'s current dense support working with + the least code. +- `Package.to_xarray()` becomes the complete export (`attrs_to_dataset`); + `Package.from_xarray` becomes a classmethod rebuilding the package + (`dataset_to_attrs`). The old instance method's "set spd from a + dataset" use moves to `Pkg.from_xarray(ds)` plus attaching, or to a + small helper. Decide during implementation, and check the editing + workflow in `docs/examples/period-data.py`. +- `Tdis.to_xarray`'s derived `kper` dataset (`perlen`/`nstp`/`tsmult`, + Timestamp `start_date_time`): it goes, unless the accessor or the time + API (`Simulation.time`) needs it. Then `Simulation.to_xarray`'s Tdis + merge goes too. +- Delete `Component._collect_child_griddata_datasets` and the merge in + `Component.to_xarray`. +- **Update callers**: `docs/examples/period-data.py` (+ the `.ipynb`), the + `to_xarray` example in `docs/dev/sdd.md`, and the tests in + `test/mf6/test_mf6_period_lists.py` and `test/mf6/test_mf6_component.py` + that use `to_xarray` options or expect the view's shape. Check + `flopy4/mf6/adapters.py` (it calls `attrs_to_dataset` directly) still + gets what it needs. + +### Phase 3: wiring and cleanup + +- `Context.from_xarray(tree)` classmethod (`datatree_to_attrs`), with + parents and dims rebinding through the normal constructors. +- `Package` satisfies `DatasetConvertible`, and `Context` satisfies + `DataTreeConvertible` (`flopy4/protocols.py`). Align the protocol + method names with what's implemented (`from_xarray` vs + `from_dataset`/`from_datatree`). Add `isinstance` checks to the tests. +- Delete `flopy4/mixins.py`, and move the `Leaf`/`Node` tests in + `test/test_attrs_xarray.py` off the mixins onto the plain functions. +- Update `docs/dev/sdd.md` ("Xarray conversion") to describe the complete + export, the accessor view, and how both differ from MF6 NetCDF input. +- Close audit item 5 in `dead-code-audit.md`, pointing here. + +## Still open (decide during implementation) + +- Bool encoding in netCDF attrs: int8 + a flag, or a convention. +- Type tag values: the DFN component name (`gwf-npf`) or the class path. +- Whether the old instance `from_xarray` survives in any form (phase 2). +- Whether `dense()` lives on DataArray too, or only on Dataset. From b4a35e245fede97a1070d843333cc9956de8d434 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 07:57:34 -0800 Subject: [PATCH 02/77] add xarray package plan and representation explainer Co-Authored-By: Claude Opus 5.5 --- docs/dev/xarray-representation.md | 174 ++++++++++++++ xarray-package-plan.md | 377 ++++++++++++++++++++++++++++++ 2 files changed, 551 insertions(+) create mode 100644 docs/dev/xarray-representation.md create mode 100644 xarray-package-plan.md diff --git a/docs/dev/xarray-representation.md b/docs/dev/xarray-representation.md new file mode 100644 index 00000000..ae4a6665 --- /dev/null +++ b/docs/dev/xarray-representation.md @@ -0,0 +1,174 @@ +# The MF6 xarray representation + +Status: **draft.** Only the first section is written. The rest is an outline, +to be filled in during `xarray-export-plan.md` phase 1. The design behind it +is in `xarray-package-plan.md` at the repo root. + +The representation is the in-memory form of an MF6 simulation as xarray +objects: what flopy4's `to_xarray` produces, what `from_xarray` accepts, and +what the accessor, `dense()` and the encoders work on. Files come from +encodings of it, and none of those encodings is the representation itself. + +## How it relates to netCDF, CF, UGRID and the MF6 NetCDF format + +### The layers + +These names are easy to mix up because they sit at different levels: + +- **netCDF** is a data model and a file format: dimensions, variables, + attributes and groups. It says nothing about what the data means. +- **CF** (Climate and Forecast) is a metadata convention on top of netCDF. + It gives data meaning through `units`, `standard_name`, `axis`, + `_FillValue`, `coordinates`, `bounds`, `grid_mapping` and time units. +- **UGRID** extends CF to unstructured meshes: mesh topology variables, and + node, edge and face dimensions. +- **xarray** uses the netCDF data model in memory, but it isn't CF. + - It applies CF rules when reading (`decode_cf`): it masks fill values, + applies scale and offset, decodes times to `datetime64`, and promotes + variables named in `coordinates` to coordinates. It reverses them when + writing. + - In memory it holds things netCDF and CF can't: indexes (pandas or + custom), `datetime64` and bool dtypes, arbitrary Python objects in + attrs, and lazy or dask-backed arrays. +- **The MF6 NetCDF format** ([wiki][mf6nc]) is CF plus UGRID plus + MF6-specific attributes (`modflow_input`, `modflow_iaux`, `layer`). It's + shaped around what MF6's input reader can load. +- **The representation** (this document) is the xarray data model plus MF6 + conventions: dims and coordinates for each grid type, a tree mirroring the + simulation, period data as given, type tags. It's shaped around what + Python users and tools need. + +xarray already keeps the in-memory dataset apart from its CF-encoded file +form. The representation does the same one level up, for MF6. + +### The picture + +``` + object model (flopy4, attrs) + │ A: to_xarray / from_xarray + ▼ + representation (DataTree, xarray data model + MF6 conventions, + CF-friendly attrs, indexes, 0-based) + ┌───────────────┼──────────────────┬──────────────────┐ + ▼ ▼ ▼ ▼ + native MF6 NetCDF UGRID (xugrid) CF dataset + (groups, full project + flatten per-model mesh bounds, + tree, Python) + CF encode UgridDataset grid_mapping + (MF6 reads it, + plus text input) +``` + +CF appears twice, in two roles: + +1. **Metadata inside the representation.** Where CF already has an answer + (`units`, `standard_name`, `axis`, `grid_mapping`, fill-value + semantics), the representation uses it rather than inventing its own. +2. **An encoding on the way to disk.** Fill values, time units, bounds, + coordinate attributes, and so on. + +Because of (1), the encoders don't each have to do CF from scratch. Most of +the CF dataset encoding is bounds plus `xr.conventions.encode_cf`, and the +MF6 NetCDF and native encodings build on the same metadata. + +### The encodings + +| | native | MF6 NetCDF | UGRID | CF | +|---|---|---|---|---| +| Read by | Python (flopy4, the accessor) | MF6, GIS and CF tools | xugrid, UGRID-aware tools (QGIS, MDAL) | CF tools (GDAL, Panoply, cf-python) | +| Scope | the whole simulation | one model's grid arrays | one model's grid arrays | one model's grid arrays | +| Hierarchy | netCDF/zarr groups mirroring the tree | flat; identity kept in names and `modflow_input` | flat | flat | +| Coverage | everything: options, dims, list data, arrays | GRIDDATA/PERIOD arrays; the rest stays in text input | arrays on DISV and layered DIS | arrays on structured grids | +| Lossless | yes | for what it covers | for what it covers | for what it covers | +| Indexing | 0-based | 1-based | 0-based (UGRID `start_index`) | n/a | + +**Native** is the representation written as is: `DataTree.to_netcdf` or +`to_zarr`, with only the encoding netCDF needs (bools, None, fill values). +MF6 never reads it. It's for round trips, caching and sharing a simulation +between Python tools. + +### MF6 NetCDF is a projection, not a serialization + +The MF6 NetCDF format and the representation can't simply be the same +thing. The difference isn't mainly the flat layout: the file is a sidecar +holding one model's grid-shaped arrays, and the namefile and package text +files still carry everything else. So the encoder projects: + +- **Project:** keep one model's GRIDDATA and PERIOD array fields. +- **Flatten:** turn package paths into variable names (`npf_k`) plus + `modflow_input` (`//`). +- **Renumber:** 0-based to 1-based. +- **Reshape:** split by layer where the layered mesh form needs it + (`npf_k_l1`, …). +- **CF encode:** fill values, time units, `grid_mapping`, CRS. + +The decoder reverses these steps. It parses the identity, rebuilds the +package nodes, and merges them into a tree whose other fields come from the +text input. This round trip is lossless for what the file covers, provided +the identity can be parsed reliably. + +The flat layout is reasonable for a file MF6 reads: + +- MF6's input data model loads one model at a time, keyed by + model/package/tag. +- UGRID defines one mesh topology per dataset. +- GDAL/QGIS, CDO and older NCO handle netCDF groups poorly or not at all. + +The flat file is a flattened tree, so we don't need groups there. What we +do need from the format is listed under "What the format could change". + +### Where the representation and MF6 NetCDF differ + +| | MF6 NetCDF format | Representation | +|---|---|---| +| Indexing | 1-based cell ids | 0-based; 1-based only in encodings and `sel_cell` | +| Layers (mesh) | one variable per layer (`npf_k_l1`, …) | a `layer` dim | +| List (stress period) data | not covered | a sparse table, plus `dense()` | +| DISU, DIS2D/DISV2D/DISV1D | not covered | covered | +| Scope | one model per file, flat | a `DataTree` of the whole simulation | +| Time | input `time` per period, output `time` per step, with no stated link | both, with the mapping | +| Indexes | none (files don't store them) | `CellIndex` and similar | +| Package identity | the `modflow_input` path string | a structured type tag | +| Fill values | vary (`9.97e36`, `3e30`, `1e30`, `-2147483647`) | one per dtype | + +### What the format could change + +Not groups, and not list data (that would be nice to have, but isn't +needed). These would make the projection well-defined and its inverse +reliable: + +- **Structured identity:** separate model, package and tag attributes, or a + documented grammar for `modflow_input`, so the decoder doesn't have to + guess package names. +- **Consistent fill values** per dtype. +- **A documented layer convention:** when variables are split by layer, and + how to reassemble them. +- **A time mapping** between period-sized input time and step-sized output + time. + +These come from reading the wiki and need checking against the MF6 source +before anything is proposed upstream. + +## Grid dims and coordinates + +To write. DIS from `rotated-grid-index-plan.md` "Coordinates on a DIS +array". A table for DISV and DISU still needs writing. DIS2D/DISV2D/DISV1D +are TBD. + +## Tree, names and type tags + +To write. + +## Period data + +To write. Period arrays along `_period`, and list data as a sparse +table (see `period-lists-plan.md`). + +## Time + +To write. See `datetime-periods-plan.md`. + +## Attributes and fill values + +To write. + +[mf6nc]: https://github.com/MODFLOW-ORG/modflow6/wiki/MODFLOW-NetCDF-Format diff --git a/xarray-package-plan.md b/xarray-package-plan.md new file mode 100644 index 00000000..6721d9ed --- /dev/null +++ b/xarray-package-plan.md @@ -0,0 +1,377 @@ +# An MF6 xarray package: layering now, extraction later + +Written 2026-11-03, against develop `000f63c3`. Status: **direction agreed, +nothing scheduled.** Everything stays in flopy4 for now. This plan is about +where to draw the boundary inside flopy4 so that pulling out a separate +package later is a packaging decision rather than a redesign. + +## Decisions + +1. **Eventually, a separate package** (working name `mf6xr`; the name is + open) implements MF6's discretization, data semantics and file formats on + xarray. It's to MF6 roughly what xugrid is to UGRID. flopy4 depends on it; + it never imports flopy4 or flopy. +2. **Our own representation first, encodings second.** The in-memory form + is plain `xr.Dataset`/`xr.DataTree` with custom indexes, close to MF6's + own concepts and lossless for every grid type. CF, UGRID (through + xugrid) and the [MODFLOW NetCDF format][mf6nc] are explicit conversions + to and from it, and none of them is the core. A fourth, **native** + encoding writes the whole tree as is (groups, full fidelity) for Python + round trips. MF6 NetCDF is a *projection* of the representation (one + model's grid arrays, flattened), not a serialization of it. See + `docs/dev/xarray-representation.md`, which explains how netCDF, CF, + UGRID, the MF6 NetCDF format and the representation relate. +3. **The boundary is a data contract, not an API.** The export schema from + `xarray-export-plan.md` phase 1 is the interface. flopy4 produces it; + the package (the accessor, `dense()`, the encoders) consumes it, using + only what's in the dataset. The export plan already states the rule: + *"If it needs the package or the grid object, phase 1 is missing + something."* +4. **Not yet.** Extract only when the criteria in "When to extract" hold. +5. **The generic attrs ↔ xarray layer could become xattree v2** (see + "xattree"). It would be a second library beside the package, not under + it: xattree knows nothing about MF6, and the package knows nothing about + attrs. Same approach: make it generic inside flopy4 first, extract later. + +[mf6nc]: https://github.com/MODFLOW-ORG/modflow6/wiki/MODFLOW-NetCDF-Format +[x36]: https://github.com/wpbonelli/xattree/issues/36 + +## How this relates to the other xarray docs + +This is the umbrella doc. The others keep their scope; this one says which +layer each belongs to and what they have to agree on. + +| Doc | Where | Scope | Layer | Status / what this plan changes | +|---|---|---|---|---| +| `xarray-export-plan.md` | branch `xarray-export-plan` (`c5e3f488`), not on develop | lossless `to_xarray`/`from_xarray`, `grid.dense()` accessor | flopy (export) + package (accessor) | Decided, not started. Phase 1 now also writes the representation spec (below). Its open questions (bool encoding, type tags) are spec questions. | +| `rotated-grid-index-plan.md` | root | DIS coords, rotation, `CellIndex`, z selection, `.grid` accessor | package (grid) | Mostly done (#469–#490). Its "Next 1" DISV input decision should follow this plan's xugrid stance. | +| `combined-index-notes.md` | root | a combined `GridIndex` on `CellIndex` | package (grid) | Deferred. Revisit if DISV needs its own index anyway (it may; see xugrid). | +| `period-lists-plan.md` | root | period lists in xarray, dense/sparse | flopy (encode) + package (densify) | Done (#476). The dense view becomes `dense()` per the export plan. | +| `tas-interpolation-plan.md`, `datetime-periods-plan.md` | root | time-array series in the view; datetime period keys | package (time) | Ready / in progress. The time coordinate they produce is part of the spec. | +| `docs/dev/spatial-index-design.md` | `origin/xarray-index` (#278) | original spatial index design | — | Superseded by `rotated-grid-index-plan.md`; housekeeping already planned there and in `dev-docs-plan.md` page 3. | +| `docs/dev/sdd.md` "Xarray conversion", "NetCDF" | develop | SDD description | both | Update when export plan phase 3 lands (below). | +| `archive/xarray-xugrid-refactor.txt`, `archive/mf6-ugrid-planning.txt` | archive | May 2026 chats behind the `xugrid` branch | — | History. The `xugrid` branch they produced conflicts with decision 2; see "xugrid". | +| `dead-code-audit.md` item 5 | root | test-only attrs/xarray layer | flopy (export) | Answered by the export plan (keep it, finish it). | +| `dev-docs-plan.md` page 3 | root | `docs/dev/grid-coordinates.md` | package (grid) | Unchanged. That page later becomes part of the package's docs. | +| [xattree #36][x36] | GitHub | functional rewrite of xattree | generic (A1) | Open. flopy4's `attrs_xarray.py` is the working prototype; see "xattree". | +| `pydantic-object-model-plan.md` | root | attrs vs pydantic dataclasses (#367) | flopy (object model) | Stale. Its outcome decides whether xattree v2 is attrs-only. | + +## The layers + +Four jobs, today all in flopy4. Job A splits into a generic half and an MF6 +half: + +| Job | What | Today | Layer | +|---|---|---|---| +| A1. attrs ↔ xarray, generic | attrs instance ↔ `Dataset`/`DataTree` from field metadata (dims, `shape`), child fields as tree nodes, protocols | `attrs_xarray.py` (most of it), `protocols.py`, part of `dimensions.py` | **xattree v2** (eventually) | +| A2. Object model ↔ xarray, MF6 | hooks for period arrays, TAS refs, fill values, aux names, references; type tags; which DFN field goes on which dims | the MF6 parts of `attrs_xarray.py`, `Package/Context.to_xarray`, `adapters.py` | **flopy** | +| B. Discretization → coordinates and indexes | dims and coordinates per grid type, cell centres, z bounds, rotation and CRS, `CellIndex`, `sel`/`sel_world`/`sel_z`/`mask_z` | `utils/grid.py`, `utils/vertical.py`, `utils/crs.py`, `accessor.py` | **package** | +| C. MF6 data semantics | sparse table ↔ dense, period fill-forward, cellid ↔ node, the `duplicates` policy, TAS evaluation | `period_lists.py` (densify half), `period_arrays.py`, the planned `dense()` | **package** | +| D. Encodings and files | MF6 NetCDF input/output, CF, UGRID, GRB/HDS/CBC readers | `netcdf.py`, `utils/netcdf_postprocess.py`, `utils/heads_reader.py`, `utils/cbc_reader.py`, `dims_from_grb` | **package** | + +Dependencies: **flopy4 → package → xarray** and **flopy4 → xattree → +attrs, xarray**, with xugrid (and flopy3, while it's still needed) +optional. The package and xattree don't depend on each other. Jobs B–D have users who never touch +flopy's object model: people post-processing MF6 output, xmipy/BMI users, +other model builders. That's what makes a separate package worth having +at all. + +### Coupling to remove (as of `000f63c3`) + +| Module | Imports that cross the boundary | What to do | +|---|---|---| +| `utils/grid.py` | subclasses flopy3 `StructuredGrid`/`VertexGrid`; imports `xugrid` at module level; `flopy4.adapters.read_binary_grid_file` (lazy) | Compute geometry from arrays (composition, not inheritance). Build a flopy3 `modelgrid` *from* it where flopy needs one. Make xugrid lazy. Move the GRB reader to D. | +| `utils/heads_reader.py`, `utils/cbc_reader.py` | `flopy.discretization`, `flopy4.adapters`, `xugrid` | Take grid geometry from the GRB (or a dataset), not a flopy grid. | +| `utils/time.py` | `flopy.discretization.modeltime` | Replace with a small time-coordinate builder from perlen/nstp/tsmult (+ start date). | +| `period_lists.py` | `item`, `item_table`, `flopy4.spec` | Split: the encoder (`ItemTable` → sparse table) stays in flopy, and densify/sparsify on tables moves to C. | +| `period_arrays.py` | `_types` (`TimeArraySeriesRef`), `constants` | Constants move with it. TAS refs are already coordinates (`_tas`, #489), so dense code shouldn't need the type. | +| `netcdf.py` | `Model`, `Package`, `spec`, pydantic `NetCDFModel` | Split: "walk the model and pick fields" stays in flopy, and "write a representation dataset in MF6 NetCDF form" moves to D. | +| `vertical.py`, `crs.py`, `accessor.py` | none | Already clean. | +| `attrs_xarray.py` (A1) | lazy `flopy4.mf6._types.TimeArraySeriesRef`, `flopy4.mf6.constants.FILL_DNODATA`; `flopy4.spec` references; `fill_forward`/`fk` metadata | Move these into hooks registered from `flopy4.mf6` (see "xattree"). | +| `dimensions.py` (A1) | `modflow_devtools.dfns.dim_value`; `flopy4.spec.is_reference` | Keep `DerivedDim` and DFN expressions in flopy; the generic part is a small resolver. | + +## The representation + +The contract from decision 3. Write it down as part of export plan phase 1 +in `docs/dev/xarray-representation.md` (later the package's spec). Its first +section, on how the representation relates to netCDF, CF, UGRID and the MF6 +NetCDF format, is drafted. Most of the rest is already decided in other +docs, so the spec mainly collects it: + +- **Grid dims and coordinates per grid type:** + - DIS `layer,row,col` with 1-D indexed `x`/`y` + - DISV `layer,cell` + - DISU `node` + - DIS2D/DISV2D/DISV1D: TBD + - `z`/`z_top`/`z_bot`, `dx`/`dy`, `projection`, `x_world`/`y_world` + - Source: `rotated-grid-index-plan.md` "Coordinates on a DIS array", plus a DISV/DISU table to write. +- **Indexing base:** 0-based everywhere (`isel` and dims); 1-based only in + encodings and in an explicit `sel_cell`. +- **Grid metadata in attrs:** grid type and sizes (`nlay`, `nrow`, `ncol`, + `ncpl`, `nodes`), `nper`, so consumers don't need the DIS object + (export plan phase 1). +- **Period data as given:** period arrays along `_period`, list data + as a sparse table (entry dim, `per` and cellid coordinates, a variable per + column, `_ts`/`_tas` coordinates). Dense is a view (`dense()`). +- **Time:** how `per` maps to a time coordinate (from TDIS, with a start date + where given); ties in with `datetime-periods-plan.md`. +- **Tree:** nodes named by component name, with a type tag (export plan open + question: DFN name `gwf-npf` vs class path; prefer the DFN name, since it + doesn't depend on Python). +- **netCDF-safe attrs:** paths as str, bools encoded (open), None and empty + child lists omitted. +- **Fill values:** one convention per dtype (see the MF6 NetCDF notes below). + +## Encodings + +Each is a pair of functions (representation ↔ encoded dataset) plus +accessor methods. None changes the representation. + +- **Native**: the representation as is, via `DataTree.to_netcdf`/`to_zarr` + with groups mirroring the tree, encoding only what netCDF needs (bools, + None, fill values). It covers everything (options, dims, list data, + arrays). MF6 never reads it; it's for Python round trips, caching and + sharing. The name is open (see "Open"). +- **MF6 NetCDF (structured and layered mesh)**: what `netcdf.py` writes + today, restated as an encoder from the representation. It's a + projection: keep one model's GRIDDATA/PERIOD arrays, flatten package paths + into names plus `modflow_input`, renumber to 1-based, split by layer where + the mesh form needs it, then CF encode. Readers reverse this and merge the + result into a tree whose other fields come from the text input, so an + MF6 NetCDF file opens as the representation. +- **UGRID via xugrid**: `ds..to_ugrid()` → `UgridDataset` for DISV and + layered DIS, and `from_ugrid` back. This is where `UgridDataset` appears, + not in `to_xarray`. +- **CF**: bounds (`x_bnds`/`y_bnds`/`z_bnds`, `rotated-grid-index-plan.md` + Next 3), `grid_mapping`, standard names. Mostly a subset of the structured + MF6 NetCDF encoding. +- **Binary output** (GRB/HDS/CBC): readers that produce the representation + directly. + +### Notes on the MODFLOW NetCDF format + +The format is ours to change too, so the representation work is a chance to +feed fixes back to MF6. The goal isn't to make the format a serialization of +the representation. It's a flat, per-model sidecar to the text input, and +that's reasonable for a file MF6 reads: the input data model loads one model +at a time, UGRID defines one mesh per dataset, and GIS tools handle groups +poorly. So don't push for groups, and treat list data as nice to have. The +goal is a well-defined projection whose inverse is reliable. That mainly +needs structured identity, consistent fill values, a documented layer +convention and a time mapping. + +Candidates, from reading the wiki (2026-11-03). They need checking against +the MF6 source before anything is proposed upstream: + +- **Per-layer variable splitting** in the layered mesh (`npf_k_l1`, + `npf_k_l2`, `z_l`, each with a `layer` attr): a variable's identity is + spread over names and attrs. xugrid and most CF tools would rather see a + `layer` dim. If that's needed for QGIS/GDAL, keep it as an encoding detail, + but say so in the spec. +- **No DISU**, and no DIS2D/DISV2D/DISV1D. The representation has to cover + them regardless, so it shouldn't wait on the format. +- **Inconsistent fill values**: `9.97e36` (GRIDDATA), `3e30` (PERIOD input + example), `1e30` (output), `-2147483647` (int). +- **`modflow_input` is a path string** (`//`) plus + `modflow_iaux` and `layer`. That's workable, but the decoder has to parse + strings to find the package, and the package name depends on the namefile + (or FloPy's generated name). +- **Input `time` is NPER-sized and output `time` is step-sized**, with no + stated way to relate the two. The representation needs both (`per` and step + time), so the format could carry the mapping. +- **List (stress period) data isn't covered**; only GRIDDATA/PERIOD arrays + in READASARRAYS/READARRAYGRID packages. Low priority: the native encoding + covers list data, and the text input carries it for MF6. +- **Rotation is only encoded through a wrapped WKT2 `PROJCRS`**; without a + CRS, rotated structured grids lose `axis`/`standard_name`. The + representation's `projection` coordinate carries xorigin/yorigin/angrot + either way (`rotated-grid-index-plan.md`). The format could do the same. +- **"No naming rules beyond CF"**: fine, but then the type tag / + `modflow_input` is the only reliable identity, which argues for making it + structured. + +## xugrid + +- **Optional, behind the UGRID encoding and DISV selection**, not in the core + representation. A `UgridDataset` is a wrapper type, not an `xr.Dataset`: + it doesn't nest in a `DataTree`, and our registered accessors don't reach + it. That conflicts with the export plan's "packages give a Dataset, + contexts a DataTree". +- **The `xugrid` branch** (May 2026, `b00d7a05`, merge-base `52daeeb0`) + returns xugrid objects from `to_xarray`. Don't merge it as is. Its useful + parts (dropping hand-written UGRID attrs, keeping CF bounds) go into the + UGRID encoder instead. +- **DISV selection** (`rotated-grid-index-plan.md` Next 1): output readers + return `UgridDataArray` today. For the representation, prefer a plain + dataset with an index (an `NDPointIndex` for nearest, or our own + containment index, maybe the `combined-index-notes.md` design). Use + `to_ugrid()` for anything xugrid does better (regridding, plotting). +- **Watch xugrid #35 / PR #373** (custom indexes instead of wrapper types). + If it lands, DISV could be "plain dataset plus xugrid's index", and the + wrapper concern goes away. +- **The pin** (`xugrid==0.15.3` in `pyproject.toml`) needs a range anyway + (`flopy-integration-plan.md` track B). Making it optional is a bigger + step; do it once grid.py no longer imports it at module level. + +## xattree + +[xattree #36][x36] proposes rewriting xattree as a functional conversion +layer: explicit `to_datatree()`/`from_datatree()`, protocols instead of +inheritance, a normal object lifecycle, pure translation functions. It +would keep the spec system, the field factories (`dim`, `coord`, `array`) +and dimension inheritance, and drop the imposter dict/list classes, +`bind_tree()` and `init_tree()`. + +**flopy4 has mostly built this already** since dropping xattree (#356, +#359): + +| #36 asks for | flopy4 today | +|---|---| +| pure `to_datatree`/`from_datatree` | `attrs_to_dataset`, `dataset_to_attrs`, `attrs_to_datatree`, `datatree_to_attrs` in `flopy4/attrs_xarray.py` | +| a `DataTreeConvertible` protocol | `flopy4/protocols.py`, same name | +| dimension inheritance | `flopy4/dimensions.py`: `DimensionProvider`/`DimensionResolver` protocols, `DerivedDim` | +| dims tagged on fields | `shape` metadata from `flopy4.mf6.spec.array()` | +| no imposter containers, no `__dict__` swapping | gone | + +So xattree v2 would start from `attrs_xarray.py` and the generic part of +`dimensions.py`. Little of the current xattree code carries over beyond +the field factory names. + +**What isn't generic yet** (see "Coupling to remove"): period-array +stacking along `_period` (keyed off `fill_forward`), the +`_name` aux dim (`fk` + `fill_forward`), `TimeArraySeriesRef` and +`FILL_DNODATA`, references stored by name, the reserved names `dims`, +`parent` and `_parent`, and DFN dimension expressions. + +**Design: a converter with hooks**, like cattrs (which flopy already uses +in `converter/`): + +- hooks keyed by a predicate on the field, each a pair (value → + `DataArray`, `DataArray` → value); +- a policy for which fields are children (tree nodes) and which are leaves; +- a policy for excluded fields (instead of hard-coded reserved names); +- a fallback dim naming (`_dim`) when a field has no `shape`. + +flopy4 registers the MF6 hooks (period arrays, TAS refs, fill values, +references). xattree doesn't set naming conventions. The representation +spec does that, through flopy's field metadata and hooks. + +**Pipeline:** object → xattree (with flopy's hooks) → a raw `DataTree` that +follows the representation spec → the package (indexes, `dense()`, +encodings). + +**Dimension inheritance may be partly native now.** `DataTree` inherits +coordinates from parent nodes, which xattree used to do by hand. That +covers dims once a tree exists. Resolving dims before conversion (shape +checks, scalar expansion) still needs something at the object level. xattree +v2 may only need a small pure `resolve_dims(obj, ancestors)`, not the +provider/resolver protocols. Check this before extracting. + +**Payoff and risks:** + +- The code is about 400 lines, so extraction won't shrink flopy much. The + main payoff is the boundary: a generic core makes the MF6 parts explicit + as hooks, and that's the same discipline the package needs. Reuse outside + flopy is secondary. +- **attrs vs pydantic.** #367 is stale but open. An attrs-only xattree + locks in attrs. attrs, dataclasses and pydantic dataclasses all expose + fields with metadata, so a thin field-introspection layer could keep the + choice open. Better still, settle #367 first. +- **Repo home.** flopy would depend on a personal repo. Before flopy + depends on xattree, consider moving it to MODFLOW-ORG, as with the package. + +## Steps inside flopy4 + +Roughly in order. Each is small and fits the existing plans. None needs a +new repo. + +1. **Write the representation spec** (`docs/dev/xarray-representation.md`, + explainer section drafted) as part of export plan phase 1. It settles + that plan's open questions (bool encoding, type tags). +2. **Rename the accessor** from `.grid` to something distinctive (e.g. + `.mf6`) before more users depend on it. `.grid` is generic enough to + collide (xugrid uses `.ugrid`, uxarray `.uxgrid`). Do it with export plan + phase 2, which adds `dense()` to it anyway. +3. **One subpackage for jobs B–D**, e.g. `flopy4/mf6/xr/` (or top-level + `flopy4/xr/`). Move the modules as they're touched rather than in one + sweep. Add an import-linter contract forbidding imports from + `flopy4.mf6.{component,package,context,model,simulation,item,item_table}` + and `flopy4.spec`, so the boundary holds while it's one repo. +4. **Drop the flopy3 grid inheritance** (`utils/grid.py`), and replace + `utils/time.py`'s use of flopy3's `ModelTime`. This is the biggest piece. + It's also what `flopy-integration-plan.md` needs, since flopy4 moving + into flopy shouldn't depend on flopy3's discretization classes. +5. **Split `netcdf.py` and `period_lists.py`** along the encode/densify line + above. +6. **Lazy xugrid** everywhere in the subpackage, then decide whether it's + an optional dependency. +7. **Update `docs/dev/sdd.md`** ("Xarray conversion", "NetCDF") to describe + the representation, the encodings, and the boundary. The current "NetCDF" + section ("adopt xugrid ... `.to_netcdf()` through the appropriate + accessor") is close to right, except that xugrid is one encoding, not the + writer. +8. **Make `attrs_xarray.py` generic** as part of export plan phase 1, which + already says to "make the generic layer complete". Add an import-linter + contract so `flopy4.attrs_xarray`, `flopy4.protocols` and the generic + part of `dimensions` can't import `flopy4.mf6`. Then move the period, + TAS, fill-value and reference handling into hooks registered from + `flopy4.mf6`. Worth doing even if xattree v2 never happens. + +## When to extract + +All of: + +- the spec has been stable for a release or two (no schema changes in + export or period-list work), +- the import-linter contract is green and the subpackage imports neither + flopy3 nor flopy4's object model, +- someone outside flopy wants it: an output-only user, imod, xmipy, a GUI, +- flopy4 is in flopy (`flopy-integration-plan.md`), so the third repo isn't + on top of the devtools release coordination already blocking that. + +Until then, a subpackage with an enforced boundary gets most of the benefit. + +**xattree v2** is a separate decision with its own criteria: the A1 +linter contract is green, the hook API has gone unchanged for a while, +and #367 is settled. Then rewrite xattree around the module, close #36 +with it, and have flopy depend on it. This is independent of the package's +timing. + +## Coordination + +- **imod-python** already builds MF6 on xarray/xugrid and reads GRB/HDS/CBC + into xarray. Before creating a package, ask Deltares whether the grid, + encoding and output layer could be shared, or at least whether we can + agree on the representation so datasets move between the tools. Don't + contact anyone without asking. +- **MF6 NetCDF format**: the notes above, as an issue or discussion on + modflow6, once checked against the source. Don't post without asking. +- **xugrid**: follow #35 / #373. Report DISV edge-tie behaviour if it + differs from DIS (`rotated-grid-index-plan.md` Next 1). +- **xattree**: ours, so no outside coordination, apart from where it lives + (personal repo vs MODFLOW-ORG). Note in #36 that flopy4's `attrs_xarray.py` + is the prototype, when there's something to point at. Don't post without + asking. + +## Open + +- Name for the native encoding. Candidates: "MF6" (parallels "MF6 + NetCDF", "UGRID", "CF", but collides with MF6's own text input format, + which flopy also writes, and MF6 can't read it), "native", or the + package's name. +- Package name, and whether the subpackage is `flopy4/xr/` or `flopy4/mf6/xr/`. +- Accessor name. +- Whether the time code (TDIS → time coordinate, TAS evaluation) is part of + the package or stays in flopy. Leaning package, since output readers need + it too. +- Does the package know DFNs (for variable metadata, long names, units)? + Prefer not: the encoder takes the metadata from attrs that flopy puts on + the representation, and devtools stays a flopy dependency only. +- Representation for DIS2D/DISV2D/DISV1D (OLF/CHF). +- Merge `xarray-export-plan.md` onto develop (or into the working tree) so + the two plans sit side by side. +- attrs vs pydantic (#367): whether xattree v2 is attrs-only or works with + any dataclass-like. +- Whether xattree v2 keeps the field factories (`dim`, `coord`, `array`) + or just reads metadata that flopy's own `spec` functions set. +- Where xattree lives: personal repo or MODFLOW-ORG. From 5217b5dfc25cd5a33879eaff935fbc3e58d61ca5 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 08:01:55 -0800 Subject: [PATCH 03/77] plan flat and layered cell layouts on the accessor Co-Authored-By: Claude Opus 5.5 --- xarray-export-plan.md | 156 ++++++++++++++++++++++++++++++++++++++++- xarray-package-plan.md | 4 +- 2 files changed, 158 insertions(+), 2 deletions(-) diff --git a/xarray-export-plan.md b/xarray-export-plan.md index 39214d25..6437e3f9 100644 --- a/xarray-export-plan.md +++ b/xarray-export-plan.md @@ -30,13 +30,21 @@ round trips. does now. Filling forward is the accessor's job. 5. **The complete export is writable to netCDF.** It is distinct from MF6's NetCDF input format (`flopy4/mf6/netcdf.py`, `NetCDFModel`), and - the docs must say which is which. + the docs must say which is which. `docs/dev/xarray-representation.md` + does: this is the hierarchical (grouped) encoding, and MF6 NetCDF is a + flat projection of the same representation. 6. **Keep `protocols.py`; put the methods on `Package`/`Context` directly; delete `mixins.py`.** The protocols are cheap, the SDD describes them, and `isinstance(x, DataTreeConvertible)` is useful. The mixins are two-line forwarders that sit awkwardly in the hierarchy (`Context` subclasses `Component`). The reverse functions in `attrs_xarray.py` stay and get finished. +7. **Both cell layouts, flat and layered, through the accessor.** The + export is layered (decision 3): `layer,row,col` on DIS, `layer,cell` + on DISV. `grid.flatten()` gives the flat `node` view, as on DISU, and + `grid.unflatten()` goes back. This is plain `stack`/`unstack`, not a + custom index. The old "index mapping nodes to layer/row/col" (#207, see + `rotated-grid-index-plan.md`) was dropped in #469 and stays dropped. ## Where things stand @@ -166,6 +174,137 @@ them directly. `flopy4/mf6/adapters.py` (it calls `attrs_to_dataset` directly) still gets what it needs. +### Phase 2b: flat and layered cell layouts on the `grid` accessor + +Decision 7. Both methods go on the `grid` accessor next to `dense()`, +work on Dataset and DataArray, and compose with it +(`ds.grid.dense().grid.flatten()`). + +How this fits `xarray-package-plan.md`: + +- `flatten()`/`unflatten()` are views, like `dense()`. They belong to the + package side (grid and data semantics), use only what's in the + dataset, and never change the representation. Writing a flat view to + netCDF is part of the CF encoding, not a new format. +- `.grid` is the current accessor name. The package plan (step 2) + renames it to something distinctive before more users depend on it, so + rename these with it. +- A cf_xarray dependency (for `compress`, below) would be the package's, + and optional. + +- **`grid.flatten(dim="node")`**: `stack(node=("layer", "row", "col"))` on + DIS, `stack(node=("layer", "cell"))` on DISV. Layer is slowest, so + position `i` along `node` is MF6 user node `i + 1`, as in the flat DFN + shape packages store internally (`Package.to_dataarray` reshapes from + it). Variables without all the cell dims (DIS `delr`/`delc`, RCHA's + one-layer arrays, list tables) are left as they are. A no-op on DISU + and on arrays already flat. +- **`grid.unflatten(dim="node")`**: `unstack` when `node` carries the + MultiIndex from `flatten()`. Without one (a flat array from somewhere + else, or the export's DFN-dim fallback on `nodes`), reshape using the + grid sizes phase 1 records in attrs (`nlay`, `nrow`, `ncol`/`ncpl`), + and raise if they're missing or don't multiply out to the length. +- **Coordinates through the round trip**: per-cell coords (`z`, + `z_top`, `z_bot`; `x`/`y` on DISV) stack along with the data. `x`/`y` + on DIS are 1-D on `col`/`row` and carry a `CellIndex`. Find out what + `stack` does with them. If they're dropped or lose the index, the flat + view broadcasts them to `node` as plain coords, and `unflatten()` + restores the 1-D form and re-runs `index_xy`. Containment `sel(x=, y=)` + is a layered-view feature; on the flat view, use `.grid.sel` (see open + questions). +- **netCDF**: xarray can't write a MultiIndex. The flat view is for + analysis; the complete export (decision 5) stays layered. If a flat + dataset needs writing, `flatten()` could take `index=False` to keep + `layer`/`row`/`col` as plain coords on `node`, which `unflatten()` can + still read. CF compression by gathering is the standard way to write + it (see prior art below). +- **Not reduced nodes.** `node` here means user nodes (every cell). + Budget output on reduced nodes (active cells only) is a different + layout; leave it out of these methods, and don't name it `node`. +- **Tests**: `flatten().unflatten()` is the identity on a DIS and a DISV + package export, on a period array (with `per`), and on a dense list + variable. Flat values match the package's internal flat arrays. + `unflatten()` from attrs alone. `CellIndex` comes back after the round + trip. DISU is a no-op. + +#### Prior art: xugrid, uxarray, CF + +We checked what the established libraries do, so that where we differ +we do it on purpose. Sources: xugrid 0.15.3 (installed in the pixi env, +read from source), uxarray `main` (read from source on GitHub), and the +CF conventions. + +| | xugrid | uxarray | Phase 2b | +|---|---|---|---| +| Entry point | `UgridDataArray.from_structured2d(da)` | `UxDataset.from_structured(ds)` | `.grid.flatten()` | +| Dims flattened | `(y, x)` only | `(lat, lon)` only | `(layer, row, col)` / `(layer, cell)` | +| Layer / vertical | kept as its own dim | kept as its own dim | folded into `node` | +| How | `da.stack({face_dim: (y, x)}, create_index=False)`, no MultiIndex | `ds.stack(n_face=[lat, lon])`, MultiIndex | `stack`, MultiIndex by default (open) | +| Order | row-major (y slowest) | row-major (lat slowest) | layer slowest, then row-major (MF6 order) | +| Way back | none by index; `rasterize`/`rasterize_like` regrid geometrically (lossy) | none | `unflatten()`, exact | +| Flat dim name | mesh-prefixed, e.g. `mesh2d_nFaces` | `n_face` | `node` | + +Where we match: + +- **Cell order.** Row-major within a layer, as in both libraries. +- **DISV layered view.** `(layer, cell)` is the UGRID layered-mesh shape: + a 2-D mesh with a separate layer dim. Both libraries use it, and so + does MF6's own netCDF mesh export (`head_l1`, `head_l2`, … on + `nmesh_face`). + +Where we depart, and what to do about it: + +1. **Folding layer into `node`.** Neither library does this. It's still + right for MF6, which numbers nodes in 3-D, and DISU uses that + numbering. Keep it, and say so in the docstring. Also consider a + horizontal-only flatten (`flatten(layers=False)` or similar): + `(layer, row, col)` → `(layer, cell)`. That is xugrid's convention + exactly, and it puts DIS data in the same shape as DISV. The UGRID + encoder needs this step for layered DIS anyway, so write it once and + use it in both places. +2. **MultiIndex by default.** The libraries disagree. xugrid skips it, + which keeps the result netCDF-writable; uxarray keeps it. This leans + toward no MultiIndex by default (`index=False`), or at least making + the no-index form fully supported, including by `unflatten()`. +3. **Inverse.** Neither library has an exact index-based inverse. + `unflatten()` is our addition. That's fine, but don't confuse it with + xugrid's `rasterize`, which resamples. +4. **Dim naming inside flopy4.** The output readers + (`flopy4/mf6/utils/heads_reader.py`, `cbc_reader.py`) already return + DISV heads and budgets as xugrid objects on + `(time, layer, nmesh_face)`, while the export uses `(layer, cell)`. + That mismatch is within flopy4, independent of the libraries. The + package plan settles it: readers produce the representation directly + and xugrid objects only come from the UGRID encoding (`to_ugrid()`). + So the readers move to `cell`, and `nmesh_face` is a UGRID encoding + detail. + +**netCDF for the flat view: CF "compression by gathering"** (CF §8.2). +This is the established encoding for a stacked dim. The flat dim is +written as an integer coordinate holding each cell's flattened index, +with a `compress = "layer row col"` attribute naming the dims it was +gathered from. cf_xarray implements both directions +(`cf_xarray.encode_multi_index_as_compress` and +`decode_compress_to_multi_index`), and its docs show a netCDF round +trip. That answers the netCDF bullet above: a flat dataset can be +written in a standard form, not only as plain coords. It also fits +reduced (active-only) budget nodes, since dropping always-missing cells +is what the convention was designed for. If reduced nodes are handled +later, use this encoding there too. cf_xarray isn't in the env. +Depend on it, or write the encoding ourselves (it's small: compute +`np.ravel_multi_index` on the way out, `np.unravel_index` on the way +back). + +References: + +- xugrid: `xugrid/core/wrap.py` (`from_structured2d`), + `xugrid/core/dataarray_accessor.py` (`rasterize`, `rasterize_like`) +- uxarray: [`_map_dims_to_ugrid`](https://github.com/UXARRAY/uxarray/blob/main/uxarray/io/utils.py), + [`Grid` API](https://uxarray.readthedocs.io/en/stable/generated/uxarray.Grid.html) +- CF: [§8.2 Compression by gathering](https://cfconventions.org/Data/cf-conventions/cf-conventions-1.2/build/ch08s02.html) +- cf_xarray: [`encode_multi_index_as_compress`](https://cf-xarray.readthedocs.io/en/latest/generated/cf_xarray.encode_multi_index_as_compress.html), + [`decode_compress_to_multi_index`](https://cf-xarray.readthedocs.io/en/latest/generated/cf_xarray.decode_compress_to_multi_index.html) + ### Phase 3: wiring and cleanup - `Context.from_xarray(tree)` classmethod (`datatree_to_attrs`), with @@ -186,3 +325,18 @@ them directly. - Type tag values: the DFN component name (`gwf-npf`) or the class path. - Whether the old instance `from_xarray` survives in any form (phase 2). - Whether `dense()` lives on DataArray too, or only on Dataset. +- Names: `flatten()`/`unflatten()`, or something else. Also whether the + flat dim is `node` (DISU's, so a flat DIS looks like DISU) or `nodes` + (the DFN fallback). Pick one and make the fallback use it too. +- Whether `.grid.sel(layer=, row=, col=)` and `.grid.sel(x=, y=)` should + work on the flat view (map to `node` positions), or require + `unflatten()` first. +- Whether `flatten()` keeps the MultiIndex by default (`sel(layer=0)` + works on the flat view) or leaves plain coords (netCDF-safe). Prior + art leans toward plain coords (xugrid's choice). +- Whether to add a horizontal-only flatten (`(layer, row, col)` → + `(layer, cell)`), matching xugrid/UGRID layered meshes. +- netCDF for flat data: CF `compress` encoding via cf_xarray (new + dependency) or our own small implementation. +- Cell dim name across export and output readers: leaning `cell` for + both, with `nmesh_face` only in the UGRID encoding (departure 4). diff --git a/xarray-package-plan.md b/xarray-package-plan.md index 6721d9ed..c2724365 100644 --- a/xarray-package-plan.md +++ b/xarray-package-plan.md @@ -112,7 +112,9 @@ docs, so the spec mainly collects it: (export plan phase 1). - **Period data as given:** period arrays along `_period`, list data as a sparse table (entry dim, `per` and cellid coordinates, a variable per - column, `_ts`/`_tas` coordinates). Dense is a view (`dense()`). + column, `_ts`/`_tas` coordinates). Dense is a view (`dense()`), + and so is the flat `node` layout (`flatten()`/`unflatten()`, export plan + phase 2b). Neither is stored. - **Time:** how `per` maps to a time coordinate (from TDIS, with a start date where given); ties in with `datetime-periods-plan.md`. - **Tree:** nodes named by component name, with a type tag (export plan open From 869f5fc4a1df6461e1aaeec4f1290239349aafc2 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 08:17:20 -0800 Subject: [PATCH 04/77] plan MF6 NetCDF as two layouts plus in-memory adapters Co-Authored-By: Claude Opus 5.5 --- docs/dev/xarray-representation.md | 217 +++++++++++++++++++----------- xarray-export-plan.md | 20 +-- xarray-package-plan.md | 156 +++++++++++++-------- 3 files changed, 252 insertions(+), 141 deletions(-) diff --git a/docs/dev/xarray-representation.md b/docs/dev/xarray-representation.md index ae4a6665..a80b524f 100644 --- a/docs/dev/xarray-representation.md +++ b/docs/dev/xarray-representation.md @@ -6,8 +6,8 @@ is in `xarray-package-plan.md` at the repo root. The representation is the in-memory form of an MF6 simulation as xarray objects: what flopy4's `to_xarray` produces, what `from_xarray` accepts, and -what the accessor, `dense()` and the encoders work on. Files come from -encodings of it, and none of those encodings is the representation itself. +what the accessor, `dense()`, the file writers and the adapters work on. +Files are written from it, and no file format is the representation itself. ## How it relates to netCDF, CF, UGRID and the MF6 NetCDF format @@ -20,8 +20,8 @@ These names are easy to mix up because they sit at different levels: - **CF** (Climate and Forecast) is a metadata convention on top of netCDF. It gives data meaning through `units`, `standard_name`, `axis`, `_FillValue`, `coordinates`, `bounds`, `grid_mapping` and time units. -- **UGRID** extends CF to unstructured meshes: mesh topology variables, and - node, edge and face dimensions. +- **UGRID** is an extension of CF for unstructured meshes: mesh topology + variables, and node, edge and face dimensions. - **xarray** uses the netCDF data model in memory, but it isn't CF. - It applies CF rules when reading (`decode_cf`): it masks fill values, applies scale and offset, decodes times to `datetime64`, and promotes @@ -30,9 +30,9 @@ These names are easy to mix up because they sit at different levels: - In memory it holds things netCDF and CF can't: indexes (pandas or custom), `datetime64` and bool dtypes, arbitrary Python objects in attrs, and lazy or dask-backed arrays. -- **The MF6 NetCDF format** ([wiki][mf6nc]) is CF plus UGRID plus - MF6-specific attributes (`modflow_input`, `modflow_iaux`, `layer`). It's - shaped around what MF6's input reader can load. +- **The MF6 NetCDF format** ([wiki][mf6nc]) is a domain convention on top + of CF and UGRID. It adds MF6 attributes (`modflow_input`, `modflow_iaux`, + `layer`) and comes in a structured form and a layered-mesh form. - **The representation** (this document) is the xarray data model plus MF6 conventions: dims and coordinates for each grid type, a tree mirroring the simulation, period data as given, type tags. It's shaped around what @@ -41,7 +41,33 @@ These names are easy to mix up because they sit at different levels: xarray already keeps the in-memory dataset apart from its CF-encoded file form. The representation does the same one level up, for MF6. -### The picture +### Conventions stack + +CF, UGRID and MF6 NetCDF aren't alternatives. They are meant to be layered: +CF is the base, UGRID extends it for meshes, and domain conventions such as +MF6 NetCDF (or ACDD, SGRID) add their own attributes on top. A file lists +all of them in one attribute, e.g. `Conventions = "CF-1.11 UGRID-1.0"`. + +This works because attributes add up. A CF tool reads the CF attributes and +ignores `modflow_input`; a UGRID tool reads the mesh topology and ignores +the rest. So one file can be valid CF, valid UGRID and valid MF6 NetCDF at +the same time. + +Layering breaks on structure, not on attributes. When a format makes a +structural choice the base conventions don't expect, generic tools stop +understanding the data even though the file is technically valid. Examples +in the MF6 NetCDF format today: + +- **Splitting variables by layer** (`npf_k_l1`, `npf_k_l2`, …). CF and + UGRID allow a `layer` dim on face data; a generic tool sees N unrelated + variables instead of one 3-D field. +- **Rotation only through a wrapped WKT2 `PROJCRS`.** CF's usual answer is + 2-D auxiliary coordinates (`x_world`/`y_world`, named in `coordinates`), + which the representation already has. +- **Groups.** CF has allowed groups since 1.8, but most UGRID and GIS tools + still read only the root group. + +### The design: one format in two layouts, plus adapters ``` object model (flopy4, attrs) @@ -49,104 +75,145 @@ form. The representation does the same one level up, for MF6. ▼ representation (DataTree, xarray data model + MF6 conventions, CF-friendly attrs, indexes, 0-based) - ┌───────────────┼──────────────────┬──────────────────┐ - ▼ ▼ ▼ ▼ - native MF6 NetCDF UGRID (xugrid) CF dataset - (groups, full project + flatten per-model mesh bounds, - tree, Python) + CF encode UgridDataset grid_mapping - (MF6 reads it, - plus text input) + ┌──────────┴────────────────┐ + ▼ files: MF6 NetCDF ▼ adapters (in memory) + ┌──────┴───────┐ to_ugrid() → UgridDataset + ▼ ▼ to_rioxarray() → rio-ready DataArray + grouped ungrouped to_cf() → CF-ready Dataset + full tree, flattened; to_modelgrid() → flopy3 modelgrid + lossless, CF + UGRID valid; + CF 1.8 MF6 reads it each library writes its own + groups ├─ structured files (UGRID netCDF, GeoTIFF) + └─ layered mesh ``` -CF appears twice, in two roles: - -1. **Metadata inside the representation.** Where CF already has an answer - (`units`, `standard_name`, `axis`, `grid_mapping`, fill-value - semantics), the representation uses it rather than inventing its own. -2. **An encoding on the way to disk.** Fill values, time units, bounds, - coordinate attributes, and so on. - -Because of (1), the encoders don't each have to do CF from scratch. Most of -the CF dataset encoding is bounds plus `xr.conventions.encode_cf`, and the -MF6 NetCDF and native encodings build on the same metadata. - -### The encodings - -| | native | MF6 NetCDF | UGRID | CF | -|---|---|---|---|---| -| Read by | Python (flopy4, the accessor) | MF6, GIS and CF tools | xugrid, UGRID-aware tools (QGIS, MDAL) | CF tools (GDAL, Panoply, cf-python) | -| Scope | the whole simulation | one model's grid arrays | one model's grid arrays | one model's grid arrays | -| Hierarchy | netCDF/zarr groups mirroring the tree | flat; identity kept in names and `modflow_input` | flat | flat | -| Coverage | everything: options, dims, list data, arrays | GRIDDATA/PERIOD arrays; the rest stays in text input | arrays on DISV and layered DIS | arrays on structured grids | -| Lossless | yes | for what it covers | for what it covers | for what it covers | -| Indexing | 0-based | 1-based | 0-based (UGRID `start_index`) | n/a | +**Files** are MF6 NetCDF, in two layouts of one spec: + +- **MF6 NetCDF (grouped)** is the representation encoded as is. Groups + mirror the tree (simulation → models → packages), and it covers + everything: options, dims, list data, arrays. It's valid CF 1.8+ with + groups, but generic tools aren't expected to use it. MF6 doesn't read it + yet. Today it's for round trips, caching, archiving and sharing a whole + simulation from Python (netCDF, or zarr for cloud storage). +- **MF6 NetCDF (ungrouped)** is the same content flattened by a fixed rule. + Group paths become name prefixes and `modflow_input`. This is the + layout MF6 reads, and the interoperability layout: it aims to be fully + valid CF and UGRID, so QGIS, GDAL, Panoply and xugrid open it directly. + The MF6 attributes go along with the data, and other tools ignore them. + It has a structured form (DIS) and a layered-mesh form (DISV, layered + DIS). + +**Adapters** hand the representation to another library's object model in +memory. They are not file formats. The library then reads, analyses, +plots and writes in its own way, e.g. UGRID netCDF through xugrid, GeoTIFF +through `rio.to_raster`. + +| Adapter | Gives | For | +|---|---|---| +| `to_ugrid()` / `from_ugrid()` | xugrid `UgridDataset` | regridding, mesh plotting, mesh operations on DISV and layered DIS | +| `to_rioxarray()` | DataArray with CRS and transform written (`.rio`) | rasters, GeoTIFF, reprojection, clipping (structured grids) | +| `to_cf()` | Dataset with CF attributes, bounds and `grid_mapping` | cf_xarray and other CF-aware Python tools | +| `to_modelgrid()` | flopy3 `StructuredGrid`/`VertexGrid` | flopy3 plotting and utilities, while still needed | + +The adapter names are placeholders. Writing an adapter's output with plain +`to_netcdf` should give something close to the ungrouped layout, not a +third format. If it doesn't, either the adapter or the ungrouped layout is +off, and that's worth fixing. + +**CF plays two roles.** It's metadata inside the representation: where CF +already has an answer (`units`, `standard_name`, `axis`, `grid_mapping`, +fill-value semantics), the representation uses it rather than inventing +its own. It's also part of writing files: fill values, time units, bounds. +Because of the first role, the writers and adapters don't each have to do +CF from scratch. + +### The two layouts compared + +| | grouped | ungrouped | +|---|---|---| +| Read by | Python (flopy4, the accessor); MF6 not yet | MF6, GIS and CF/UGRID tools, xarray/xugrid | +| Hierarchy | groups mirroring the tree | flat; identity in name prefixes and `modflow_input` | +| Coverage today | everything | one model's GRIDDATA/PERIOD arrays; the rest in text input | +| Coverage goal | everything | everything (to confirm with the MF6 developers) | +| Conventions | CF 1.8+ with groups, MF6 | CF, UGRID (mesh form), MF6 | +| Indexing | 1-based on disk, 0-based in memory | 1-based (UGRID `start_index = 1`) | -**Native** is the representation written as is: `DataTree.to_netcdf` or -`to_zarr`, with only the encoding netCDF needs (bools, None, fill values). -MF6 never reads it. It's for round trips, caching and sharing a simulation -between Python tools. +### Ungrouped is a flattening -### MF6 NetCDF is a projection, not a serialization +The ungrouped layout is meant to cover the whole simulation eventually; it +has been developed in steps. So the lasting difference between the layouts +is only groups vs flattening, and the cleanest spec defines the ungrouped +layout as a mechanical flattening of the grouped one. Then there's one spec, +not two. -The MF6 NetCDF format and the representation can't simply be the same -thing. The difference isn't mainly the flat layout: the file is a sidecar -holding one model's grid-shaped arrays, and the namefile and package text -files still carry everything else. So the encoder projects: +Writing the ungrouped layout means: -- **Project:** keep one model's GRIDDATA and PERIOD array fields. - **Flatten:** turn package paths into variable names (`npf_k`) plus `modflow_input` (`//`). - **Renumber:** 0-based to 1-based. -- **Reshape:** split by layer where the layered mesh form needs it - (`npf_k_l1`, …). -- **CF encode:** fill values, time units, `grid_mapping`, CRS. - -The decoder reverses these steps. It parses the identity, rebuilds the -package nodes, and merges them into a tree whose other fields come from the -text input. This round trip is lossless for what the file covers, provided +- **Lay out layers:** a `layer` dim, or one variable per layer where the + layered mesh form needs it (`npf_k_l1`, …). Prefer the dim (it's what + CF/UGRID tools expect). If MF6 or a tool needs the split form, make it + an option on the writer, not a separate format. +- **Write CF:** fill values, time units, `grid_mapping`, CRS. + +Reading reverses these steps: parse the identity, rebuild the tree, and +(while coverage is partial) merge in the fields that come from text input. + +**Dimension names are the hard part of flattening a whole simulation.** +One model per file avoids it, but a whole simulation in one flat file means +two GWF models each with their own `nlay`/`ncpl`, a GWT model on the same +grid as its GWF model, exchanges between models, and list data needing +per-package entry dims. Without groups, every dim needs a prefix +(`gwf1_nlay`) or a rule for when dims are shared. That's the problem groups +solve, and the flattening rule has to answer it either way. A shared-grid +rule (GWF and GWT on one grid share dims) would keep the file usable in GIS +tools. + +**Today** the ungrouped layout covers only part of the model, so reading it +yields a partial tree, and the text input fills in the rest. That's a +projection of the representation, and it's lossless for what it covers if the identity can be parsed reliably. -The flat layout is reasonable for a file MF6 reads: - -- MF6's input data model loads one model at a time, keyed by - model/package/tag. -- UGRID defines one mesh topology per dataset. -- GDAL/QGIS, CDO and older NCO handle netCDF groups poorly or not at all. - -The flat file is a flattened tree, so we don't need groups there. What we -do need from the format is listed under "What the format could change". - -### Where the representation and MF6 NetCDF differ +### Where the representation and today's MF6 NetCDF differ -| | MF6 NetCDF format | Representation | +| | MF6 NetCDF format (today) | Representation | |---|---|---| -| Indexing | 1-based cell ids | 0-based; 1-based only in encodings and `sel_cell` | +| Indexing | 1-based cell ids | 0-based; 1-based only on disk and in `sel_cell` | | Layers (mesh) | one variable per layer (`npf_k_l1`, …) | a `layer` dim | | List (stress period) data | not covered | a sparse table, plus `dense()` | +| Options, scalars, simulation components | not covered | covered | | DISU, DIS2D/DISV2D/DISV1D | not covered | covered | | Scope | one model per file, flat | a `DataTree` of the whole simulation | | Time | input `time` per period, output `time` per step, with no stated link | both, with the mapping | | Indexes | none (files don't store them) | `CellIndex` and similar | | Package identity | the `modflow_input` path string | a structured type tag | +| Rotation | wrapped WKT2 `PROJCRS` only | `projection` coordinate, `x_world`/`y_world` | | Fill values | vary (`9.97e36`, `3e30`, `1e30`, `-2147483647`) | one per dtype | ### What the format could change -Not groups, and not list data (that would be nice to have, but isn't -needed). These would make the projection well-defined and its inverse -reliable: +To make the ungrouped layout a full, conventions-clean flattening of the +grouped one: +- **A flattening rule for the whole simulation**, including how dims are + named or shared across models. - **Structured identity:** separate model, package and tag attributes, or a - documented grammar for `modflow_input`, so the decoder doesn't have to + documented grammar for `modflow_input`, so the reader doesn't have to guess package names. +- **Coverage:** list data, options and scalars, simulation-level + components, DISU and DIS2D/DISV2D/DISV1D. +- **CF/UGRID validity as a stated goal:** a `layer` dim rather than split + variables (or a documented reason for splitting), and 2-D auxiliary + coordinates for rotated grids. - **Consistent fill values** per dtype. -- **A documented layer convention:** when variables are split by layer, and - how to reassemble them. - **A time mapping** between period-sized input time and step-sized output time. +- **Groups**, eventually, if MF6 is to read the grouped layout. -These come from reading the wiki and need checking against the MF6 source -before anything is proposed upstream. +These come from reading the wiki and need checking against the MF6 source, +and the coverage goal needs confirming with the MF6 developers, before +anything is proposed upstream. ## Grid dims and coordinates diff --git a/xarray-export-plan.md b/xarray-export-plan.md index 6437e3f9..030ba7bb 100644 --- a/xarray-export-plan.md +++ b/xarray-export-plan.md @@ -31,8 +31,8 @@ round trips. 5. **The complete export is writable to netCDF.** It is distinct from MF6's NetCDF input format (`flopy4/mf6/netcdf.py`, `NetCDFModel`), and the docs must say which is which. `docs/dev/xarray-representation.md` - does: this is the hierarchical (grouped) encoding, and MF6 NetCDF is a - flat projection of the same representation. + does: this is MF6 NetCDF (grouped), and `netcdf.py` writes MF6 NetCDF + (ungrouped), a flattening of the same representation. 6. **Keep `protocols.py`; put the methods on `Package`/`Context` directly; delete `mixins.py`.** The protocols are cheap, the SDD describes them, and `isinstance(x, DataTreeConvertible)` is useful. The mixins are @@ -185,7 +185,8 @@ How this fits `xarray-package-plan.md`: - `flatten()`/`unflatten()` are views, like `dense()`. They belong to the package side (grid and data semantics), use only what's in the dataset, and never change the representation. Writing a flat view to - netCDF is part of the CF encoding, not a new format. + netCDF is a CF matter (compression by gathering, below), not a new + format. - `.grid` is the current accessor name. The package plan (step 2) renames it to something distinctive before more users depend on it, so rename these with it. @@ -259,9 +260,10 @@ Where we depart, and what to do about it: numbering. Keep it, and say so in the docstring. Also consider a horizontal-only flatten (`flatten(layers=False)` or similar): `(layer, row, col)` → `(layer, cell)`. That is xugrid's convention - exactly, and it puts DIS data in the same shape as DISV. The UGRID - encoder needs this step for layered DIS anyway, so write it once and - use it in both places. + exactly, and it puts DIS data in the same shape as DISV. Both writers + need this step for layered DIS anyway, so write it once and + use it in both places (the UGRID adapter and the ungrouped mesh + layout). 2. **MultiIndex by default.** The libraries disagree. xugrid skips it, which keeps the result netCDF-writable; uxarray keeps it. This leans toward no MultiIndex by default (`index=False`), or at least making @@ -275,8 +277,8 @@ Where we depart, and what to do about it: `(time, layer, nmesh_face)`, while the export uses `(layer, cell)`. That mismatch is within flopy4, independent of the libraries. The package plan settles it: readers produce the representation directly - and xugrid objects only come from the UGRID encoding (`to_ugrid()`). - So the readers move to `cell`, and `nmesh_face` is a UGRID encoding + and xugrid objects only come from the UGRID adapter (`to_ugrid()`). + So the readers move to `cell`, and `nmesh_face` is a UGRID adapter detail. **netCDF for the flat view: CF "compression by gathering"** (CF §8.2). @@ -339,4 +341,4 @@ References: - netCDF for flat data: CF `compress` encoding via cf_xarray (new dependency) or our own small implementation. - Cell dim name across export and output readers: leaning `cell` for - both, with `nmesh_face` only in the UGRID encoding (departure 4). + both, with `nmesh_face` only in the UGRID adapter (departure 4). diff --git a/xarray-package-plan.md b/xarray-package-plan.md index c2724365..3e512bd8 100644 --- a/xarray-package-plan.md +++ b/xarray-package-plan.md @@ -13,14 +13,20 @@ package later is a packaging decision rather than a redesign. it never imports flopy4 or flopy. 2. **Our own representation first, encodings second.** The in-memory form is plain `xr.Dataset`/`xr.DataTree` with custom indexes, close to MF6's - own concepts and lossless for every grid type. CF, UGRID (through - xugrid) and the [MODFLOW NetCDF format][mf6nc] are explicit conversions - to and from it, and none of them is the core. A fourth, **native** - encoding writes the whole tree as is (groups, full fidelity) for Python - round trips. MF6 NetCDF is a *projection* of the representation (one - model's grid arrays, flattened), not a serialization of it. See - `docs/dev/xarray-representation.md`, which explains how netCDF, CF, - UGRID, the MF6 NetCDF format and the representation relate. + own concepts and lossless for every grid type. Files and other + libraries' objects are explicit conversions to and from it, and none of + them is the core: + - **Files are [MF6 NetCDF][mf6nc] in two layouts of one spec.** + *Grouped* is the whole tree, lossless. *Ungrouped* is the same content + flattened. MF6 reads it, and it aims to be valid CF and UGRID, so GIS + tools open it directly. CF and UGRID are conventions MF6 NetCDF builds + on, not separate formats. + - **Adapters** (`to_ugrid`, `to_rioxarray`, `to_cf`, `to_modelgrid`) + hand the representation to xugrid, rioxarray, cf_xarray or flopy3 in + memory. Those libraries write their own files. + + `docs/dev/xarray-representation.md` explains how netCDF, CF, UGRID, the + MF6 NetCDF format and the representation relate. 3. **The boundary is a data contract, not an API.** The export schema from `xarray-export-plan.md` phase 1 is the interface. flopy4 produces it; the package (the accessor, `dense()`, the encoders) consumes it, using @@ -124,43 +130,72 @@ docs, so the spec mainly collects it: child lists omitted. - **Fill values:** one convention per dtype (see the MF6 NetCDF notes below). -## Encodings - -Each is a pair of functions (representation ↔ encoded dataset) plus -accessor methods. None changes the representation. - -- **Native**: the representation as is, via `DataTree.to_netcdf`/`to_zarr` - with groups mirroring the tree, encoding only what netCDF needs (bools, - None, fill values). It covers everything (options, dims, list data, - arrays). MF6 never reads it; it's for Python round trips, caching and - sharing. The name is open (see "Open"). -- **MF6 NetCDF (structured and layered mesh)**: what `netcdf.py` writes - today, restated as an encoder from the representation. It's a - projection: keep one model's GRIDDATA/PERIOD arrays, flatten package paths - into names plus `modflow_input`, renumber to 1-based, split by layer where - the mesh form needs it, then CF encode. Readers reverse this and merge the - result into a tree whose other fields come from the text input, so an - MF6 NetCDF file opens as the representation. -- **UGRID via xugrid**: `ds..to_ugrid()` → `UgridDataset` for DISV and - layered DIS, and `from_ugrid` back. This is where `UgridDataset` appears, - not in `to_xarray`. -- **CF**: bounds (`x_bnds`/`y_bnds`/`z_bnds`, `rotated-grid-index-plan.md` - Next 3), `grid_mapping`, standard names. Mostly a subset of the structured - MF6 NetCDF encoding. -- **Binary output** (GRB/HDS/CBC): readers that produce the representation - directly. +## Files and adapters + +None of these changes the representation. Each is a pair of functions +(representation ↔ file or object) plus accessor methods. + +**Files: MF6 NetCDF, two layouts of one spec.** CF, UGRID and MF6 NetCDF +stack: CF is the base, UGRID extends it, and MF6 NetCDF adds attributes +that other tools ignore. So the ungrouped layout can be valid CF, UGRID and +MF6 NetCDF at once, and there's no separate "CF" or "UGRID" file format. + +- **Grouped**: the representation as is, via `DataTree.to_netcdf`/`to_zarr`, + with groups mirroring the tree and only the encoding netCDF needs (bools, + None, fill values, 1-based). It covers everything (options, dims, list + data, arrays). It's valid CF 1.8+ with groups. MF6 doesn't read it yet. + Today it's for Python round trips, caching, archiving and sharing. +- **Ungrouped (structured and layered mesh)**: what `netcdf.py` writes + today, restated as a flattening of the grouped layout. Package paths + become names plus `modflow_input`, indices go to 1-based, layers become a + `layer` dim (split variables only as a writer option, if MF6 or a tool + needs them), and then CF is written. Readers reverse this. While + coverage is partial (one model's GRIDDATA/PERIOD arrays), they merge the + result into a tree whose other fields come from text input. It's meant to + cover the whole simulation eventually (to confirm). CF/UGRID validity is + a goal: bounds (`x_bnds`/`y_bnds`/`z_bnds`, `rotated-grid-index-plan.md` + Next 3), `grid_mapping`, standard names, and UGRID mesh topology on the + mesh form. + +**Adapters: other libraries' objects, in memory.** Each library then writes +its own files (UGRID netCDF through xugrid, GeoTIFF through +`rio.to_raster`). Writing an adapter's output with plain `to_netcdf` should +give something close to the ungrouped layout, not a third format. + +- **`to_ugrid()` / `from_ugrid()`** → `UgridDataset` for DISV and layered + DIS. This is where `UgridDataset` appears, not in `to_xarray`. +- **`to_rioxarray()`** → a DataArray with CRS and transform written, for + structured grids. Rotation needs checking: rioxarray takes a rotated + affine transform, but its 1-D x/y coordinates assume no rotation. +- **`to_cf()`** → a Dataset with CF attributes, bounds and `grid_mapping`, + for cf_xarray and other CF-aware tools. It's mostly the same metadata + the ungrouped writer uses. +- **`to_modelgrid()`** → a flopy3 grid, while flopy3 still needs one + (replaces the inheritance in `utils/grid.py`, step 4). + +The names are placeholders. + +**Binary output** (GRB/HDS/CBC): readers that produce the representation +directly. ### Notes on the MODFLOW NetCDF format The format is ours to change too, so the representation work is a chance to -feed fixes back to MF6. The goal isn't to make the format a serialization of -the representation. It's a flat, per-model sidecar to the text input, and -that's reasonable for a file MF6 reads: the input data model loads one model -at a time, UGRID defines one mesh per dataset, and GIS tools handle groups -poorly. So don't push for groups, and treat list data as nice to have. The -goal is a well-defined projection whose inverse is reliable. That mainly -needs structured identity, consistent fill values, a documented layer -convention and a time mapping. +feed fixes back to MF6. The ungrouped layout has been developed in steps and +is meant to cover the whole simulation eventually (to confirm with the MF6 +developers). So the goal is one spec: the grouped layout is the +representation encoded, and the ungrouped layout is a fixed flattening of +it that stays valid CF and UGRID. Flat is right for the layout MF6 reads +today: MF6 loads one model at a time, UGRID defines one mesh per dataset, +and GIS tools handle groups poorly. Groups only matter if MF6 is to read +the grouped layout. + +The hard part of flattening a whole simulation is dimension names, not +variable names. Two GWF models each have their own `nlay`/`ncpl`; a GWT +model shares its GWF model's grid; exchanges span models; list data needs +per-package entry dims. Without groups every dim needs a prefix or a rule +for sharing. Settle that rule first. A shared-grid rule keeps the file +usable in GIS tools. Candidates, from reading the wiki (2026-11-03). They need checking against the MF6 source before anything is proposed upstream: @@ -168,8 +203,8 @@ the MF6 source before anything is proposed upstream: - **Per-layer variable splitting** in the layered mesh (`npf_k_l1`, `npf_k_l2`, `z_l`, each with a `layer` attr): a variable's identity is spread over names and attrs. xugrid and most CF tools would rather see a - `layer` dim. If that's needed for QGIS/GDAL, keep it as an encoding detail, - but say so in the spec. + `layer` dim. If that's needed for QGIS/GDAL, make it a writer option and + say so in the spec. - **No DISU**, and no DIS2D/DISV2D/DISV1D. The representation has to cover them regardless, so it shouldn't wait on the format. - **Inconsistent fill values**: `9.97e36` (GRIDDATA), `3e30` (PERIOD input @@ -182,8 +217,9 @@ the MF6 source before anything is proposed upstream: stated way to relate the two. The representation needs both (`per` and step time), so the format could carry the mapping. - **List (stress period) data isn't covered**; only GRIDDATA/PERIOD arrays - in READASARRAYS/READARRAYGRID packages. Low priority: the native encoding - covers list data, and the text input carries it for MF6. + in READASARRAYS/READARRAYGRID packages. Nor are options, scalars or the + simulation-level components. All in scope if the layout is to cover the + whole simulation. - **Rotation is only encoded through a wrapped WKT2 `PROJCRS`**; without a CRS, rotated structured grids lose `axis`/`standard_name`. The representation's `projection` coordinate carries xorigin/yorigin/angrot @@ -194,7 +230,7 @@ the MF6 source before anything is proposed upstream: ## xugrid -- **Optional, behind the UGRID encoding and DISV selection**, not in the core +- **Optional, behind the UGRID adapter and DISV selection**, not in the core representation. A `UgridDataset` is a wrapper type, not an `xr.Dataset`: it doesn't nest in a `DataTree`, and our registered accessors don't reach it. That conflicts with the export plan's "packages give a Dataset, @@ -202,7 +238,7 @@ the MF6 source before anything is proposed upstream: - **The `xugrid` branch** (May 2026, `b00d7a05`, merge-base `52daeeb0`) returns xugrid objects from `to_xarray`. Don't merge it as is. Its useful parts (dropping hand-written UGRID attrs, keeping CF bounds) go into the - UGRID encoder instead. + UGRID adapter and the ungrouped mesh writer instead. - **DISV selection** (`rotated-grid-index-plan.md` Next 1): output readers return `UgridDataArray` today. For the representation, prefer a plain dataset with an index (an `NDPointIndex` for nearest, or our own @@ -308,10 +344,10 @@ new repo. 6. **Lazy xugrid** everywhere in the subpackage, then decide whether it's an optional dependency. 7. **Update `docs/dev/sdd.md`** ("Xarray conversion", "NetCDF") to describe - the representation, the encodings, and the boundary. The current "NetCDF" - section ("adopt xugrid ... `.to_netcdf()` through the appropriate - accessor") is close to right, except that xugrid is one encoding, not the - writer. + the representation, the two MF6 NetCDF layouts, the adapters, and the + boundary. The current "NetCDF" section ("adopt xugrid ... `.to_netcdf()` + through the appropriate accessor") is close to right, except that xugrid + is one adapter, not the writer. 8. **Make `attrs_xarray.py` generic** as part of export plan phase 1, which already says to "make the generic layer complete". Add an import-linter contract so `flopy4.attrs_xarray`, `flopy4.protocols` and the generic @@ -347,7 +383,9 @@ timing. agree on the representation so datasets move between the tools. Don't contact anyone without asking. - **MF6 NetCDF format**: the notes above, as an issue or discussion on - modflow6, once checked against the source. Don't post without asking. + modflow6, once checked against the source. First confirm the intent: + is the ungrouped layout meant to cover the whole simulation, and would + MF6 ever read a grouped one? Don't post without asking. - **xugrid**: follow #35 / #373. Report DISV edge-tie behaviour if it differs from DIS (`rotated-grid-index-plan.md` Next 1). - **xattree**: ours, so no outside coordination, apart from where it lives @@ -357,17 +395,21 @@ timing. ## Open -- Name for the native encoding. Candidates: "MF6" (parallels "MF6 - NetCDF", "UGRID", "CF", but collides with MF6's own text input format, - which flopy also writes, and MF6 can't read it), "native", or the - package's name. +- The flattening rule for a whole simulation in the ungrouped layout: + how dims are prefixed or shared across models (see the format notes). +- Default layout per writer: ungrouped for model input and tool-facing + output, grouped when saving a whole simulation from flopy? Users should + only ever have to pick the layout. +- Adapter names and which ones to have (`to_ugrid`, `to_rioxarray`, + `to_cf`, `to_modelgrid`). The format names are settled: MF6 NetCDF + (grouped) and MF6 NetCDF (ungrouped). - Package name, and whether the subpackage is `flopy4/xr/` or `flopy4/mf6/xr/`. - Accessor name. - Whether the time code (TDIS → time coordinate, TAS evaluation) is part of the package or stays in flopy. Leaning package, since output readers need it too. - Does the package know DFNs (for variable metadata, long names, units)? - Prefer not: the encoder takes the metadata from attrs that flopy puts on + Prefer not: the writers take the metadata from attrs that flopy puts on the representation, and devtools stays a flopy dependency only. - Representation for DIS2D/DISV2D/DISV1D (OLF/CHF). - Merge `xarray-export-plan.md` onto develop (or into the working tree) so From d15e00ee97787a2f6b69933bcb0890d787c35e4a Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 08:41:57 -0800 Subject: [PATCH 05/77] consolidate xarray planning docs Fill in the representation spec's outline (grid dims and coordinates, tree and type tags, period data, time, attributes and fill values) from the plans where each piece was designed, and move the MF6 NetCDF format notes into it. The package plan keeps direction and all open questions; the export plan keeps the implementation steps. Check in the plans the docs still depend on, and drop references to local-only files. Co-Authored-By: Claude Opus 5.5 --- combined-index-notes.md | 177 ++++++++++++++++++++ datetime-periods-plan.md | 141 ++++++++++++++++ docs/dev/xarray-representation.md | 209 ++++++++++++++++++++++-- rch-evt-layer-plan.md | 73 +++++++++ rotated-grid-index-plan.md | 259 ++++++++++++++++++++++++++++++ tas-interpolation-plan.md | 188 ++++++++++++++++++++++ xarray-export-plan.md | 43 ++--- xarray-package-plan.md | 225 ++++++++------------------ 8 files changed, 1125 insertions(+), 190 deletions(-) create mode 100644 combined-index-notes.md create mode 100644 datetime-periods-plan.md create mode 100644 rch-evt-layer-plan.md create mode 100644 rotated-grid-index-plan.md create mode 100644 tas-interpolation-plan.md diff --git a/combined-index-notes.md b/combined-index-notes.md new file mode 100644 index 00000000..77648026 --- /dev/null +++ b/combined-index-notes.md @@ -0,0 +1,177 @@ +# Combined grid index: rebuilding on `CellIndex` + +Notes from the dead-code review (develop @ 613b5c50). `flopy4/mf6/indexes.py` +is being deleted. This covers what it would take to bring back a combined +index on top of `CellIndex`, and how that compares to what we do now. It +builds on `rotated-grid-index-plan.md` ("Why not an index (yet)"), which +already prototyped a z index. + +## What `indexes.py` did + +- `alias(ds, "nlay", "lay")`: a `PandasIndex` over `RangeIndex(n)`, giving a + dim named after its size (`nlay`) a label coordinate (`lay`). This was for + when array dims were `nlay`/`nrow`/`ncol`/`nper`. +- `MetaIndex`: wraps several indexes as one. `create_variables` and `sel` + delegate to each child and merge the results with `merge_sel_results`. +- `grid_index`/`time_index`: build those for `lay`/`row`/`col` and `kper`. + +None of it is needed now. Dims are `layer`/`row`/`col` (#469), and selecting +by position is `isel`. It also doesn't fit the current xarray `Index` API: +`from_variables` has no keyword-only `options`, and it returns a dict +instead of an `Index`. + +The thing to take from it is the shape, one index delegating to child +indexes, not the code. + +## Why a plain meta-index adds nothing today + +`MetaIndex.sel` asks each child separately and merges what they return. +xarray already does exactly that across separate indexes. `da.sel(x=..., y=...)` +sends `x` to the `CellIndex` on `col` and `y` to the one on `row`, then merges +the two dim indexers. Wrapping the two `CellIndex`es in a `MetaIndex` would +behave the same, with more code. + +A combined index only helps when the children **depend on each other**: one +label's answer depends on what another label chose. On a DIS grid, z is the +case that needs it. The layer containing z depends on which `(row, col)` x/y +picked (see the `ZIndex` prototype in `rotated-grid-index-plan.md`). Merging +independent answers gives wrong results for scalar x/y and raises +`IndexError` for lists. + +So "rebuild the combined index" really means a **`GridIndex` owning `x`, +`y`, and optionally `z`/`z_top`/`z_bot`**, which answers all three labels in +one `sel` call. + +## What a `GridIndex` on `CellIndex` would need + +### Structure + +- `class GridIndex(xr.Index)` holding: + - `xindex: CellIndex | None` (on `col`) + - `yindex: CellIndex | None` (on `row`) + - `z_top`/`z_bot`: `Variable`s on `(layer, row, col)`, or `None` for + arrays without z (1-layer arrays, RCHA/EVTA; see `rch-evt-layer-plan.md`). + - `z` (centres), which it owns so that `sel(z=...)` reaches it, but which + it answers by containment over the bounds, not by centre. +- Composition, not inheritance: x and y keep all their behaviour from + `CellIndex`, and `GridIndex` delegates per dim, as `MetaIndex` did. + +### Methods + +- **`from_variables(cls, variables, *, options)`**: build the children with + `CellIndex.from_variables({"x": ...}, options={"bounds": options["x_bounds"]})` + and the same for y. Keep z bounds as variables. Accept any subset of the + coordinates. +- **`create_variables`**: the children's variables, plus `z`/`z_top`/`z_bot` + wrapped as index variables. +- **`sel(labels, method=None, tolerance=None)`**, the reason this exists: + 1. Split `labels` into x, y and z. + 2. Ask `xindex.sel`/`yindex.sel` for the `col`/`row` indexers. They handle + containment, edge ties, slices and `method="nearest"` as now. + 3. If z is given: `isel` the z bounds by those indexers, then call + `vertical.layer_at(z, z_top_sel, z_bot_sel)` to get the layer per chosen + vertical column. Raise `KeyError` where no layer contains z, to match + `.grid.sel`. + 4. Return one `IndexSelResult`. For points (DataArray labels sharing a + dim), all three indexers are vectorized over the points dim. For scalar + x/y with scalar z, they're ints. For z alone, `layer` is a `(row, col)` + indexer (the horizontal slice, as `sel_z`). + - `method`/`tolerance` apply to x/y only. z is always containment, as now. +- **`isel(indexers)`**: + - Delegate `col`/`row` to the children. + - `isel` the z bounds on `layer`/`row`/`col`. + - When a dim becomes scalar, drop that part and keep the rest. This is the + main thing to verify: with one index across three dims, returning `None` + drops the **whole** index, so `isel(layer=0)` would also lose x/y + selection. Prototype what xarray does when `isel` returns an index that + no longer owns some of its coordinates. +- **`join`/`reindex_like`/`equals`**: delegate per dim to the children + (keeps #486's edge checks), and compare the z bounds. +- **`concat`**: + - Along `col`/`row`: delegate. + - Along `layer`: concatenate the z bounds. + - Along anything else (e.g. `time`): the index is unchanged. +- **`roll`, `__getitem__`, `_replace`, `copy`, `__repr__`**, pickling: as + `CellIndex` does. + +### Wiring + +- `index_xy` sets one `GridIndex` with + `obj.set_xindex(["x", "y", "z", "z_top", "z_bot"], GridIndex, x_bounds=..., y_bounds=...)` + instead of two `CellIndex`es, falling back to x/y only when z isn't there. +- `CellIndex` stays, as the child and for any 1-D-only use. +- `.grid.sel(..., z=...)` and `sel_world(x, y, z)` keep working. They can + move to plain `sel` once it gives the same answers. +- Reopening from netCDF drops the index either way. `index_xy` rebuilds it + (rotated-grid plan, Next item 2). + +### Tests + +- Re-run every `CellIndex` test (`test/mf6/test_mf6_component.py`, around + `:2686`) against `GridIndex`: containment, ties, outside → `KeyError`, + slices, nearest, align/concat edges, roll. +- Then add: + - `sel(z=)` alone vs `sel_z` + - `sel(x=, y=, z=)` vs `.grid.sel` for scalars and point DataArrays + - `isel(layer=k)` then `sel(x=, y=)` + - arrays without z + - chunked (dask) arrays +- Pin xarray. Multi-coordinate custom indexes and `IndexSelResult` are + public but still changing. + +### Rough size + +`CellIndex` is about 130 lines, because most of an `Index` subclass is +keeping state right through `isel`/`join`/`concat`/`roll`. A `GridIndex` +that delegates x/y is probably 150–250 more, mostly the z bookkeeping, plus a +test file of about the same size as `CellIndex`'s. DISV isn't covered: it has +no `CellIndex`, and horizontal containment there goes through xugrid. + +## Current approach vs a custom combined index + +**Current:** one `CellIndex` per horizontal coordinate. z, world coordinates +and their combinations are free functions in `utils/grid.py` and +`utils/vertical.py`, on the `.grid` accessor. + +| | Current (per-coordinate `CellIndex` + accessor) | Combined `GridIndex` | +|---|---|---| +| x/y containment | `da.sel(x=, y=)` | Same | +| z | `.grid.sel(z=)`, `.grid.sel_z`, `.grid.mask_z`; `da.sel(z=)` raises `KeyError` | `da.sel(z=)` and `da.sel(x=, y=, z=)` | +| World coordinates on a rotated grid | `.grid.sel_world`, which converts and then calls `sel` | Unchanged; still an accessor function (world x/y aren't 1-D) | +| Discoverability | Users have to know `.grid` exists (covered in the selection guide) | Plain `sel`, what xarray users try first | +| Code to maintain | `CellIndex` plus small functions over plain coordinates | `CellIndex` plus a three-dim index whose state must survive every xarray operation | +| Behaviour through `isel`/`concat`/`align`/arithmetic | z coordinates are unindexed, so xarray handles them as ordinary coordinates; nothing to get wrong | Every operation goes through our code; partial `isel` across three dims is the risk | +| xarray API exposure | One-dim `PandasIndex` subclass (fairly stable) | Multi-coordinate custom `Index` (experimental, evolving) | +| Failure modes | A missing accessor call gives a clear `KeyError: no index for 'z'` | A bookkeeping bug gives silently wrong cells | +| Arrays without z, DISV | Same functions, branch on what's present | Needs an "x/y only" mode; DISV still separate | +| netCDF round trip | `index_xy` re-attaches the indexes | Same | +| Capability | Everything | Nothing new; a nicer spelling of the same selections | + +**Recommendation:** keep the current approach. A combined index only adds +the `da.sel(z=...)` spelling, and in exchange takes on the hardest part of +custom xarray indexes: keeping state right across several dims through +partial `isel`. Revisit it if: + +- users ask for plain `sel(z=...)`; or +- the DISV/xugrid work ends up needing its own index anyway, so one design + can cover both; or +- xarray settles its multi-coordinate index API (and ideally gains a + documented way to keep part of an index through `isel`). + +Unrelated to the index design: `alias` was 0-based (`RangeIndex(n)`), the +same as `isel` now. Keep selection 0-based everywhere and convert only at +the MF6 file boundary. One 1-based exception already exists: +`StructuredGrid.to_xarray` gives `layer` a 1..nlay coordinate +(`utils/grid.py:504`, and `:657`, `:1171`), so `ds.sel(layer=1)` there means +`isel(layer=0)`. This is the file boundary, not an exception: those +datasets are what `NetCDFModel` writes as MF6 netCDF input. They match +MF6's own netCDF export, which defines a `layer` int variable of `1..nlay` +with the same attrs (`DisNCStructured.f90`, `DisNCMesh.f90`, `DisvNCMesh.f90`, +modflow6 @ 48f34498f3). Keep it. + +Bug found along the way: `vertical.sel_z` replaces `layer` with the chosen +**positions** (`vertical.py:99`). On a dataset with 1-based `layer` labels +(e.g. MF6 netCDF output opened with `xr.open_dataset`), the result's +`layer` is off by one: selecting z in the second layer returns its value +with `layer=1`. It should keep the labels when `layer` has them +(`obj[dim].values[idx]`), and positions only when it has none. diff --git a/datetime-periods-plan.md b/datetime-periods-plan.md new file mode 100644 index 00000000..238df717 --- /dev/null +++ b/datetime-periods-plan.md @@ -0,0 +1,141 @@ +# Datetime period keys (#518), then `create_tdis` (#226) + +## Goal + +Let period-keyed package data (fill-forward fields: CHD/WEL/... lists, +RCHA/WELG/... arrays, OC) be keyed by datetimes as well as 0-based period +indices, e.g. + +```python +chd = Chd(stress_period_data={ + "2000-01-01": [((0, 0, 0), 1.0)], + pd.Timestamp("2000-03-01"): [((0, 0, 0), 2.0)], +}) +``` + +and then let a simulation build its TDIS from the datetimes its packages use +(imod-python's `create_time_discretization`), the time half of #226. + +## Current state + +- `Package._check_periods()` (`flopy4/mf6/package.py`, run from + `Package.write`) raises on any non-integer, negative, or >= NPER period + key. Before it, a datetime key was silently dropped on write, and keys + past NPER were written as PERIOD blocks MF6 never reads. +- Keys are stored as given; nothing normalizes them at construction. +- `Tdis.from_timestamps` / `Time.from_timestamps` build a TDIS from N + timestamps -> N-1 periods (the last timestamp is the end time). +- `Tdis.start_date_time` is an ISO string; `time_units` may be None. + +## Design + +**Keep datetime keys as given, resolve to period indices late.** TDIS stays +the one source of truth for what period n means. `create_tdis` doesn't +rewrite package keys; it only builds a TDIS that the keys line up with. +Datetime keys therefore also work without `create_tdis`, against a +hand-built TDIS with `start_date_time` set. + +### Key normalization (construction / reassignment) + +- Accept `datetime.datetime`, `datetime.date`, `np.datetime64`, + `pd.Timestamp`, and ISO strings; normalize all to `pd.Timestamp` (a + sortable, hashable, `==`-sane type with nanosecond precision). +- Integers (`int`, `np.integer`, not `bool`) stay as `int`. +- Strings: parse with `pd.Timestamp`; reject digit-only strings (`"0"`, + `"3"`) with an error rather than guess between "period 0" and a date. +- **No mixing** ints and datetimes within one package (all its + fill-forward fields together, e.g. RIVG's stage/cond/rbot), raising at + construction. Fill-forward order and "which period is first" are + ambiguous otherwise. Different packages in one simulation may differ. +- Hook: wherever fill-forward dicts are coerced, i.e. `_coerce_list_field` + (lists, `PeriodRows`) and `_init_period_arrays` (arrays, including the + dense `(nper, ...)` split in `period_arrays.split_periods`, which already + produces ints). One helper, e.g. `period_arrays.normalize_period_key`. +- Audit `int(kper)` casts (`period_arrays.py:182`) so a datetime can never + be coerced to an integer silently. + +### Resolution against TDIS + +- New `Tdis.period_index(t) -> int` (in `TdisMethods`), and a vectorized + `period_starts() -> pd.DatetimeIndex`: `start_date_time` plus the + cumulative sum of `perlen` in `time_units`. +- Errors when: `start_date_time` is unset; `time_units` is None/"unknown"; + `t` isn't exactly a period start (say which periods it falls between); + `t` is at or after the end of the last period. "years" is 365.25 days + (check against MF6 docs before committing to that). +- `Package._period_index(key) -> int`: ints pass through; datetimes go + through the TDIS found via the package's parent chain (the simulation). + No TDIS reachable -> error naming the package and key. + +### Write path + +- `_check_periods` widens: datetimes are valid iff they resolve via + `_period_index`; then the existing negative / >= NPER checks apply to + the resolved index. +- `_unstructure_package` (`converter/egress/unstructure.py`) maps every + fill-forward key through `value._period_index` before collecting into + `spd_period` / `readarray_period`, so the `kper + 1` block numbering and + sort are on ints. Two datetime keys resolving to the same period (only + possible with sub-second differences) -> error. +- `external.py` names external files `_p{kper + 1}`; it must see the + resolved index too. +- `_normalize_kper`'s `"*"` wildcard is unreachable after the write-time + check (nothing produces it); delete it in this change. + +### `Simulation.create_tdis` (#226) + +```python +sim.create_tdis(end="2001-01-01", nstp=1, tsmult=1.0) +``` + +- Collect datetime keys from every package in every model (and exchanges), + plus `end`, and build `self.tdis = Tdis.from_timestamps(...)` with + `time_units="days"`, `start_date_time` = earliest key. +- `end` is required (keys give period starts only), must be after the + latest key. +- Integer-keyed packages are allowed alongside but contribute no times; + warn if the resulting NPER is below one past their largest key, since + their meaning changes with the new TDIS. +- Optionally accept extra `times=` to split periods with no stress change + (imod allows this, e.g. for output timing). +- Time-series (TS6/TAS6) times are numeric offsets from simulation start, + not period keys; out of scope. + +### Dense / xarray views (follow-up) + +`Package.period_array`, `period_lists` dense tables and `to_xarray` derive +NPER from `resolve_dims("nper")` or `max(keys) + 1` and sort keys. With +datetime keys: + +- with a TDIS: resolve keys to indices first, then as today; optionally add + a `time` coordinate from `period_starts()`; +- without: sort the timestamps and index them 0..n-1, `time` coordinate + from the keys. + +Separate PR; the write path doesn't depend on it. + +## Steps + +1. Key normalization + no-mixing check, with construction tests (each + accepted type, digit-only string rejected, mixed keys rejected). +2. `Tdis.period_starts` / `period_index`, with unit tests (units, off-grid + date, past end, missing start date). +3. Write-path resolution (`_period_index`, `_check_periods`, + `_unstructure_package`, external file naming), drop the `"*"` wildcard. + Tests: datetime-keyed CHD and RCHA write the same files as their + int-keyed equivalents; off-grid and no-TDIS cases raise. +4. `Simulation.create_tdis`, with tests mirroring the imod 1D transport + example (packages first, TDIS last), and an mf6 run of one example if + the binary is available. +5. Follow-up PR: dense / xarray views with datetime keys. +6. Docs: a short example in the user guide; update #226 / #518. + +## Open questions + +- `datetime.date` vs `datetime.datetime`: treat a date as midnight? + (Probably yes, as pandas does.) +- Should `create_tdis` also be callable on a model-free basis + (`Tdis.from_packages(*pkgs, end=...)`), with the simulation method as a + wrapper? Useful for tests and for building one TDIS for several sims. +- Round-trip: a loaded simulation always has int keys. Do we ever want to + offer "re-key by date" (`pkg.with_datetime_keys()`)? Not needed now. diff --git a/docs/dev/xarray-representation.md b/docs/dev/xarray-representation.md index a80b524f..99a8171f 100644 --- a/docs/dev/xarray-representation.md +++ b/docs/dev/xarray-representation.md @@ -1,8 +1,11 @@ # The MF6 xarray representation -Status: **draft.** Only the first section is written. The rest is an outline, -to be filled in during `xarray-export-plan.md` phase 1. The design behind it -is in `xarray-package-plan.md` at the repo root. +Status: **draft.** This is the spec the export work builds to. Each +section says what exists today and what is planned. Where a point is still +open, it says so; the open questions are collected in +`xarray-package-plan.md` ("Open"). The design behind it is in +`xarray-package-plan.md`, the implementation steps in +`xarray-export-plan.md`, both at the repo root. The representation is the in-memory form of an MF6 simulation as xarray objects: what flopy4's `to_xarray` produces, what `from_xarray` accepts, and @@ -199,43 +202,217 @@ grouped one: - **A flattening rule for the whole simulation**, including how dims are named or shared across models. - **Structured identity:** separate model, package and tag attributes, or a - documented grammar for `modflow_input`, so the reader doesn't have to - guess package names. + documented grammar for `modflow_input`. Today the reader has to parse + `//` strings, and the package name depends on the + namefile (or the name FloPy generated). The format sets no naming rules + beyond CF, so this attribute is the only reliable identity. - **Coverage:** list data, options and scalars, simulation-level components, DISU and DIS2D/DISV2D/DISV1D. - **CF/UGRID validity as a stated goal:** a `layer` dim rather than split - variables (or a documented reason for splitting), and 2-D auxiliary - coordinates for rotated grids. + variables (`npf_k_l1`, `z_l`, each with a `layer` attr), or a + documented reason for splitting; 2-D auxiliary coordinates for rotated + grids, which without a CRS lose `axis`/`standard_name` today; and + bounds, `grid_mapping` and standard names throughout. - **Consistent fill values** per dtype. - **A time mapping** between period-sized input time and step-sized output time. - **Groups**, eventually, if MF6 is to read the grouped layout. -These come from reading the wiki and need checking against the MF6 source, +These come from reading the wiki (2026-11-03) and need checking against +the MF6 source, and the coverage goal needs confirming with the MF6 developers, before anything is proposed upstream. ## Grid dims and coordinates -To write. DIS from `rotated-grid-index-plan.md` "Coordinates on a DIS -array". A table for DISV and DISU still needs writing. DIS2D/DISV2D/DISV1D -are TBD. +### Dims + +| Grid | Cell dims | Flat view | Status | +|---|---|---|---| +| DIS | `(layer, row, col)` | `node` | done (#469) | +| DISV | `(layer, cell)` | `node` | done | +| DISU | `(node,)` | (already flat) | done | +| DIS2D, DISV2D, DISV1D (OLF, CHF) | TBD | TBD | open | + +- **Dims are named for the grid's axes**, not the DFN sizes (`nlay`, + `nrow`, `ncol`, `ncpl`, `nodes`), since they're typed in selections all + the time and xarray has no dim aliases. The DFN names appear only as + sizes in attrs, and as fallback dims for an array whose grid doesn't + resolve. The mapping is `flopy4.mf6.constants.GRID_DIMS`. +- **`cell`, not `node`, on DISV**, because in UGRID a node is a vertex. + `node` means an MF6 user node: a cell numbered in 3-D, as on DISU. +- **No index coordinates on `layer`/`row`/`col`/`cell`.** Positions are + the indices, 0-based (`isel`). Selecting by MF6's 1-based numbers is a + separate accessor method (`sel_cell`, planned). +- **Arrays not on the cells** get one dim per shape name: DIS `delr` on + `col`, `delc` on `row`. Arrays with one layer's worth of values (DIS/DISV + `top`, RCHA `recharge`) keep a size-1 `layer` dim. +- **The flat view** (`flatten()`/`unflatten()`, export plan phase 2b) + stacks the cell dims into `node`, layer slowest, so position `i` is user + node `i + 1`. It's a view, not stored. + +### Coordinates on a DIS array + +From `rotated-grid-index-plan.md`, where the design and its alternatives +are recorded. + +| Coord | Dims | Index | Rotated grid | Unrotated grid | +|---|---|---|---|---| +| `x` | `col` | `CellIndex` | local, zero-based | world | +| `y` | `row` | `CellIndex` | local, zero-based | world | +| `dx`/`dy` | scalar if uniform, else `col`/`row` | none | cell sizes (`dy` negative) | same | +| `projection` | scalar | none | grid mapping | grid mapping | +| `x_world`/`y_world` | `(row, col)` | none | world centres | omitted (same as `x`/`y`) | +| `z` | `(layer, row, col)` | none | cell-centre elevation | same | +| `z_top`/`z_bot` | `(layer, row, col)` | none | cell top and bottom | same | + +- **`x`/`y` are in the grid's CRS.** Unrotated, that's the base CRS, so + they're world coordinates. Rotated, it's a derived CRS (an EPSG:9624 + affine wrapping the base), so they're local. Rotation lives in the CRS, + not in the coordinate values. This is CF's `rotated_latitude_longitude` + pattern, and what MF6's own netCDF writer does. +- **`projection`** has the value `1`, and attrs `xorigin`/`yorigin`/ + `angrot` always, plus `crs_wkt` and `grid_mapping_name` when there's a + CRS. Gridded variables name it in their `grid_mapping` attr. + `xorigin`/`yorigin`/`angrot` aren't CF, but without a CRS they're the + only record of where the grid is. +- **`CellIndex`** is a `PandasIndex` that also keeps each cell's edges. + `sel` without a `method` gives the cell containing the point. A point on + an edge goes to the lower-numbered cell (west, north), as in flopy3. A + point outside raises `KeyError`. +- **z coordinates** come from `top`/`botm` (`vertical.z_coords`), with + `z_top[0] = top`, `z_top[k] = botm[k-1]` and `z_bot = botm`. They're on + every array on the grid's layers. `top` gets layer 0's. RCHA/EVTA arrays + don't get them unless the model has one layer, since the layer their + values go to depends on `irch`/`ievt` (`rch-evt-layer-plan.md`). + +### DISV and DISU + +- **DISV:** `x`/`y` on `cell`, world coordinates from `cell2d` (rotated by + `angrot` about the origin), unindexed. `z`/`z_top`/`z_bot` on + `(layer, cell)`. No containment index yet; whether DISV input gets an + `NDPointIndex`, its own index, or goes through `to_ugrid()` is open + (`rotated-grid-index-plan.md` Next 1). +- **DISU:** `x`/`y` on `node` from `cell2d`. Per-node z coordinates, from + DISU's own `top`/`bot`, are open. + +### What isn't stored + +Indexes aren't written to files. An array read back from netCDF has its +coordinates but needs `index_xy` (or an accessor method, planned) to get +its `CellIndex` back. Cell bounds (`x_bnds`/`y_bnds`/`z_bnds`, CF +`bounds`) are planned for the writers, so a file can be indexed without +the GRB. ## Tree, names and type tags -To write. +Planned, in export plan phase 1. Today `attrs_to_datatree` names nodes by +field and position (`models0`, `chd0`) and can't recover concrete types. + +- **Packages give an `xr.Dataset`, contexts an `xr.DataTree`.** The + simulation is the root. Models, exchanges, solutions and TDIS are its + children, and packages are children of their model. A package looks the + same whether exported alone or pulled out of its model's tree: it carries + its own grid coordinates. +- **Nodes are named by the component's name** (`gwf_1/npf`), not by field + and position. +- **Every node has a type tag**, `attrs["type"]`, so a reader can pick the + concrete class for `list[Model]` and union fields. Leaning toward the DFN + component name (`gwf-npf`), since it doesn't depend on Python; the class + path is the alternative (open). +- **Grid metadata in attrs:** the grid type and sizes (`nlay`, `nrow`, + `ncol`, `ncpl`, `nodes`) and `nper`, so `dense()`, `unflatten()` and the + writers never need the DIS object. +- **References** between components (a solution's models, an exchange's + models) are stored by name. +- **Empty child lists** are omitted. ## Period data -To write. Period arrays along `_period`, and list data as a sparse -table (see `period-lists-plan.md`). +Data given per stress period is stored **as given**: only the periods the +user set, never filled forward. Filling forward is a view (`dense()`). + +### Period arrays + +RCHA's `recharge`, WELG's `q` and the like (built, in `attrs_xarray`): + +- stacked along a `_period` dim, labelled by the periods given; +- a value from a time-array series is NaN, with the series' name in a + `_tas` coordinate along `_period` (`None` where a period has + no series). + +### Period lists + +CHD's, WEL's, SFR's stress period data (the sparse form is built, as +`to_xarray(sparse=True)`; export phase 1 makes it the stored form): + +- a sparse table along an `entry` dim, one entry per row; +- a `per` coordinate, 0-based; +- one coordinate per cellid component: `cellid_layer`, `cellid_row`, + `cellid_col` on DIS; +- a `setting` coordinate for keystring packages (SFR, LAK, MAW); +- a variable per value column and per aux name. An array column (EVT's + `pxdp`) gets a dim sized from its `shape`; +- a `_ts` coordinate naming the time series where a value comes + from one (the value is NaN), and a `boundname` coordinate. + +Unset values are `FILL_DNODATA` (floats), `FILL_INT64` (ints) or `None`. + +**Other row lists** (TDIS `perioddata`, `packagedata`, `connectiondata`) +get the same treatment along a `_row` dim (planned). + +### The dense view + +`dense()` (planned, on the accessor; today it's `to_xarray()`'s default) +places each period's rows by their key and fills forward: + +| Row key | Dense dims | Examples | +|---|---|---| +| one `cellid` | the grid | CHD, WEL, DRN | +| a foreign key to another table | that table's dim (`nreaches`), or the grid with `grid=True` | SFR, LAK, MAW, UZF | +| the package's own index | a dim named for it (`bndno`) | SPC | +| anything else (two cellids, no key) | no dense form; the table as is | HFB, MVR, OC | + +- Periods fill forward as in MF6: a block replaces the last one, but a + keystring setting holds until it's set again. +- Two rows in one cell raise, unless `duplicates="sum" | "first" | "last"` + merges them (lossy). +- All of this is decided from field metadata (`cellid`, `fk`, `pk`, + `union`, `shape`), with no per-package code. ## Time -To write. See `datetime-periods-plan.md`. +Partly open. + +- **`per`** is the 0-based stress period index, in list tables and in + `_period` labels. +- **Datetime period keys** (`datetime-periods-plan.md`, #518): packages + may be keyed by datetimes. Keys stay as given and are resolved to `per` + against TDIS when needed, so TDIS stays the one source of truth for what + period *n* means. +- **Time coordinates:** period start times come from TDIS + (`start_date_time` plus the cumulative `perlen` in `time_units`); output + is per time step. The representation should carry both and the mapping + between them (the period of each step). Not built. +- **Time-array series** stay NaN plus a name in the representation. + Evaluating them (MF6 averages over each step, STEPWISE or LINEAR only) is + a view; see `tas-interpolation-plan.md`. +- Whether the time code (TDIS to time coordinates, TAS evaluation) belongs + to the package or to flopy is open, leaning package, since output readers + need it too. ## Attributes and fill values -To write. +- **netCDF-safe attrs**, so the grouped layout is just `to_netcdf`: paths + as strings, references as names, `None` omitted, empty child lists + omitted. Bools are open: int8 with a flag, or a documented convention. +- **Variable metadata** (`long_name`, `units` where known) is put on by + flopy from the DFNs. The package side doesn't read DFNs. +- **Fill values:** one per dtype. Today the code uses `FILL_DNODATA` + (`3e30`, MF6's DNODATA) for unset floats in list data, `FILL_INT64` + (`-2147483647`, netCDF's int default) for ints, and `FILL_FLOAT64` + (`9.97e36`, netCDF's double default) when writing MF6 NetCDF. Which float + value the representation uses for "not set", and how it maps to + `_FillValue` on write, is open. [mf6nc]: https://github.com/MODFLOW-ORG/modflow6/wiki/MODFLOW-NetCDF-Format diff --git a/rch-evt-layer-plan.md b/rch-evt-layer-plan.md new file mode 100644 index 00000000..bf902f53 --- /dev/null +++ b/rch-evt-layer-plan.md @@ -0,0 +1,73 @@ +# Elevations for RCHA/EVTA arrays + +Follow-up to `z-attach` (z coordinates on arrays on the grid's layers). + +## Where things stand + +Package arrays on all of the grid's layers get `z_top`/`z_bot`/`z`, the +elevations of their cells (`Package._with_cell_centers`, via +`vertical.z_coords`). DIS/DISV `top` has a size-1 `layer` dim and gets +layer 0's, since it's the top layer's cells' tops. + +RCHA's `recharge`/`irch`/`aux` and EVTA's `surface`/`rate`/`depth`/`ievt`/ +`aux` also come out with a size-1 `layer` dim, `(per, layer=1, row, col)` +or `(per, layer=1, cell)`, but get no z coordinates, except in a 1-layer +model, where the shapes happen to match `botm`. Their values don't belong +to a fixed layer: + +- `irch`/`ievt` (period arrays, `shape=("ncpl",)`, `index=True`, fill + forward) give the layer each vertical column's recharge/ET goes to. + Without them it's layer 1 (0 here). +- Unless `fixed_cell` is set, recharge/ET in an inactive cell + (`idomain < 1`) moves down to the highest active cell below it. + +So the cell a value belongs to depends on the period, on `irch`/`ievt`, +on `fixed_cell` and on `idomain`. Layer 0's elevations would be wrong +wherever any of those move it. + +## Plan + +1. **Resolve the target layer**: a function, say + `rch_layer(irch, idomain, fixed_cell) -> DataArray` on + `(per, row, col)`/`(per, cell)`, giving each column's layer per period: + - `irch`'s value (check whether it's stored 1- or 0-based: it's an + `index=True` field), else 0; + - without `fixed_cell`, the first layer at or below it with + `idomain > 0`; + - where there's none, -1, and NaN elevations. + EVTA uses the same function with `ievt`. +2. **Take the elevations at that layer**: index `z_coords(top, botm)` by + the layer along `layer` (as `vertical.sel_z` does with `layer_at`), + giving `z_top`/`z_bot`/`z` on `(per, row, col)`, plus the chosen + `layer` as a coordinate, as `sel_z` keeps it. +3. **Attach them** in `_with_cell_centers` for RCHA/EVTA period arrays, + dropping the size-1 `layer` dim's place in the coordinates' dims. + Unlike other z coordinates these vary with `per`; check that + `sel_z`/`mask_z` handle the extra dim (they should broadcast). +4. **Changing `irch`/`ievt`**: the coordinates are computed when the array + is built, so they're as fresh as `to_dataarray`. Note it in the + docstring rather than caching. + +## Open questions + +- **Dry cells.** MF6 may also move recharge/ET below a cell that goes dry + during the run. That isn't knowable from input, so the input + coordinates describe where the input puts it. Check the MF6 source + (`gwf-rch`, `bnd` `nodeu`/`reassign`) for exactly what moves and when. +- **EVTA `surface`** is itself an elevation, like `top`, but the ET + surface, not a cell face. It gets the target cell's coordinates like the + others; its own value is data. +- **List-based RCH/EVT** have cellids with a layer, so they have a fixed + cell. Check whether their gridded form (`to_xarray(grid=True)`) gets z + coordinates; it should, from the cellid's layer. +- **DISU**, as for the rest of the z work, waits on horizontal selection. + +## Tests + +- `irch` given, varying by column: the elevations are those of each + column's `irch` layer. +- No `irch`: layer 0's, the same as `top`'s. +- An inactive target with and without `fixed_cell`: moved down vs. NaN. +- `irch` changing between periods: the coordinates differ by `per`. +- DISV EVTA with `ievt`. +- `sel_z`/`mask_z` on a recharge array. diff --git a/rotated-grid-index-plan.md b/rotated-grid-index-plan.md new file mode 100644 index 00000000..f9e269a5 --- /dev/null +++ b/rotated-grid-index-plan.md @@ -0,0 +1,259 @@ +# Grid coordinates, rotation and spatial selection + +This plan is the current version of two earlier ones: + +- **The spatial index design** (`docs/dev/spatial-index-design.md` on `origin/xarray-index`, Nov 2025–Feb 2026), with upstream #207 (structured grid index), #208 (`.grid` accessor) and #120 (closed background). It proposed spatial dims for DIS, a `.grid` accessor, a custom `GeospatialIndex` (from flopy PR 2654), containment-vs-nearest point queries, z and ball queries. +- **The rotated-grid selection plan** (this file's earlier form): local x/y plus a grid mapping for rotated DIS, world coordinates as an overlay. + +The "Where the old design stands" section below records what was kept, changed or dropped from the first. + +## Framing + +Keep the grid geometry pure and treat world coordinates as an overlay: + +- Index in the grid's own frame. `x`/`y` are 1-D on `col`/`row`, always indexed, rotated or not. +- The rotation lives in the CRS (or, without one, in grid-mapping metadata), not in the coordinate values. +- World coordinates for a rotated grid are derived, 2-D, auxiliary, and unindexed. + +This is the CF `rotated_latitude_longitude` pattern: 1-D `rlon`/`rlat` dimension coordinates in the rotated frame, a `grid_mapping` variable describing the transform, and 2-D `lon`/`lat` auxiliary coordinates. + +It's also what MF6 and our own writer already do: + +- `DisNCStructured` writes zero-based local x/y for rotated grids. When a CRS is configured, it wraps it in a derived projected CRS (EPSG:9624 affine) encoding xorigin/yorigin/angrot. +- `StructuredGrid._structured_dataset` (`flopy4/mf6/utils/grid.py`) mirrors that, using `wrap_rotated_crs` in `flopy4/mf6/utils/crs.py`. + +So "`x`/`y` are coordinates in the grid's CRS" holds everywhere: +- **Unrotated:** the grid CRS is the base CRS, so `x`/`y` are world coordinates (offset by the origin). +- **Rotated:** the grid CRS is the derived CRS, so `x`/`y` are local and zero-based. + +No renaming of `x`/`y` is needed. + +## Where the old design stands + +| Topic | Old design | Now | Status | +|---|---|---|---| +| DIS dims | `x`/`y`/`z` as dims, `col`/`row`/`layer` auxiliary | Reversed: dims `layer`/`row`/`col`, `x`/`y` 1-D indexed coords | Changed (#469). Stock xarray covers it with `set_xindex`. | +| Layer/row/col selection (#207's original ask) | Custom index mapping nodes to l/r/c | Dim names plus `isel` (0-based) | Done (#469). 1-based `sel` by MF6 numbers is still open; see accessor. | +| Rotation | Not covered | Local x/y, `projection` grid mapping, `x_world`/`y_world`, `sel_world` | Done (#485) | +| Custom index | `GeospatialIndex` as an xarray `Index` | `CellIndex` (a `PandasIndex` subclass with cell edges) for DIS; xugrid for DISV/DISU | Changed | +| Point semantics | Containment by default, `nearest_center` opt-in, `nearest` ambiguous | `sel` without a method is containment; `method="nearest"` is the nearest centre | Done (#485); answers the old open question by making `nearest` mean nearest centre | +| Edge ties | Lowest-numbered cell (flopy3) | Lowest-numbered: west (lower col), north (lower row) | Decided, matches flopy3. Keep DISV consistent if possible. | +| `.grid` accessor (#208) | The single entry point; DISV redirects to `.ugrid` | `.grid.sel_world`, plus `sel`/`sel_z`/`mask_z` on `z-attach` | Started (#487); `sel`/`sel_z`/`mask_z` in #490; see Next for the rest | +| CRS | "Future" | `crs_wkt`/`grid_mapping_name` on `projection` | Done (#485) | +| `z` selection, vertical profiles | 1-D layer-representative `z` or 3-D aux + custom index; open questions on z-only `sel` | One cell per vertical column: `sel_world(x, y, z)`, `sel_z`, `mask_z` over `z_top`/`z_bot` coords on all layered arrays | Functions merged (#488); attaching and accessor in #490 (`z-attach`). Plain `.sel(z=...)` needs a custom index (see Why not an index). DISU and `z_bnds` open | +| Ball/range queries | `query_ball` via cKDTree (3-D) / geopandas (2-D) | None | Open, accessor methods | +| DISV/DISU spatial selection | `GeospatialIndex` | Output is `UgridDataArray` (`ugrid.sel`); inputs have unindexed x/y on `cell`/`node` | Open | +| Quasi-3D `laycbd` z centres (comment on #207) | Needed for non-MF6 grids | Ignored | Open, only matters if non-MF6 grids are supported | + +## Coordinates on a DIS array + +| coord | dims | index | rotated | unrotated | status | +|---|---|---|---|---|---| +| `x` | `col` | `CellIndex` | local, zero-based | world | #469; `CellIndex` #485 | +| `y` | `row` | `CellIndex` | local, zero-based | world | #469; `CellIndex` #485 | +| `dx`/`dy` | scalar if uniform, else `col`/`row` | none | cell sizes (`dy` negative) | same | output since #469; package arrays and `StructuredGrid` #485 | +| `projection` (scalar) | — | none | grid mapping | grid mapping | #469; `crs_wkt` #485 | +| `x_world`/`y_world` | `(row, col)` | none | `to_world(...)` | omitted (same as `x`/`y`) | #485 | +| `z` | `(layer, row, col)` | none | cell-centre elevation | same | `StructuredGrid` only before; all layered arrays on `z-attach`, computed from the bounds | +| `z_top`/`z_bot` | `(layer, row, col)` | none | cell top and bottom | same | `z-attach` | + +- `projection`'s value is `1`. Its attrs are `xorigin`/`yorigin`/`angrot` always, plus `crs_wkt` (wrapped by `wrap_rotated_crs` when rotated) and `grid_mapping_name` with a CRS. Gridded variables name it in their `grid_mapping` attr, set by `index_xy`. +- `xorigin`/`yorigin`/`angrot` stay: they're non-standard, but they're the only record of placement without a CRS. +- The netCDF heads reader keeps the file's CRS attrs on `projection`; the grb still gives the placement. +- The z coordinates come from the discretization's `top`/`botm` (package arrays, `StructuredGrid`/`VertexGrid`) or the grb's TOP/BOTM (heads, budgets), via `vertical.z_coords`. DIS/DISV `top` (size-1 `layer`) gets layer 0's, since it's the top layer's cells' tops. Other arrays with one layer's worth of values, such as RCHA/EVTA's, don't get them unless the model has one layer, since their layer depends on `irch`/`ievt` (see `rch-evt-layer-plan.md`). Each has a `long_name` ("cell top elevation" etc.). DISV arrays get the same three on `(layer, cell)`. + +## Selection + +### `CellIndex` (DIS, all grids) + +A `PandasIndex` over the centres that also keeps each cell's `(lower, upper)` edges, built by `index_xy` from `dx`/`dy`, or midway between centres without them. + +- `sel` with no `method`: the cell containing the point. Scalar and point-wise (DataArray) labels both work. + - A point on an edge goes to the lower-numbered cell (west/north), as in flopy3. + - Points outside raise `KeyError`. + - The edges are widened by `tolerance`, a billionth of the axis's extent by default, so rotation rounding doesn't push an edge point outside. +- With `method=`, or a slice: selection by centre, as with `PandasIndex`. +- The index is kept by `isel` (slices and arrays), `roll`, pickle, deepcopy, `chunk`, `concat` along non-grid dims, and arithmetic between same-grid arrays. +- Aligning (`xr.align` inner/outer, `a + b`) and `concat` along `col`/`row` keep each cell's edges (#486). Aligning grids with a cell at the same centre but different edges raises `ValueError`. + +Why a `PandasIndex` subclass rather than the alternatives considered earlier: +- `pd.IntervalIndex`: prints and plots as intervals, doesn't serialise to netCDF, and has no `method="nearest"`. +- Float centres alone: the nearest centre can be a neighbouring cell when `delr`/`delc` vary. + +### `sel_world` (rotated DIS) + +```python +def sel_world(obj, x, y, z=None, **kwargs): + # rotated: to_local(...) through projection, then sel; otherwise just sel + # with z: then sel_z(..., outside="raise") on the vertical column +``` + +- On a rotated grid it converts to local coordinates through `to_local` (the inverse of `to_world`), then calls `sel`, so it gets containment from `CellIndex`. It no longer needs its own tolerance. +- Scalar and point-wise inputs. **Slices don't work**: a world box isn't a local box on a rotated grid, and `to_local` can't take a slice. For box selection, mask on `x_world`/`y_world` (`where`) in the accessor. + +### DISV/DISU + +Unchanged. +- MF6's mesh exports write world-coordinate vertices, and so do our grb readers since #469. +- Output is a `UgridDataArray`, so containing-face selection is xugrid's `ugrid.sel` / `sel_points`. That needs `numba_celltree`, which isn't in the dev env. +- Input arrays on `cell`/`node` have unindexed x/y. Wrap them as `UgridDataArray`, or give them an `NDPointIndex` (nearest centre only). Not decided. + +## Alternatives considered + +- **`xarray.indexes.NDPointIndex`** (KD-tree on 2-D world centres): nearest centre only, wrong next to a wide cell, lost on any `isel`. No for DIS; maybe for DISV/DISU inputs. +- **Custom `CoordinateTransformIndex` on world coordinates** (x/y world-valued, 2-D when rotated): an experimental xarray API, x/y changing shape with rotation, no slicing on rotated grids, and disagreement with MF6's netCDF. The overlay plus `CellIndex` gives the same selection. +- **`GeospatialIndex` as an xarray `Index`** (the old design): it isn't needed for DIS. It might still be the way to do DISV/DISU inputs, or ball queries, if xugrid doesn't fit. +- **Prior art:** rasterix `RasterIndex` is affine-only (uniform spacing). rioxarray keeps the transform/CRS in `spatial_ref` with 1-D x/y, the convention followed here. xproj records the CRS but doesn't transform. + +## Known limitations + +- Arithmetic between differently placed grids keeps the first's `projection` (xarray compares the value, always 1, not the attrs) and drops whichever world coordinate differs. Documented in `index_xy`. +- Arrays reopened from netCDF have their coordinates but not their indexes; they need `index_xy`. The z coordinates come back as they are, since they're unindexed. + +## z selection + +### Status + +- **#488 (`z-select`, merged 2026-10-17):** `flopy4/mf6/utils/vertical.py` with `layer_at`, `sel_z`, `mask_z`, over `z_top`/`z_bot` coordinates or bounds passed in. +- **`z-attach` (PR #490; checked 2026-10-19):** + - `vertical.z_coords(top, botm, dims)` builds `z`, `z_top` and `z_bot`. It replaces #488's `layer_bounds`. + - Attaches them to package arrays, `StructuredGrid`/`VertexGrid` and their arrays, and DIS/DISV heads and budgets. + - Adds `z=` to `sel_world`, and `sel_z`/`mask_z` to `.grid`. + - Tested on a rotated DIS grid, a `VertexGrid`, and real DIS/DISV heads. + - Also on the branch since: DIS/DISV `top` gets its layer's elevations; `long_name`s on the z coordinates; `.grid.sel`, `sel`'s indexers plus `z`; world-x/y slices and lists on a rotated grid raise, pointing to masking on `x_world`/`y_world` or point DataArrays; and the selection guide (Next item 6). +- **Open:** `z_bnds` in netCDF (with the `x_bnds`/`y_bnds` work), and DISU. + +A **vertical column** is the cells at one `(row, col)`, or one DISV `cell`, through all layers. ("Column" alone is ambiguous next to the `col` dim.) + +### The old open question goes away + +The old design asked what `sel(z=50)` means when z varies across (row, col), since there could be many matches. Within one vertical column, though, layers stack without overlapping, so **at most one cell contains a given elevation**. Two answers follow: + +- **With x/y:** the point is in one vertical column, so z picks at most one layer and the result is one cell, the same as horizontal containment. +- **z alone:** one cell per vertical column, so the result is a (row, col) map, a horizontal slice at that elevation. Vertical columns where z is above the top or below the bottom are NaN. This is the useful "head at elevation 50" map, without interpolation. + +So z needs neither a layer-representative 1-D z (which is wrong wherever layers aren't flat) nor a "return all matches" mode. + +### Why not an index (yet) + +xarray's `sel` passes each keyword to the index owning that coordinate. `x` and `y` have `CellIndex`es, on `col` and `row`; `z` has none, and the default `PandasIndex` only takes 1-D coordinates, while `z` is on `(layer, row, col)`. So `head.sel(z=...)` raises `KeyError: no index found for coordinate 'z'`, alone, with `x`/`y`, or even on one vertical column. (On one column, `set_xindex("z")` works, but `method="nearest"` then means the nearest centre, not the containing cell.) + +So z selection is free functions over `z_top`/`z_bot`, on the accessor: +`.grid.sel(..., z=...)` (`sel`'s indexers, then `sel_z` on what's left), `.grid.sel_world(x, y, z)`, `.grid.sel_z`, `.grid.mask_z`. + +**A custom index could make plain `.sel(z=...)` work.** This section used to say it couldn't, because neither result is a rectangular subset. That's wrong: `Index.sel` can return a vectorized indexer. Prototyped on xarray 2025.12 (2026-10-19): + +```python +class ZIndex(xr.Index): + # owns z_top and z_bot: da.set_xindex(["z_top", "z_bot"], ZIndex) + @classmethod + def from_variables(cls, variables, *, options): + return cls(xr.DataArray(variables["z_top"]), xr.DataArray(variables["z_bot"])) + + def sel(self, labels, **kwargs): + idx, found = layer_at(labels["z_top"], self.z_top, self.z_bot, "layer") + return IndexSelResult({"layer": idx.variable}) # a layer per (row, col) +``` + +- **z alone works:** `da.sel(z_top=12.0)` gives the horizontal slice on `(row, col)`, a different layer per column, the same as `sel_z`. +- **z with x/y doesn't:** xarray asks each index separately and merges their dim indexers. The z index picks a layer for every column of the whole grid, not knowing which column x/y chose. Then: + - scalar x/y give a wrong result on `(row, col)`; + - lists raise `IndexError: Dimensions of indexers mismatch`. + +So a working `.sel(x=, y=, z=)` needs **one index owning `x`, `y`, `z`, `z_top` and `z_bot`**, which sees all three labels in one `sel` call: horizontal containment first, as `CellIndex` does, then the layer in the chosen columns. It would replace the two `CellIndex`es on DIS arrays. Work and open questions: + +- **Keyword:** users want `sel(z=...)`, so the index owns `z` too (the centres) and answers `z` labels with containment over its bounds, not with the centres. +- **Partial selections:** `sel(x=...)` alone, `sel(y=..., z=...)`, slices of x/y (containment for scalars, by centre for slices, as now), and `method="nearest"` (x/y only). +- **Everything `CellIndex` keeps working through:** `isel` (slices, arrays, and a single layer, after which z is 2-D), `align`/`concat` along `col`/`row` with edges (#486), `roll`, pickle, `chunk`, arithmetic. +- **Arrays without z:** 1-layer arrays and RCHA/EVTA (see `rch-evt-layer-plan.md`) still need x/y selection, so the index must work with or without the z coordinates, or there are two index types. +- **DISV:** no `CellIndex`; horizontal selection goes through xugrid. The index would need a cell-containment structure for `(cell)` too, or z stays accessor-only on DISV. +- **netCDF:** like `CellIndex`, it isn't saved; reopened arrays need `index_xy` (Next item 2). +- **Experimental API:** custom indexes over several coordinates and `IndexSelResult` are public but still evolving in xarray. Pin and test. +- **Payoff:** `.sel(x=, y=, z=)` and `.sel(z=)` read naturally, but they add no capability over `.grid.sel`. Do it only if users ask for plain `sel`, or with the DISV/xugrid work if that needs its own index anyway. + +### Data: z bounds on arrays + +Containment needs each cell's top and bottom, not its centre, so two unindexed coords sit alongside `z`: + +- `z_top` and `z_bot` on `(layer, row, col)` (DIS) or `(layer, cell)` (DISV), with `z_top[0] = top`, `z_top[k] = botm[k-1]` and `z_bot = botm`. +- Why two coords rather than a CF `z_bnds`: a DataArray's coords can only use the array's own dims, so a `(layer, row, col, bnd)` coordinate can't be attached to a `(layer, row, col)` array. Write them out as `z_bnds` with `z:bounds = "z_bnds"` when saving to netCDF (with the `x_bnds`/`y_bnds` work), and read them back into the two coords. +- Sources: `Dis`/`Disv` `top`/`botm` for package arrays, and the grb's TOP/BOTM for heads/budget arrays (`get_coords` and the heads reader). +- They're plain numpy, 2× the size of one layer stack. That's small next to a time series. `z` (the centres) is computed from them, so the two can't disagree. +- They follow `isel(layer=...)` and `sel(x=, y=)` for free, since they share the array's dims. So after a horizontal point selection, a column keeps its own bounds. + +### API + +On `.grid`: + +- **`sel(indexers=None, method=None, tolerance=None, drop=False, **indexers_kwargs)`:** `sel` with `z` taken out first, then `sel_z(..., outside="raise")` on the result: the cell containing z in each vertical column the other indexers leave. x/y are in the grid's frame, as with `sel`; `method`/`tolerance` apply only to the other indexers. `sel_world` is this after converting world x/y. +- **`sel_world(x, y, z=None, **kwargs)`:** an optional `z` added to the existing method. It selects horizontally as now, then picks the layer whose `z_bot <= z <= z_top`. With DataArray x/y/z sharing a dim, it works point-wise: horizontal `sel` gives `(point, layer)`, then the layer is chosen per point by vectorized `isel`. With `method="nearest"`, only the horizontal pick is nearest; z is always containment. A nearest-in-z without bounds is the cell-centre `z`, which isn't what anyone asking for an elevation means. +- **`sel_z(z, outside="nan")`:** the horizontal slice. It returns the array with `layer` replaced by the containing layer per vertical column: `(…, row, col)` for DIS, `(…, cell)` for DISV, with NaN where no cell contains z (`outside="raise"` raises `KeyError` instead). The chosen layer is kept as a `layer` coord, so users can see where it came from. `z` can also be a DataArray, on the vertical columns' dims (e.g. a surface) or on a points dim. +- **`mask_z(zmin=None, zmax=None, how="overlap")`:** a boolean on `(layer, row, col)` for use in `where`; a missing end is open. `how` is `"overlap"` (the cell intersects the range; the default, since "which cells touch the water-table zone" is the usual question), `"within"` (the cell is entirely in the range), or `"center"` (`z` is in the range). Thickness-weighted fractions are out of scope. +- **Vertical profile:** no new method. `sel_world(x, y)` already returns the `(…, layer)` column, and with `z`/`z_top`/`z_bot` attached it carries its elevations, which is what a profile plot needs. A `plot_profile` helper can come later. + +### Rules + +- **Ties:** a z exactly on a layer interface goes to the upper (lower-numbered) layer, matching flopy3's `StructuredGrid.intersect` (it scans layers from the top, inclusive at both ends, and takes the first match) and our horizontal rule. The model top and bottom are inclusive. +- **Outside:** for a point outside, `sel_world` raises `KeyError`, as horizontal selection does. `sel_z` gives NaN for those vertical columns. Point-wise selection with some points outside raises too. A `drop`/NaN option can be added if anyone asks. +- **idomain:** selection is geometric. Inactive (0) and vertical pass-through (-1) cells are found like any other, and their values are whatever the array holds (NaN/1e30 in heads). Zero-thickness cells never win a tie against the cell above them, because the upper cell matches first. +- **Not covered:** dry or saturated-thickness selection (a water-table-aware z) is hydraulics, not geometry. Make it a separate method if wanted. Transient z (CSUB subsidence) is out of scope; bounds are static. +- **Rotation:** z isn't rotated, so this composes with `sel_world`'s local/world conversion unchanged. +- **Units:** z is in model length units. There's no vertical CRS handling. + +### DISV and DISU + +- **DISV:** the same, on `(layer, cell)`. The horizontal step goes through xugrid (see DISV/DISU), and the z step is identical, so write the layer pick once over a trailing `layer` dim. +- **DISU:** nodes have their own top/bot and there's no layer dim, so "the vertical column" is the set of nodes whose footprints contain the point. A point query is the horizontal containment and then `bot <= z <= top` among those nodes. `sel_z`/`mask_z` become masks over nodes. Defer this until DISU horizontal selection exists. + +## Done + +1. **#469:** layer/row/col dims, local 1-D `x`/`y` for rotated DIS, the `projection` coordinate, grb readers that honour `angrot`. +2. **#485 (`rotated-world`):** + - `x_world`/`y_world`, `to_local`, `sel_world`. + - `crs_wkt`/`grid_mapping_name` on `projection`; the heads reader keeps the file's CRS. + - `CellIndex` containment selection on all DIS arrays; `dx`/`dy` on package arrays and `StructuredGrid`. +3. **#486 (`cellindex-align`):** `CellIndex` keeps cell edges through `align` and `concat` along `col`/`row`. +4. **#487 (`grid-accessor`):** the `.grid` accessor, with `sel_world`. +5. **#488 (`z-select`):** z-selection functions. **#490 (`z-attach`)** attaches them and adds them to the accessor; see z selection. + +## Next + +In the order I'd take them: + +1. **DISV/DISU spatial selection.** + - Add `numba_celltree` to the pixi dev env. + - Test `ugrid.sel` / `sel_points` on the `UgridDataArray`s from our grb and heads readers, which now carry `z`/`z_top`/`z_bot` (check `ugrid.sel` keeps them, so `sel_z` works after it). + - Check xugrid's behaviour for a point on an edge between faces: can DISV match DIS ties (lower-numbered cell), or does it need a wrapper? + - Decide what DISV/DISU *input* arrays get (a `UgridDataArray` wrapper or `NDPointIndex`), and how `.grid` dispatches on dims: `row`/`col` structured, `cell`/`node` or a `UgridDataArray` unstructured, redirecting to `.ugrid` where it can. +2. **Accessor leftovers (#208):** + - 1-based selection by MF6 cell numbers (cellids from list packages), since dims carry no labels. Not as `ds.grid.sel(layer=1, ...)`: `grid.sel` is now `sel` plus `z`, 0-based like `sel`. A separate method, e.g. `ds.grid.sel_cell(layer=1, row=5, col=10)` or `ds.grid.sel_cell(cellid)`. + - `index_xy` for arrays reopened from netCDF (e.g. `ds.grid.index()`, or doing it on access). + - A world-box selection that masks on `x_world`/`y_world` for rotated grids. + - Whether to expose the old `method="contains" | "nearest_center"` names, or just document `sel`'s `None`/`"nearest"`. + - Possibly `ds.grid.crs` (from `projection.crs_wkt`), `plot`, `plot_profile`, `neighbors`. +3. **netCDF bounds:** write `x_bnds`/`y_bnds` from `CellIndex` and `z_bnds` from `z_top`/`z_bot` (with `bounds` attrs), and read them back, so MF6 netCDF can be indexed without a grb. This is the writer and the heads reader. +4. **Ball/range queries:** `query_ball` via `scipy.spatial.cKDTree` on centres (3-D, using `z` now that layered arrays carry it) and geopandas (2-D), as free functions with accessor methods. On rotated grids they read `x_world`/`y_world`. +5. **Budget-reader coordinates for faces.** + - Now: flows indexed by the cell they leave (DIS right/front/lower, DISV lower) carry that cell's x/y/z coordinates, as in DIS since #469; DISV horizontal flows, on mesh edges, carry none. + - DISV edges: x/y from UGRID edge coordinates (xugrid's edge midpoints), and a vertical extent from the two cells on either side, `z_top = min` of their tops and `z_bot = max` of their bottoms (one cell on the grid boundary), from `edge_face_connectivity`. Then `mask_z`/`sel_z` work on edges. + - Keep the cell-based convention for cell-indexed flows, so budgets line up with heads cell for cell, and document it: a lower flow crosses the face at the cell's `z_bot`, so `sel_z(z)` on lower flows gives the lower-face flow of the cell containing z, not the flow through a face at z. + - After DISV selection, since both use xugrid's edge connectivity. +6. **Selection vignette.** Drafted on `z-attach` (`df8e4d93`) as `docs/examples/selection.py` + `.ipynb`, covering the points below; review it with that PR. Original outline: `docs/guide/selection.py` (jupytext, paired `.ipynb`, under the Guide part of `_toc.yml`), on a small rotated DIS model with heads: + 1. Scalar `sel`/`sel_world`: containment, ties to the lower-numbered cell, outside raises `KeyError`. + 2. Lists vs DataArrays on a shared `point` dim: the outer product vs one cell per point, side by side, with result shapes. Lists silently give the outer product, which is the subtle part. + 3. With `z`: a vertical column per point, then a cell per point; a scalar `z` with point x/y. + 4. `sel_z` as a horizontal slice, and `mask_z` with `where`. + 5. Rotated grids: why world slices don't work, and masking on `x_world`/`y_world` instead. +7. **The `projection` attrs limitation** (see Known limitations). +8. **Quasi-3D `laycbd`** z centres, only if non-MF6 grids are in scope. +9. **Upstream housekeeping** (don't post without asking): + - Rescope or close #207: layer/row/col and z are done; DISV is what's left. + - Close draft #278 (`xarray-index`, the old spatial index design), which this plan replaces. + - Point #208 at this plan. + - Supersede or delete `docs/dev/spatial-index-design.md` on `xarray-index`, or replace it with a version of this plan under `docs/dev/`. + +### Parallel work + +- `z-attach` touches `grid.py`, `package.py`, `adapters.py`, the heads and budget readers, and the accessor. Until it merges, avoid starting work in those files. +- **Can start now:** the DISV/xugrid groundwork (env and tests only), `query_ball` as free functions in their own module, and drafting the upstream housekeeping text. +- **After `z-attach` merges:** the accessor leftovers, netCDF bounds, and the `projection` fix. The DISV input/dispatch decision can then go in alongside the accessor leftovers. diff --git a/tas-interpolation-plan.md b/tas-interpolation-plan.md new file mode 100644 index 00000000..a4d34b6d --- /dev/null +++ b/tas-interpolation-plan.md @@ -0,0 +1,188 @@ +# Evaluate time-array series in the period view + +Status: ready, not started (rechecked 2026-10-19 against develop `e0d94b86`: +no `Tas.average`, `step_array` or `dense_steps` yet; `period_array`, +`period_arrays.dense` and `_check_tas_refs` are where the plan expects). +Since written, #489 marks series values `NaN` in the xarray view with the +series name in a `_tas` coordinate; a step view could fill those. +Written 2026-10-07, against develop `a6c52c8`. MF6 semantics checked +in `../modflow6` (`src/Utilities/TimeSeries/TimeArraySeries.f90`, +`TimeArraySeriesManager.f90`). Decisions are under "Decisions". + +## Problem + +A period array (RCHA's `recharge`, EVTA's, SPCA's) can come from a +time-array series: `recharge={0: TimeArraySeriesRef("rch")}`. +`Package.period_array()` and `period_arrays.dense()` raise on these +because they can't compute the values. So anything that wants the values +in effect can't use a model that has them: plots, adapters, and analysis. + +## What's available + +- The series is already loaded with the reference. RCHA, EVTA and SPCA + have `tas: list[Tas]` children (`TAS6 FILEIN`), and each `Tas` holds + `time_series_name`, `interpolation_method`, `sfac` and + `tas_array: {time: array}`. +- `Package._check_tas_refs` already matches references to series by name. +- `Time` (`flopy4/mf6/utils/time.py`, a flopy `ModelTime`) gives each + step's end time as `totim`; a step starts at the previous one's end, or + at 0. `Simulation.time` builds it from TDIS. + +## MF6 semantics (checked) + +- **Only STEPWISE and LINEAR.** `tas_init` rejects anything else + ("Unknown interpolation method"). The DFN (`utl-tas.dfn`) lists + `linearend` as valid, but MF6 doesn't accept it for array series. +- **Evaluated once per time step, as an average.** `tasmgr_ad` calls + `GetAverageValues(begintime=totimc, endtime=totimc+delt)`. That is the + integral of the interpolant over the step, divided by `delt`. A + zero-length step takes the value at that time. + - **STEPWISE:** the integrand holds each array until the next time. If + a series time falls inside a step, the step gets the time-weighted + mix of both arrays, not the value at the step's start. + - **LINEAR:** the integrand interpolates linearly between times, so the + step gets the trapezoid average. +- **SFAC is applied when the array is read**, so it scales every value. +- **Series times are simulation time** (`totim`), starting at 0. +- **Coverage:** + - If the series ends before a step ends, MF6 stops with "Error + encountered while performing integration". This is true for STEPWISE + too. + - If the step starts before the series' first time, + `get_latest_preceding_node` finds no record, and the integral stays + zero with no error. The step gets zeros. + +## Design + +1. **`TimeArraySeriesRef.resolve(package) -> Tas`** finds the series by + name (case-insensitive) among `package.tas`, and raises as + `_check_tas_refs` does if none matches. The reference stays a name; + resolving it when used keeps it valid if `tas` is edited. +2. **`Tas.average(t0, t1) -> np.ndarray`** gives MF6's step value: the + integral of the interpolant over `[t0, t1]` divided by `t1 - t0`, or + the value at `t0` when they're equal. SFAC is applied. + - It lives in a method-only mixin (`TasMethods`, listed in `MIXINS`), + with the integration as plain functions so they can be tested on + their own. + - It raises if the series ends before `t1`, as MF6 does. + - It returns zeros with a warning if `t0` is before the first time. + MF6 does this silently; the warning flags what's likely an input + mistake. + - It raises on a method other than STEPWISE or LINEAR, naming the + method. +3. **A step view, as its own method.** `Package.step_array(name, + time=None)` returns `(nstp_total, ...)`: one array per time step, the + values MF6 applies. + - Plain period arrays repeat their period's value across its steps, + with carry-forward as today. + - A period given by a reference is evaluated per step with + `Tas.average`. A reference carried forward keeps being evaluated at + each later step. + - Aux arrays work the same way, per name: `{name: (nstp_total, ...)}`, + as `period_array` gives `{name: (nper, ...)}`. + - Values stay in input units, as in `period_array`. MF6 also converts + RCH/EVT fluxes to flows (cell area) and applies `auxmultname` after + substitution. The view does neither. +4. **The period view is unchanged.** `period_array` keeps its signature + and its `(nper, ...)` shape. A period from a time-array series has no + single value in MF6, so it still raises for one. The error now points + to `step_array`. No averaged or derived value is invented. +5. **Where the time steps come from.** `step_array` takes an optional + `time: Time`. + - Without it, the package finds TDIS through its parents (the + simulation's `tdis`) if it's attached to one. + - Pass `time` for a package that isn't attached. + - Raise if both are present, so it's never ambiguous which time + discretization applies. + - Raise if neither is present. +6. **`period_arrays.dense_steps()`**, next to `dense()`, takes the step + bounds and a resolver (a callable from a reference to a `Tas`). + `dense()` keeps raising on references. +7. **NetCDF export is a non-goal.** Exported files keep the reference in + the package file, since MF6 reads the series itself. + +## Steps + +1. Add `Tas.average` and its integration functions. Unit tests against + hand-worked values: + - STEPWISE and LINEAR over a step that lies within one interval; + - a step spanning a series time (the mix for STEPWISE, piecewise + trapezoids for LINEAR); + - a zero-length step; + - SFAC; + - the series ending early (raises), and starting late (zeros, warns); + - an unsupported method (raises). +2. Add `TimeArraySeriesRef.resolve`, and have `_check_tas_refs` use it. +3. Add the time lookup (`time=` or TDIS through the parents, raising on + both or on neither). +4. Add `dense_steps` and `Package.step_array`, aux included. Update the + period view's error message. +5. **Test against MF6:** + - Run test027 (`TimeseriesTest`), read the RCH budget per step from + the CBC file, and compare it with `step_array("recharge")` times + cell area. + - Add a small synthetic model per method with a series time inside a + multi-step period, so the STEPWISE mix is exercised. + - Cover an aux array from a series, and SPCA's `concentration`, in + the unit tests. MF6's budget doesn't show aux, so these aren't + compared against a run. +6. **Upstream:** report that `utl-tas.dfn` lists `linearend`, which MF6 + rejects for array series. Either the DFN or the Fortran should change. + +## Decisions + +- **Time source:** the package's TDIS if it's attached to a simulation, + or an explicit `time`. Both present raises. +- **No period average.** The step view gives MF6's values. The period + view gives exact values or raises; it never returns a derived summary. +- **Aux and SPCA are in scope.** MF6 accepts a series for RCHA + `recharge`/`aux`, EVTA `rate`/`aux` and SPCA + `concentration`/`temperature` (the DFNs' `time_series true`; aux + through `aux_tasnames` in `Mf6FileLayerArray.f90`). flopy4 already + types all of them `... | TimeArraySeriesRef`, and `split_tas` already + handles aux. EVTA `surface`/`depth` can't come from a series. +- **A separate `step_array`, not a `by=` keyword.** A keyword would make + `period_array`'s shape depend on an argument, and `time=` would apply + to only one of its modes. +- **Warn on zeros before the series starts.** MF6 is silent; the warning + is flopy4's addition, since it most likely marks an input mistake. + +## Follow-on: time series in period lists + +Not started (added 2026-10-18). The same gap exists for TS6 in list-based +packages. A row value can come from a series, `TimeSeriesRef("headseries")`, +and `to_xarray` holds it as NaN, with the series name in the `_ts` +coordinate. So the values in effect aren't available for these either. This +follows the TAS work and reuses its time lookup (design 5) and step +bounds. + +MF6 semantics (checked in `TimeSeries.f90` and `TimeSeriesManager.f90`): + +- **STEPWISE, LINEAR and LINEAREND.** Unlike TAS, LINEAREND is accepted. + STEPWISE and LINEAR give the step average, as with TAS. LINEAREND gives + the value at the step's end (`get_value_at_time(time1)`). +- **Per-series method and SFAC.** `METHODS`/`SFACS` set them per name, and + `METHOD`/`SFAC` set them for all names. SFAC is applied when each record + is read. +- **DNODATA in a TS file is a gap.** That record is skipped for that series + only, so the series interpolates across it. +- **Names are case-insensitive** (`UPCASE` on lookup). +- **No extension past the end.** `extendTsToEndOfSimulation` defaults to + false, and no package sets it. A series that ends before a step ends is + an error, as with TAS. +- **Aux values can come from a series too** (`auxvarTsLinks`), so the + evaluation covers aux columns. + +Sketch: + +1. `TimeSeriesRef.resolve(package) -> (Ts, index)` finds the series among + `package.ts` by name. It shares the name lookup with the TS name check. +2. `Ts.average(name, t0, t1) -> float` gives MF6's step value for one + series, reusing the TAS integration functions on scalars, with + LINEAREND added. +3. A step view of a period list, with one value per step for each row + given by a series. Rows with plain values repeat their period's value, + as `step_array` does. Whether this is a dense per-step dataset or rows + per step is open. It should match whatever `step_array` settles on. +4. The period view is unchanged, as for TAS (design 4). +5. Test against test027 and the test001e TS models' CBC budgets per step. diff --git a/xarray-export-plan.md b/xarray-export-plan.md index 030ba7bb..aa439d1c 100644 --- a/xarray-export-plan.md +++ b/xarray-export-plan.md @@ -1,17 +1,22 @@ # xarray export plan -Follows up dead-code audit item 5 (`dead-code-audit.md`). Status: design -**decided** (2026-11-02), implementation not started. Nothing in `mixins.py`, -`protocols.py` or the reverse half of `attrs_xarray.py` gets deleted, and -no tests get dropped, except as the phases below say. - -Scratch scripts used for the comparison (rerun them to see current -behavior): -`/private/tmp/claude-502/-Users-wbonelli-dev-pyphoenix-project/3ec4ddc6-e42e-49eb-bc12-ce39a377ee2c/scratchpad/compare.py` and -`/private/tmp/claude-502/-Users-wbonelli-dev-pyphoenix-project/3ec4ddc6-e42e-49eb-bc12-ce39a377ee2c/scratchpad/compare2.py`. -They build a 1x4x5 DIS GWF model (NPF, IC, CHD, WEL, RCHA, OC, IMS, TDIS) -and print `pkg.to_xarray()` next to `attrs_to_dataset(pkg)`, plus both -round trips. +Status: design **decided** (2026-11-02), implementation not started. The +spec it builds to is `docs/dev/xarray-representation.md`; the direction +and the higher-level open questions are in +`xarray-package-plan.md`. + +This started from a dead-code audit question: is the test-only half of +the attrs/xarray layer (`mixins.py`, `protocols.py`, `dataset_to_attrs`, +`datatree_to_attrs`) the planned generic conversion layer, or leftover +from before xattree was dropped (#356)? The answer is to keep it and +finish it. Nothing in `mixins.py`, `protocols.py` or the reverse half of +`attrs_xarray.py` gets deleted, and no tests get dropped, except as the +phases below say. + +The comparison behind "Where things stand" used a 1x4x5 DIS GWF model +(NPF, IC, CHD, WEL, RCHA, OC, IMS, TDIS), printing `pkg.to_xarray()` next +to `attrs_to_dataset(pkg)`, plus both round trips. Phase 1's tests +replace it. ## Decisions @@ -80,7 +85,7 @@ sparse forms. - **No coordinates**, and arrays sit on flat DFN dims. - **Tdis appears twice** in `sim.to_xarray()`: as root vars, and as `/tdis`. -### Already fixed (uncommitted on `develop`) +### Already fixed (branch `to-xarray-constants`) `Package.to_dataarray` (and so the view) raised on structured DIS (reshaping `delr`/`delc` onto the grid) and on any scalar griddata field @@ -90,11 +95,12 @@ shape (a 0-d array if the shape is unknown). Arrays not on the cells get one dim per shape name, via `_AXIS_DIMS` (`ncol` → `col`, `nrow` → `row`). Tests: `test_to_xarray_constants_attached_later` and `test_dis_to_xarray_axis_arrays` in `test/mf6/test_mf6_component.py`. -Should go in as its own small PR before phase 1. +Committed on local branch `to-xarray-constants` (`07679ca0`, off +develop `000f63c3`). Should go in as its own small PR before phase 1. ## Phases -Each phase is roughly one PR. Per `CLAUDE.local.md`, verify with +Each phase is roughly one PR. Verify with `pixi run -e dev smoke` plus the test files touched; CI runs the rest. ### Phase 1: make the generic layer complete (`flopy4/attrs_xarray.py`) @@ -319,7 +325,7 @@ References: `test/test_attrs_xarray.py` off the mixins onto the plain functions. - Update `docs/dev/sdd.md` ("Xarray conversion") to describe the complete export, the accessor view, and how both differ from MF6 NetCDF input. -- Close audit item 5 in `dead-code-audit.md`, pointing here. +- Close the dead-code audit question (above): the layer was kept and finished. ## Still open (decide during implementation) @@ -327,9 +333,8 @@ References: - Type tag values: the DFN component name (`gwf-npf`) or the class path. - Whether the old instance `from_xarray` survives in any form (phase 2). - Whether `dense()` lives on DataArray too, or only on Dataset. -- Names: `flatten()`/`unflatten()`, or something else. Also whether the - flat dim is `node` (DISU's, so a flat DIS looks like DISU) or `nodes` - (the DFN fallback). Pick one and make the fallback use it too. +- Names (`flatten()`/`unflatten()`, `node` vs `nodes`) are decided with + the accessor name; see `xarray-package-plan.md` "Open". - Whether `.grid.sel(layer=, row=, col=)` and `.grid.sel(x=, y=)` should work on the flat view (map to `node` positions), or require `unflatten()` first. diff --git a/xarray-package-plan.md b/xarray-package-plan.md index 3e512bd8..728d8e15 100644 --- a/xarray-package-plan.md +++ b/xarray-package-plan.md @@ -29,7 +29,8 @@ package later is a packaging decision rather than a redesign. MF6 NetCDF format and the representation relate. 3. **The boundary is a data contract, not an API.** The export schema from `xarray-export-plan.md` phase 1 is the interface. flopy4 produces it; - the package (the accessor, `dense()`, the encoders) consumes it, using + the package (the accessor, `dense()`, the writers and adapters) + consumes it, using only what's in the dataset. The export plan already states the rule: *"If it needs the package or the grid object, phase 1 is missing something."* @@ -44,23 +45,31 @@ package later is a packaging decision rather than a redesign. ## How this relates to the other xarray docs -This is the umbrella doc. The others keep their scope; this one says which -layer each belongs to and what they have to agree on. - -| Doc | Where | Scope | Layer | Status / what this plan changes | -|---|---|---|---|---| -| `xarray-export-plan.md` | branch `xarray-export-plan` (`c5e3f488`), not on develop | lossless `to_xarray`/`from_xarray`, `grid.dense()` accessor | flopy (export) + package (accessor) | Decided, not started. Phase 1 now also writes the representation spec (below). Its open questions (bool encoding, type tags) are spec questions. | -| `rotated-grid-index-plan.md` | root | DIS coords, rotation, `CellIndex`, z selection, `.grid` accessor | package (grid) | Mostly done (#469–#490). Its "Next 1" DISV input decision should follow this plan's xugrid stance. | -| `combined-index-notes.md` | root | a combined `GridIndex` on `CellIndex` | package (grid) | Deferred. Revisit if DISV needs its own index anyway (it may; see xugrid). | -| `period-lists-plan.md` | root | period lists in xarray, dense/sparse | flopy (encode) + package (densify) | Done (#476). The dense view becomes `dense()` per the export plan. | -| `tas-interpolation-plan.md`, `datetime-periods-plan.md` | root | time-array series in the view; datetime period keys | package (time) | Ready / in progress. The time coordinate they produce is part of the spec. | -| `docs/dev/spatial-index-design.md` | `origin/xarray-index` (#278) | original spatial index design | — | Superseded by `rotated-grid-index-plan.md`; housekeeping already planned there and in `dev-docs-plan.md` page 3. | -| `docs/dev/sdd.md` "Xarray conversion", "NetCDF" | develop | SDD description | both | Update when export plan phase 3 lands (below). | -| `archive/xarray-xugrid-refactor.txt`, `archive/mf6-ugrid-planning.txt` | archive | May 2026 chats behind the `xugrid` branch | — | History. The `xugrid` branch they produced conflicts with decision 2; see "xugrid". | -| `dead-code-audit.md` item 5 | root | test-only attrs/xarray layer | flopy (export) | Answered by the export plan (keep it, finish it). | -| `dev-docs-plan.md` page 3 | root | `docs/dev/grid-coordinates.md` | package (grid) | Unchanged. That page later becomes part of the package's docs. | -| [xattree #36][x36] | GitHub | functional rewrite of xattree | generic (A1) | Open. flopy4's `attrs_xarray.py` is the working prototype; see "xattree". | -| `pydantic-object-model-plan.md` | root | attrs vs pydantic dataclasses (#367) | flopy (object model) | Stale. Its outcome decides whether xattree v2 is attrs-only. | +This is the umbrella doc for direction: decisions, the boundary, when to +extract, coordination, and all open questions. The other docs keep their +scope: + +- **`docs/dev/xarray-representation.md`** is the spec: what the + representation is, how it relates to netCDF, CF, UGRID and MF6 NetCDF, + and the notes on the MF6 NetCDF format. +- **`xarray-export-plan.md`** is the implementation checklist for the + export, the accessor and the cell layouts. +- The plans below are where specific pieces were designed. The spec now + summarizes what they decided. + +| Doc | Scope | Layer | Status | +|---|---|---|---| +| `xarray-export-plan.md` | lossless `to_xarray`/`from_xarray`, `dense()`, `flatten()` | flopy (export) + package (accessor) | Decided, not started. Phase 1 fills in the spec. Needs branch `to-xarray-constants` merged first. | +| `rotated-grid-index-plan.md` | DIS coords, rotation, `CellIndex`, z selection, `.grid` accessor | package (grid) | Mostly done (#469–#490). Its Next list is still live; Next 1 (DISV) should follow the xugrid stance below. | +| `rch-evt-layer-plan.md` | z coordinates for RCHA/EVTA arrays | package (grid) | Not started. | +| `combined-index-notes.md` | a combined `GridIndex` on `CellIndex` | package (grid) | Deferred. Revisit if DISV needs its own index anyway (see xugrid). | +| `datetime-periods-plan.md` | datetime period keys, `create_tdis` | flopy + package (time) | In progress (#518, #226). | +| `tas-interpolation-plan.md` | time-array series in the dense view | package (time) | Ready, not started. | +| period lists (#476) | period lists in xarray, dense/sparse | flopy (encode) + package (densify) | Done. The spec's "Period data" records the result; the dense view becomes `dense()`. | +| `docs/dev/spatial-index-design.md` on `origin/xarray-index` (#278) | original spatial index design | — | Superseded by `rotated-grid-index-plan.md`, whose Next 9 covers the housekeeping. | +| `docs/dev/sdd.md` "Xarray conversion", "NetCDF" | SDD description | both | Update after export phase 1 (representation) and once the writers and adapters exist (NetCDF); see step 7. | +| [xattree #36][x36] | functional rewrite of xattree | generic (A1) | Open. flopy4's `attrs_xarray.py` is the working prototype; see "xattree". | +| #367 | attrs vs pydantic dataclasses | flopy (object model) | Stale. Its outcome decides whether xattree v2 is attrs-only. | ## The layers @@ -98,135 +107,38 @@ at all. ## The representation -The contract from decision 3. Write it down as part of export plan phase 1 -in `docs/dev/xarray-representation.md` (later the package's spec). Its first -section, on how the representation relates to netCDF, CF, UGRID and the MF6 -NetCDF format, is drafted. Most of the rest is already decided in other -docs, so the spec mainly collects it: - -- **Grid dims and coordinates per grid type:** - - DIS `layer,row,col` with 1-D indexed `x`/`y` - - DISV `layer,cell` - - DISU `node` - - DIS2D/DISV2D/DISV1D: TBD - - `z`/`z_top`/`z_bot`, `dx`/`dy`, `projection`, `x_world`/`y_world` - - Source: `rotated-grid-index-plan.md` "Coordinates on a DIS array", plus a DISV/DISU table to write. -- **Indexing base:** 0-based everywhere (`isel` and dims); 1-based only in - encodings and in an explicit `sel_cell`. -- **Grid metadata in attrs:** grid type and sizes (`nlay`, `nrow`, `ncol`, - `ncpl`, `nodes`), `nper`, so consumers don't need the DIS object - (export plan phase 1). -- **Period data as given:** period arrays along `_period`, list data - as a sparse table (entry dim, `per` and cellid coordinates, a variable per - column, `_ts`/`_tas` coordinates). Dense is a view (`dense()`), - and so is the flat `node` layout (`flatten()`/`unflatten()`, export plan - phase 2b). Neither is stored. -- **Time:** how `per` maps to a time coordinate (from TDIS, with a start date - where given); ties in with `datetime-periods-plan.md`. -- **Tree:** nodes named by component name, with a type tag (export plan open - question: DFN name `gwf-npf` vs class path; prefer the DFN name, since it - doesn't depend on Python). -- **netCDF-safe attrs:** paths as str, bools encoded (open), None and empty - child lists omitted. -- **Fill values:** one convention per dtype (see the MF6 NetCDF notes below). +The contract from decision 3, specified in +`docs/dev/xarray-representation.md`: grid dims and coordinates, the tree +and type tags, period data as given (dense and flat are views), time, and +attributes and fill values. Export plan phase 1 makes `to_xarray` produce +it and settles the spec's open points as it goes. ## Files and adapters -None of these changes the representation. Each is a pair of functions -(representation ↔ file or object) plus accessor methods. - -**Files: MF6 NetCDF, two layouts of one spec.** CF, UGRID and MF6 NetCDF -stack: CF is the base, UGRID extends it, and MF6 NetCDF adds attributes -that other tools ignore. So the ungrouped layout can be valid CF, UGRID and -MF6 NetCDF at once, and there's no separate "CF" or "UGRID" file format. - -- **Grouped**: the representation as is, via `DataTree.to_netcdf`/`to_zarr`, - with groups mirroring the tree and only the encoding netCDF needs (bools, - None, fill values, 1-based). It covers everything (options, dims, list - data, arrays). It's valid CF 1.8+ with groups. MF6 doesn't read it yet. - Today it's for Python round trips, caching, archiving and sharing. -- **Ungrouped (structured and layered mesh)**: what `netcdf.py` writes - today, restated as a flattening of the grouped layout. Package paths - become names plus `modflow_input`, indices go to 1-based, layers become a - `layer` dim (split variables only as a writer option, if MF6 or a tool - needs them), and then CF is written. Readers reverse this. While - coverage is partial (one model's GRIDDATA/PERIOD arrays), they merge the - result into a tree whose other fields come from text input. It's meant to - cover the whole simulation eventually (to confirm). CF/UGRID validity is - a goal: bounds (`x_bnds`/`y_bnds`/`z_bnds`, `rotated-grid-index-plan.md` - Next 3), `grid_mapping`, standard names, and UGRID mesh topology on the - mesh form. - -**Adapters: other libraries' objects, in memory.** Each library then writes -its own files (UGRID netCDF through xugrid, GeoTIFF through -`rio.to_raster`). Writing an adapter's output with plain `to_netcdf` should -give something close to the ungrouped layout, not a third format. - -- **`to_ugrid()` / `from_ugrid()`** → `UgridDataset` for DISV and layered - DIS. This is where `UgridDataset` appears, not in `to_xarray`. -- **`to_rioxarray()`** → a DataArray with CRS and transform written, for - structured grids. Rotation needs checking: rioxarray takes a rotated - affine transform, but its 1-D x/y coordinates assume no rotation. -- **`to_cf()`** → a Dataset with CF attributes, bounds and `grid_mapping`, - for cf_xarray and other CF-aware tools. It's mostly the same metadata - the ungrouped writer uses. -- **`to_modelgrid()`** → a flopy3 grid, while flopy3 still needs one - (replaces the inheritance in `utils/grid.py`, step 4). - -The names are placeholders. - -**Binary output** (GRB/HDS/CBC): readers that produce the representation -directly. - -### Notes on the MODFLOW NetCDF format - -The format is ours to change too, so the representation work is a chance to -feed fixes back to MF6. The ungrouped layout has been developed in steps and -is meant to cover the whole simulation eventually (to confirm with the MF6 -developers). So the goal is one spec: the grouped layout is the -representation encoded, and the ungrouped layout is a fixed flattening of -it that stays valid CF and UGRID. Flat is right for the layout MF6 reads -today: MF6 loads one model at a time, UGRID defines one mesh per dataset, -and GIS tools handle groups poorly. Groups only matter if MF6 is to read -the grouped layout. - -The hard part of flattening a whole simulation is dimension names, not -variable names. Two GWF models each have their own `nlay`/`ncpl`; a GWT -model shares its GWF model's grid; exchanges span models; list data needs -per-package entry dims. Without groups every dim needs a prefix or a rule -for sharing. Settle that rule first. A shared-grid rule keeps the file -usable in GIS tools. - -Candidates, from reading the wiki (2026-11-03). They need checking against -the MF6 source before anything is proposed upstream: - -- **Per-layer variable splitting** in the layered mesh (`npf_k_l1`, - `npf_k_l2`, `z_l`, each with a `layer` attr): a variable's identity is - spread over names and attrs. xugrid and most CF tools would rather see a - `layer` dim. If that's needed for QGIS/GDAL, make it a writer option and - say so in the spec. -- **No DISU**, and no DIS2D/DISV2D/DISV1D. The representation has to cover - them regardless, so it shouldn't wait on the format. -- **Inconsistent fill values**: `9.97e36` (GRIDDATA), `3e30` (PERIOD input - example), `1e30` (output), `-2147483647` (int). -- **`modflow_input` is a path string** (`//`) plus - `modflow_iaux` and `layer`. That's workable, but the decoder has to parse - strings to find the package, and the package name depends on the namefile - (or FloPy's generated name). -- **Input `time` is NPER-sized and output `time` is step-sized**, with no - stated way to relate the two. The representation needs both (`per` and step - time), so the format could carry the mapping. -- **List (stress period) data isn't covered**; only GRIDDATA/PERIOD arrays - in READASARRAYS/READARRAYGRID packages. Nor are options, scalars or the - simulation-level components. All in scope if the layout is to cover the - whole simulation. -- **Rotation is only encoded through a wrapped WKT2 `PROJCRS`**; without a - CRS, rotated structured grids lose `axis`/`standard_name`. The - representation's `projection` coordinate carries xorigin/yorigin/angrot - either way (`rotated-grid-index-plan.md`). The format could do the same. -- **"No naming rules beyond CF"**: fine, but then the type tag / - `modflow_input` is the only reliable identity, which argues for making it - structured. +Each is a pair of functions (representation ↔ file or object) plus +accessor methods. None changes the representation. The spec's first +section has the full picture; in short: + +- **MF6 NetCDF (grouped):** the representation as is, through + `DataTree.to_netcdf`/`to_zarr`, with only the encoding netCDF needs. + Lossless. MF6 doesn't read it yet. +- **MF6 NetCDF (ungrouped), structured and layered mesh:** what + `netcdf.py` writes today, restated as a fixed flattening of the grouped + layout, aiming to be valid CF and UGRID. Readers reverse it and, while + coverage is partial, merge in the fields that come from text input. CF + bounds come with it (`rotated-grid-index-plan.md` Next 3). +- **Adapters:** `to_ugrid()`/`from_ugrid()` (xugrid, DISV and layered + DIS), `to_rioxarray()` (structured grids; rioxarray takes a rotated + affine transform but its 1-D x/y assume no rotation, so check this), + `to_cf()` (cf_xarray), `to_modelgrid()` (flopy3, replacing the + inheritance in `utils/grid.py`, step 4). Names are placeholders. +- **Binary output** (GRB/HDS/CBC): readers that produce the + representation directly. + +The format is ours to change too. The spec's "What the format could +change" lists candidates. The first thing to settle is the flattening +rule for a whole simulation, chiefly how dims are prefixed or shared +across models. ## xugrid @@ -248,7 +160,7 @@ the MF6 source before anything is proposed upstream: If it lands, DISV could be "plain dataset plus xugrid's index", and the wrapper concern goes away. - **The pin** (`xugrid==0.15.3` in `pyproject.toml`) needs a range anyway - (`flopy-integration-plan.md` track B). Making it optional is a bigger + before flopy4 can go into flopy. Making it optional is a bigger step; do it once grid.py no longer imports it at module level. ## xattree @@ -323,9 +235,9 @@ provider/resolver protocols. Check this before extracting. Roughly in order. Each is small and fits the existing plans. None needs a new repo. -1. **Write the representation spec** (`docs/dev/xarray-representation.md`, - explainer section drafted) as part of export plan phase 1. It settles - that plan's open questions (bool encoding, type tags). +1. **Finish the representation spec** (`docs/dev/xarray-representation.md`, + drafted) as part of export plan phase 1, settling its open points (bool + encoding, type tags, the float fill value). 2. **Rename the accessor** from `.grid` to something distinctive (e.g. `.mf6`) before more users depend on it. `.grid` is generic enough to collide (xugrid uses `.ugrid`, uxarray `.uxgrid`). Do it with export plan @@ -337,8 +249,8 @@ new repo. and `flopy4.spec`, so the boundary holds while it's one repo. 4. **Drop the flopy3 grid inheritance** (`utils/grid.py`), and replace `utils/time.py`'s use of flopy3's `ModelTime`. This is the biggest piece. - It's also what `flopy-integration-plan.md` needs, since flopy4 moving - into flopy shouldn't depend on flopy3's discretization classes. + It's also needed for moving flopy4 into flopy, which shouldn't depend + on flopy3's discretization classes. 5. **Split `netcdf.py` and `period_lists.py`** along the encode/densify line above. 6. **Lazy xugrid** everywhere in the subpackage, then decide whether it's @@ -364,7 +276,7 @@ All of: - the import-linter contract is green and the subpackage imports neither flopy3 nor flopy4's object model, - someone outside flopy wants it: an output-only user, imod, xmipy, a GUI, -- flopy4 is in flopy (`flopy-integration-plan.md`), so the third repo isn't +- flopy4 is in flopy, so the third repo isn't on top of the devtools release coordination already blocking that. Until then, a subpackage with an enforced boundary gets most of the benefit. @@ -382,7 +294,7 @@ timing. encoding and output layer could be shared, or at least whether we can agree on the representation so datasets move between the tools. Don't contact anyone without asking. -- **MF6 NetCDF format**: the notes above, as an issue or discussion on +- **MF6 NetCDF format**: the spec's format notes, as an issue or discussion on modflow6, once checked against the source. First confirm the intent: is the ungrouped layout meant to cover the whole simulation, and would MF6 ever read a grouped one? Don't post without asking. @@ -404,7 +316,10 @@ timing. `to_cf`, `to_modelgrid`). The format names are settled: MF6 NetCDF (grouped) and MF6 NetCDF (ungrouped). - Package name, and whether the subpackage is `flopy4/xr/` or `flopy4/mf6/xr/`. -- Accessor name. +- Accessor name, and with it the names in export plan phase 2b, decided + together: `flatten()`/`unflatten()` or something else, and whether the + flat dim is `node` (DISU's) or `nodes` (the DFN fallback, which should + then use the same name). - Whether the time code (TDIS → time coordinate, TAS evaluation) is part of the package or stays in flopy. Leaning package, since output readers need it too. @@ -412,8 +327,8 @@ timing. Prefer not: the writers take the metadata from attrs that flopy puts on the representation, and devtools stays a flopy dependency only. - Representation for DIS2D/DISV2D/DISV1D (OLF/CHF). -- Merge `xarray-export-plan.md` onto develop (or into the working tree) so - the two plans sit side by side. +- When to get this branch's docs onto develop: now, as plans, or with + export phase 1. - attrs vs pydantic (#367): whether xattree v2 is attrs-only or works with any dataclass-like. - Whether xattree v2 keeps the field factories (`dim`, `coord`, `array`) From 256d8c2440cb8b0e53b0390606c619ecc2d48193 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 11:30:12 -0800 Subject: [PATCH 06/77] add road-to-MMP plan and flopy integration plan The MMP plan sequences what's left before MMP into parallel tracks with a gate, and records the release shape: alphas and betas from this repo via GitHub, release candidates from flopy's 4.x branch, GA as the last candidate. map.md gains the release-candidate stage. The integration plan is refreshed against develop 8d25e4b7 and flopy aca52382, and its phases and checkpoints follow the same release shape. Co-Authored-By: Claude Opus 5.5 --- docs/_toc.yml | 1 + docs/dev/map.md | 8 +- docs/dev/mmp-plan.md | 151 +++++++++++ flopy-integration-plan.md | 521 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 680 insertions(+), 1 deletion(-) create mode 100644 docs/dev/mmp-plan.md create mode 100644 flopy-integration-plan.md diff --git a/docs/_toc.yml b/docs/_toc.yml index 66d493ec..f51723a7 100644 --- a/docs/_toc.yml +++ b/docs/_toc.yml @@ -29,6 +29,7 @@ parts: - file: dev/srs - file: dev/sdd - file: dev/map + - file: dev/mmp-plan - file: dev/protocols-plan - file: dev/array-converter-design - file: dev/test diff --git a/docs/dev/map.md b/docs/dev/map.md index 64fb567a..87baae2e 100644 --- a/docs/dev/map.md +++ b/docs/dev/map.md @@ -35,8 +35,14 @@ All core and most peripheral functionality, and may involve: Release to wider test audience at USGS and Deltares via `pip install` from github URL. Begin beta versioning. +What's left to reach the MMP, and in what order, is in [mmp-plan.md](mmp-plan.md). + +### Release candidates + +Integrated into the existing repository on a `4.x` branch, cut once MMP feedback has settled the API. Released as `flopy==4.0.0rcN` via PyPI and conda-forge. See `flopy-integration-plan.md`. + ### GA -Feature-parity with 3.x. Integrated into existing repository. Generally available via standard channels (PyPI, Conda). +Feature-parity with 3.x. The last release candidate. Generally available via standard channels (PyPI, Conda). 3.x enters maintenance-only mode for a limited time after which support will be dropped and all effort moved to 4.x. diff --git a/docs/dev/mmp-plan.md b/docs/dev/mmp-plan.md new file mode 100644 index 00000000..804a9485 --- /dev/null +++ b/docs/dev/mmp-plan.md @@ -0,0 +1,151 @@ +# Road to MMP + +What's left between today and the MMP, in what order, and what comes +after it. Written 2026-11-03 against `develop` (`8d25e4b7`). This doc +sequences the work and links to the plans that hold the detail, so it +stays short and the plans stay the source of truth. Update the status +column as things land. + +## Release shape + +Decided 2026-11-03. + +| stage | released from | installed by | gate | +|---|---|---|---| +| alphas | this repo | `pip install` from GitHub | as needed | +| betas, MMP the last of them | this repo | `pip install` from GitHub | "MMP gate" below | +| release candidates `flopy==4.0.0rcN` | flopy's `4.x` branch | PyPI, conda-forge | the integration plan's "Before the first release candidate" | +| GA | flopy, the last release candidate | PyPI, conda-forge | the integration plan's "Before GA" | + +Why this order, and not integrating into flopy before MMP: + +- **GitHub installs don't need the packaging work.** PyPI rejects git + dependencies, `pip install git+...` doesn't. The devtools release, the + xugrid range and Python 3.14 gate the release candidates, not MMP. +- **Testers can run both engines side by side.** `flopy4` and `flopy` + are different packages, so testers can compare them on their own + models in one environment. +- **Feedback lands at this repo's pace,** without flopy's review and CI + on every beta fix. +- **The `4.x` branch stays short.** It's cut after MMP feedback has + settled the API, so it carries a stable codebase, not beta churn. + +The costs: integration issues (the shared-code protocol, tier A tests +against both engines) surface after MMP, not before; testers change +`import flopy4` to `import flopy` at the release candidates (a scripted +rename, documented in `migrating.md`). The integration plan rates the +code move as low risk, and the engine risk is exercised by the betas +either way. + +**Open: version numbers.** `flopy4` is at `0.0.1.dev0`. Versioning the +pre-releases `4.0.0aN`/`4.0.0bN` lines them up with `flopy==4.0.0` and +avoids a jump from 0.x. + +## What MMP means + +From `map.md`: all core and most peripheral functionality, released to a +wider test audience at USGS and Deltares, beta versioning. Its list +includes rough feature parity with 3.x, user-testing feedback, adapted +3.x docs and examples, comprehensive testing, and finalized maintenance, +release and MF6 version support plans. + +The SRS requirements due at MMP, plus the one MVP requirement still open: + +| req | MoSCoW | status (2026-11-03) | remaining | +|---|---|---|---| +| FUNC-7 round-trip with identical results (MVP) | C | Not done. `docs/dev/test.md` lists flopy3/4 comparison and round-trip tests as unchecked. | Track A | +| FUNC-8 larger-than-memory | M | Partial. dask is an optional extra; external arrays load eagerly (#516); no xarray IO backend (#368). | Track C | +| FUNC-10 wide range of MF6 versions | S | Mostly. `flopy4 mf6 sync`, `docs/mf6-versions.md`; tested against the latest release and MF6 `develop` only. | Decide whether "wide range" needs a test matrix of older releases. | +| FUNC-11 spatial and temporal units | C | Not audited. Grid adapters carry length units. | Audit, or defer past MMP (it's a C). | +| FUNC-14 reproduce the spec | S | Partial. `_contract.py` records the DFN commit; #299 open. | Track B | +| NFR-1 public vs internal API in docs | M | Not done. No API reference in the docs. | Track D | +| NFR-6 documented, type-hinted public API | S | Not done, same reason. | Track D | +| NFR-7 consolidated entry points | S | Partial. #342 (unified output loading) and #232 open. | Track B | + +The GitHub MMP milestone has 24 open issues. Many are questions +(#229, #240, #258, #259, #436, #523) that are either decisions for +Track B or can move past MMP. Triage it early, so the milestone matches +this doc. + +## MMP gate + +- [ ] Tier E (equivalence harness) passes on the registry corpus, with + any accepted gaps listed. FUNC-7 falls out of this. +- [ ] Round-trip load/write tests over the corpus. +- [ ] The bare-list `stress_period_data` bug fixed. +- [ ] The user-facing API decisions in Track B made and implemented. + Testers write code against them, so changing them after MMP breaks + them. +- [ ] Datetime period keys (#518) in, or explicitly deferred. +- [ ] API reference in the docs, public vs internal marked (NFR-1, NFR-6). +- [ ] Concepts doc (#217), and `migrating.md` covering the compat layer. +- [ ] Tier C scoreboard running here against MF6's autotests and the + examples, so the release notes can say how much flopy3 code runs. +- [ ] Install from GitHub works on Windows, Linux and macOS (SYS-2), with + the command documented. +- [ ] MMP milestone triaged: each open issue fixed, decided, or moved. + +## Tracks + +All run at once. The table says what each needs to deliver by MMP; work +after MMP is under "After MMP". + +| track | items | by MMP | detail | +|---|---|---|---| +| **A. Engine correctness** | bare-list bug; equivalence harness (tier E) over the registry; round-trip tests; fix what they find | everything | `flopy-integration-plan.md` "Tests", `docs/dev/test.md`. **Longest and least predictable. Start first.** | +| **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open" (branch `xarray-plan`) | +| **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md` (branch `xarray-plan`) | +| **D. Docs** | API reference; concepts (#217); 3.x examples adapted; `migrating.md` for compat; SDD update once export phase 1 lands | all | `map.md`, `srs.md`, `sdd.md` | +| **E. Compat layer** | generated `ModflowXxx` constructors and the Simulation/Model aliases, growing out of `flopy4/mf6/legacy.py`; tier C scoreboard against MF6 autotests and examples | running, with numbers | `flopy-integration-plan.md` "The compat layer" | +| **F. 3.x prep** (flopy `develop`) | 3.x items 1 to 5: utilities out of `flopy/mf6`, the shared-code protocol, upstream patches, deprecations, forward-compatible keywords | nothing required; items 1 and 2 by the `4.x` cut | `flopy-integration-plan.md` "3.x work" | +| **G. Packaging** | devtools release; drop the git dependency; xugrid range; Python 3.14 and the 3.11 question; numpy floor; lazy xugrid import | nothing required (GitHub installs); all by the first release candidate | `flopy-integration-plan.md` Phase 0 item 3; `xarray-package-plan.md` (lazy xugrid, grid composition) | + +Tracks F and G aren't needed for MMP, but start them now: they're small, +each piece is releasable on its own, and they decide how quickly the +release candidates follow MMP. + +### Dependencies between tracks + +- Track A's harness finds engine bugs, and fixing them may touch Track C + (period data, arrays). Keep Track C's changes small until the harness + is running. +- Track E's compat constructors wait for Track B's naming decisions, or + they get regenerated. +- Track D's API reference waits for Track B, and the SDD update waits + for export phase 1 (Track C). + +### The longest chain + +Equivalence harness → engine fixes → tier E passes → MMP. The other +tracks should finish inside it. If one starts to hold up the gate, +that's the time to add people to it. + +## After MMP + +From `flopy-integration-plan.md`, Phases 2 and 3: + +1. Work through MMP feedback. Make integration decision 2 (compat data + access), informed by the scoreboard numbers and the feedback. +2. Cut `4.x` once the API has settled and the integration plan's "Before + cutting `4.x`" list is done. +3. Move the old engine to the legacy path, bring `flopy4` in as + `flopy.mf6`, run tiers A to E there. +4. Release candidates `flopy==4.0.0rcN` from PyPI and conda-forge. +5. GA, then 3.x maintenance for a set time. + +## Decisions to make + +From the plans this doc links to, ordered by when they're needed: + +| decision | needed by | where | +|---|---|---| +| Pre-release version numbers (`4.0.0bN`?) | first alpha | here | +| `output.head`: which side renames (integration 3) | MMP | `flopy-integration-plan.md` | +| Remaining Track B API questions | MMP | Track B | +| FUNC-10 test matrix; FUNC-11 in or out of MMP | MMP | SRS | +| Compat data access (integration 2) | `4.x` cut | `flopy-integration-plan.md` | +| Legacy engine and its namespace (integration 1) | `4.x` cut | `flopy-integration-plan.md` | +| Where the top-level modules go (integration 6) | `4.x` cut | `flopy-integration-plan.md`, `xarray-package-plan.md` | +| numpy floor and Python range (integration 5) | first release candidate | `flopy-integration-plan.md` | +| 3.x keywords in 3.x (integration 4) | 3.x item 5 | `flopy-integration-plan.md` | +| 3.x support window (integration 7) | GA | `flopy-integration-plan.md` | diff --git a/flopy-integration-plan.md b/flopy-integration-plan.md new file mode 100644 index 00000000..0a598aa4 --- /dev/null +++ b/flopy-integration-plan.md @@ -0,0 +1,521 @@ +# Integrating flopy4 into flopy + +How hard it will be to replace flopy's mf6 engine with this project's, and +how to go about it. Written 2026-10-28 against pyphoenix `develop` +(`000f63c3`) and flopy `develop` (`b9ee6fa2`, 3.12.0.dev0). Refreshed +2026-11-03 against pyphoenix `develop` (`8d25e4b7`) and flopy `develop` +(`aca52382`, 3.13.0.dev0, after the 3.12.0 release). Covers discussions +#68 (rollout), #35 (MF6 versions) and #37 (refactoring 3.x), and issue +#113 (flopy3 interfaces). + +### Since the first draft + +- flopy 3.12.0 is out, and `develop` now requires Python `>=3.12` and + numpy `>=2.2` (#2843, #2844, per SPEC 0). The dependency gap is + smaller: flopy4's numpy floor is a minor version above flopy's, not two + majors, and the Python mismatch is now only flopy4's `<3.14` cap. +- `flopy4 mf6 sync --developmode` landed (#522), so MF6's CI switch no + longer waits on it. +- The flopy3 adapter gained stress-period list data, geodataframe export, + `get_package` by name or type, and grids for every discretization + (#531). packagedata/connectiondata are still not wrapped, on purpose + (see 3.x item 2). +- The devtools registry grew to 261 models; 260 are known to load. +- Unchanged: the bare-list `stress_period_data` bug (re-run 2026-11-03, + same error), the git dependency on devtools (PyPI still has 1.9.2), + the exact xugrid pin (0.15.3 is still the latest release, so the pin + is current, just not a range), the `StructuredGrid` `ia`/`ja` TODO, and + the unchecked items in `docs/dev/test.md`. Nothing from "3.x work" has + started on flopy `develop`. + +## Summary + +- **The surgery is smaller than it looks.** Everything flopy4 needs from + flopy outside `flopy/mf6` (grids, `ModelTime`, `MfGrdFile`, plotting and + export through `ModelInterface`/`PackageInterface`) it already imports + from flopy3 and uses today. The old engine touches the rest of flopy at + about a dozen places, all known (listed below). What's hard is the + user-facing API, not the code layout. +- **Don't put the new engine into 3.x**, in place of the old one or beside + it. The new required dependencies and the Python cap break + 3.x users in a minor release, and the API differences are about + behaviour, not just names, so adapters can't hide them. Use 3.x instead + to make the swap cheap (see "3.x work"). +- **In 4.x, move the old engine, don't delete it.** Put the new engine at + `flopy.mf6`, the old one, frozen, at a legacy path (e.g. + `flopy.mf6.legacy`), and add a thin compatibility layer for the common + flopy3 idioms. The compat layer either behaves exactly as flopy3 did or + raises an error naming the replacement. It never gets close and moves on. +- **Tests: re-point, don't rewrite.** The old engine's own tests move with + it to the legacy path, unchanged. The shared-code tests (grids, plot, + export, binary files, modpath) must pass against both engines. The + flopy3 mf6 tests also run against the compat layer as a scoreboard, + allowed to fail, to measure it. flopy4's ~1190 tests come over as the new + engine's own suite. The release gate is whether the same models produce + the same heads and budgets, not whether old tests pass. +- **The best test corpus isn't in flopy.** MF6's autotests (420 files) and + modflow6-examples (76 scripts) use the flopy3 mf6 API more, and more + typically, than flopy's own tests. They compare simulation results, and + MF6's CI is where flopy's versioning hurts most. Make them the + acceptance tests for the compat layer and for version decoupling. +- **Keep the 4.x branch short.** Finish the core here, upstream every + change that isn't the swap itself to `develop`, and only then cut the + branch, once MMP feedback has settled the API. Merge `develop` into it + weekly. +- **Release shape** (decided 2026-11-03; see `docs/dev/mmp-plan.md`). + Alphas and betas, up to and including MMP, ship from this repo as + `flopy4`, installed from GitHub, so testers can run both engines side by + side. The 4.x branch ships release candidates `flopy==4.0.0rcN` (pip + skips pre-releases unless asked), and GA is the last of them. + +## Where we stand + +### flopy + +| | lines of Python | +|---|---| +| `flopy/` | 195k | +| `flopy/mf6/` | 80k | +| `flopy/mf6/modflow/` (generated classes) | 42k | +| `flopy/mf6/` handwritten (data layer, base classes, utils) | 38k | + +Where the old engine reaches outside `flopy/mf6`: + +- `flopy/plot/map.py`, `crosssection.py`, `plotutil.py`: on + `p.parent.version == "mf6"`, read `pkg.stress_period_data.array[kper]`, + `pkg.packagedata.array`, `pkg.connectiondata.array` (recarrays with a + `cellid` field) and `pkg.package_type`. +- `flopy/modpath/mp7.py`: `isinstance(flowmodel, MFModel)`, `model_type`, + `get_package("DIS")`, `model_ws`. +- `flopy/utils/gridgen.py`: imports `ModflowGwfdis` to build DIS packages. +- `flopy/utils/binaryfile/reverse.py`: `MFSimulation`, `ModflowTdis`. +- `flopy/discretization/{structured,vertex,unstructured}grid.py`, + `flopy/utils/zonbud.py`: import `MfGrdFile` from + `flopy/mf6/utils/binarygrid_util.py`. It doesn't depend on the engine; + it's just stored in the mf6 directory. + +Inside `flopy/mf6/utils`, the modules that work without the engine +(`binarygrid_util`, `mfsimlistfile`, `postprocessing`, `lakpak_utils`, +`mfenums`) can move out of the directory. The ones that need it +(`model_splitter` (4175 lines, 109 references to engine internals), +`output_util`, `mfobservation`, `testutils`, codegen) can't. + +### flopy4 + +- Already a flopy3 client: `flopy4.mf6.adapters.Flopy3Model` / + `Flopy3Package` implement `ModelInterface`/`PackageInterface` (with + `.array`) to reuse flopy3's plotting and export. Since #531 that covers + stress-period list data (fill-forward, aux columns, duplicate cells + summed or averaged as flopy3 does), `to_geodataframe`, `get_package` by + name or type, and a grid for every discretization type. Not wrapped: + packagedata/connectiondata, which only flopy3's `plot_bc` reads for + SFR/LAK/MAW/UZF. `to_grid()` returns flopy3 grids, and the output + readers use flopy3 grids. One upstream TODO is open: + `flopy4/adapters.py:27` (`StructuredGrid` `ia`/`ja`). +- Loads 260 of the 261 models in the devtools registry (`KNOWN_PASSING` + in `test/mf6/test_mf6_load_all_models.py`; the rest are `xfail`). As of + 2026-10-28 it loaded faster than flopy3 on 127 of the 165 benchmark + models (median flopy4/flopy3 time ratio 0.44), and slower on large + inline arrays. The benchmarks weren't re-run for the refresh. +- Not done yet (`docs/dev/test.md`): running a model written by flopy3 + and by flopy4 and comparing the results, round-trip tests over the + corpus, and porting the flopy3 autotests and MF6 examples. +- 1190 test functions (2157 collected with parameters). + +### API distance: constructors + +flopy3 and flopy4 have 100 component classes in common, with 1101 flopy3 +constructor keywords between them (leaving out flopy3 plumbing like +`loading_package` and `pname`): + +| | count | +|---|---| +| same name in flopy4 | 836 (76%) | +| `*_filerecord` / `*printrecord` → `*_file` / `*print` | 106 | +| `timeseries` / `timearrayseries` / `observations` → `ts` / `tas` / `obs` children | 66 | +| `dev_*` options (flopy4 codegen leaves them out unless `developmode`) | 36 | +| `maxbound` (derived in flopy4) | 12 | +| `parent_*` (flopy3 plumbing) | 12 | +| other (`perioddata`, `packagedata` on DIS, nam `packages`/`list`, IMS `rcloserecord`, TS records…) | 33 | + +Only 20 classes match exactly, but nearly every difference follows a rule +a generator can apply. `LegacyKwargs` (`flopy4/mf6/legacy.py`) already +does this for period data. The constructor gap can be closed with +generated code. + +### API distance: everything else + +Running a flopy3 quickstart against flopy4: + +| flopy3 idiom | flopy4 | +|---|---| +| `MFSimulation(sim_name=, sim_ws=, exe_name=)` | `Simulation(name=, workspace=)` | +| `ModflowGwfdis(gwf, ...)`: the parent positional | `Dis(parent=gwf, ...)`: keyword only | +| `ModflowGwf(sim, modelname=)` | `Gwf(parent=sim, name=)` | +| `stress_period_data=[...]` (a bare list for period 0) | raises (`'list' object has no attribute 'items'`). [`migrating.md`](docs/migrating.md) says a list means period 0, so this looks like a bug. | +| `sim.write_simulation()` / `run_simulation()` | `write()` / `run()` | +| `gwf.modelgrid` | `gwf.dis.to_grid()` | +| `gwf.get_package("chd")`, `sim.get_model()` | none (`gwf["chd0"]`) | +| `gwf.output.head()` → `HeadFile` | `gwf.output.head` → `DataArray`, a property. **Same name, different kind of thing.** | +| `npf.k.array`, `.get_data()`, `.set_data()` | `npf.k` is the value itself (a `float` here) | +| `gwf.plot()`, `pkg.export()` | via `Flopy3Model` | +| default filenames `m.chd`, `m_0.chd`, `s.ims` | `m.chd0.chd`, `m.chd1.chd`, `s.ims0.ims` | + +Top-level methods are cheap to alias. The data layer is not: flopy3's +`MFArray`/`MFList`/`MFTransientList` objects (storage modes, external +file management, `DataStorageType`, `lazy_io`, `use_pandas`, +`set_record`, `mfstructure` introspection) are an API in themselves, and +reimplementing them on top of flopy4 means rewriting flopy3. Same for +`model_splitter`. + +### Who uses the flopy3 mf6 API, and how + +| corpus | size | most-used beyond constructors | +|---|---|---| +| flopy `autotest` | 948 tests; 552 in the 41 files touching `flopy.mf6` | `test_mf6.py` and `regression/test_mf6.py` use the data layer heavily; the shared-code tests mostly use mf6 to build fixtures | +| MF6 `autotest` | 420 of 453 test files | `register_ims_package` (340), `modelgrid` (109), `get_model` (86), `output.*` (78), `write_simulation` (77), `MFSimulation.load` (51 files), `run_simulation`, `get_package`; `set_data`/`simulation_data` in only 14 files | +| modflow6-examples | 76 scripts | `.array`, `.get_data()` (partly on `HeadFile`), `output.*`, `simulation_data` (17) | + +MF6's tests are mostly "build it with constructors, run it, read +`output`, compare". That's the subset a compat layer can cover exactly. +MF6 also regenerates flopy's classes from its own DFNs (`pixi run +update-flopy`), with `dev_*` options, which is the version coupling #35 +wants to remove. + +### Dependencies and platforms + +| | flopy 3.x | flopy4 | +|---|---|---| +| required | numpy, matplotlib, pandas | + attrs, cattrs, jinja2, lark, xarray*, xugrid (`==0.15.3`), scipy, sparse, pydantic, modflow-devtools (git) | +| numpy | `>=2.2.0` (3.13.0.dev0; `>=1.20.3` through 3.12.0) | `>=2.3.4,<2.6` | +| Python | `>=3.12` (3.13.0.dev0; `>=3.11` through 3.12.0), 3.14 tested | `>=3.11,<3.14` | +| devtools | `>=1.7,<2` (codegen/test), `modflow_devtools.dfn` | unreleased 1.10.0.dev1 from git (PyPI has 1.9.2), `modflow_devtools.dfns` (marked experimental), DFN schema `2.0.0.dev3` | + +\* through xugrid; xarray is listed as optional, but `attrs_xarray` and +the array fields need it. + +## Question 1: the new engine in 3.x? + +**Replacing the old engine: no.** Even with adapters: + +- **Dependencies.** Ten or so new required packages, an exact pin, a + dependency installed from git, and Python 3.14 dropped. (Since flopy + raised its own floors to Python 3.12 and numpy 2.2, the numpy floor is + only a minor version apart.) Any one of these breaks + environments on `pip install -U flopy`, conda-forge and locked-down + sites especially. In a minor release that breaks semver in spirit, if + not in letter. +- **Behaviour.** The data layer is where flopy3 scripts live, and its + semantics (mutable data objects, storage modes, external-file control, + recarray views) aren't the new design's. An adapter that comes close + is worse than one that fails: it writes different files and moves on. +- **Files written.** Default names differ, and the formatting differs. + PEST workflows (pyemu `PstFrom`, hand-made templates) and scripts that + open input files by name depend on the exact layout. +- **Downstream.** `model_splitter`, pyemu and modflow-setup are built on + the old internals. + +**Shipping it beside the old engine (e.g. `flopy.mf6next` under an +extra): possible, but not recommended.** It gets the new engine to users +early, with no long-lived branch. But it puts code that's still changing +fast on 3.x's release schedule and compatibility promises, means both +engines in every 3.x maintenance release, and needs a second namespace +switch at 4.0. `flopy4` betas from this repo get it to users just as +early, with none of that. + +### 3.x work + +This is what makes the 4.x swap mechanical. All of it lands on flopy +`develop`, and none of it needs the new engine. It takes up #37's +refactoring list where it helps the swap. + +1. **Move engine-agnostic utilities out of `flopy/mf6`.** Move + `binarygrid_util`, `mfsimlistfile`, `postprocessing`, `lakpak_utils` + and `mfenums` to e.g. `flopy/utils/mf6/`, with re-exports at the old + paths. Then `flopy/mf6` holds only the engine and can be moved + wholesale. +2. **A small neutral protocol for shared code.** Replace the + `version == "mf6"` branches in plot/modpath with a few methods that + flopy3's mf6 classes implement now and flopy4's natively later: e.g. + `boundary_cellids(kper)`, `period_records(kper)` (a recarray or + DataFrame), `get_dis()`, `model_ws`. Same for `gridgen` (build a + generic grid spec, not a `ModflowGwfdis`) and `binaryfile/reverse.py` + (read TDIS with a small reader). This is #113's question, answered on + the flopy3 side. +3. **Upstream what flopy4 patches**: `StructuredGrid` `ia`/`ja`, and + anything else the adapters subclass. +4. **Deprecate in 3.x what 4.x won't have**, with warnings that point to + the replacement: `DataStorageType`, `lazy_io`, `use_pandas`, + `ExtFileAction` and fine-grained external-file control, `mfstructure` + introspection, `set_record`, and so on, as was done for `check` (#2357). + Users see the warnings for a release or two before anything breaks. +5. **Optionally, accept 4.x idioms in 3.x**: `parent=` as a keyword, + `sim.write()`/`sim.run()`, the `stress_period_data` names (the inverse + of `legacy.py`), and `*_file` for `*_filerecord`. Then users can write + code that runs on both, and migrate before they upgrade. It's cheap + for constructor keywords and top-level methods (codegen templates); + values are out of scope. +6. **Release modflow-devtools** with both `dfn` (for 3.x) and a stable + `dfns`, and settle the DFN schema version. This blocks any flopy + release that includes the new engine. + +## Question 2: how to do the replacement + +### Layout in 4.x + +``` +flopy/ + mf6/ <- this project's flopy4/mf6 (generated + handwritten) + compat/ <- generated flopy3 names: ModflowGwfdis, MFSimulation, ... + legacy/ <- 3.x's flopy/mf6, frozen (git mv, history kept) + _core/ or similar <- flopy4's top-level modules (uio, spec, dimensions, + attrs_xarray, mixins, protocols) and flopy4/adapters.py + discretization/, plot/, export/, utils/, modpath/, ... <- unchanged +``` + +- `flopy.mf6.ModflowGwfdis` etc. stay importable, served from `compat`. +- `import flopy.mf6.legacy as mf6` gets the old engine back exactly, for + code that uses the data layer or `model_splitter`. This is TensorFlow's + `tf.compat.v1`. Freeze it at the last 3.x DFNs: it doesn't follow sync, + and users who need newer MF6 features migrate. Remove it in 5.0, or + after a set number of minors. +- The legacy path also means that if pyemu and modflow-setup lag, users + have a way out: they can switch an import instead of pinning + `flopy<4`. +- Rename imports (`flopy4.` → `flopy.`) with a script, not by hand. The + generated modules are regenerated, not edited. +- `flopy4` depends on `flopy>=3.9.5` today. After the merge those are + ordinary internal imports. + +### The compat layer + +Generated from the DFNs alongside the classes (so `sync` keeps it +current), plus handwritten pieces for Simulation and Model: + +- **Constructors.** A `ModflowXxx` subclass or factory per component: + the parent positional, the keyword renames in the table above, flopy3 + value shapes (`head_filerecord="m.hds"`, `saverecord=`, a bare list for + period 0). It produces a real flopy4 object. `legacy.py` is the + pattern to extend. +- **Simulation/Model methods**: `write_simulation`, `run_simulation` + (returning `(success, buff)`), `MFSimulation.load(sim_ws=...)`, + `get_model`, `get_package`, `register_ims_package`, `modelgrid`, + `model_ws`, `set_sim_path`, `name_file`. These are mostly aliases. +- **Data access is the decision.** flopy3's `npf.k` is an object with + `.array`/`.get_data()`; flopy4's is the value. Either: + (a) don't support it: `.array` on a value raises with a hint (`npf.k` + is the array now); or (b) compat classes return a small read-only view + with `.array`/`.get_data(kper)`. The new classes stay clean, but then + the same package behaves differently depending on which class built + it. I lean toward (a) plus the legacy path. Decide with the scoreboard + numbers, not ahead of them. +- **`gwf.output.head()`**: `head` is a `DataArray` property in flopy4 and + a method returning `HeadFile` in flopy3. Pick one before 4.0. Something + callable that also acts as a DataArray is possible but too clever. I'd + rename one side (e.g. flopy4's `output.head` → `output.heads`, or + `output.head()` returning the DataArray) so compat can keep the old + name. +- **Rule**: same result or a clear error. Every compat class or method + warns (`FutureWarning`, once per call site) and links to + `docs/migrating.md`. + +### Tests + +| tier | what | in 4.x | +|---|---|---| +| A. shared code | grid, plot, export, binary files, gridintersect, modpath, zonbud, utils (~400 tests that never touch mf6, plus the shared-code tests that use mf6 for fixtures) | must pass. Run the mf6 fixtures through both engines (a fixture parameter) | +| B. old engine internals | `test_mf6_set_data_*`, `max_columns`, `header_preservation`, `DataStorageType`, external-file and `lazy_io` tests, `model_splitter`, much of `test_mf6.py` / `regression/test_mf6.py` | re-point to `flopy.mf6.legacy` (an import rewrite); they protect the frozen engine | +| C. compat scoreboard | tier B's end-to-end tests (build/load → write → run → compare), MF6 autotests, modflow6-examples, all against `flopy.mf6` compat | report, don't gate: an `xfail(strict=False)` manifest like `test_mf6_load_all_models.py`. A test that starts passing gets pinned | +| D. new engine | this repo's ~1190 tests, benchmarks | move in, must pass | +| E. equivalence | every registry model: load/write/run with legacy and with new, compare heads and budgets; also compare written input files after normalization | **the gate for 4.0** | + +The flopy tests aren't so tied to the old engine that they need +rewriting. They split cleanly, and the ones that are tied (B) should +stay tied, to the legacy path. Tier C, the old tests against the compat +layer, is what #113 suggested ("get all the old tests and examples"). +It's useful as measurement, but shouldn't be what decides the release: +a red test there means the compat layer needs work or we should document +a break. + +Start E here, before the branch: it's the open item in +`docs/dev/test.md`, and it finds engine bugs, which the compat layer +can't fix. + +## MF6 versions in 4.x + +The swap is what makes #35 possible: 3.x can't decouple, because its +classes are generated into the package and regenerating them is a +developer step. + +- A 4.x release ships classes synced to the latest MF6 release. + `flopy mf6 sync ` regenerates them, plus the + compat classes (they're generated too, so they're in + `GENERATED_FILES`). `docs/mf6-versions.md` comes over as is, renamed. +- The legacy engine is frozen and doesn't sync. Say so in its docs. +- **MF6's CI** replaces `update-flopy` (`generate_classes --dfnpath ...`) + with `flopy mf6 sync doc/mf6io/mf6ivar/dfn --developmode`. Sync + leaves out `dev_*` options by default, which MF6's tests use (36 of + them in the constructor table); `--developmode` (#522) keeps them. This + is the first real user of decoupling, and the best test of it. +- `sync` writes into site-packages. That's fine for 4.0. Revisit the + per-user directory (`user-dir-classes-plan.md`) once `mf6 --spec` + lands, as that plan says. +- `Simulation.run`'s version check comes over unchanged. Compat's + `run_simulation` gets it too. + +## Order of work + +**Phase 0 (here, through MMP):** + +1. Fix the bare-list `stress_period_data` bug the probe found (still + open on `8d25e4b7`). +2. Equivalence harness (tier E) over the registry, starting with the + models that both load. +3. Release modflow-devtools (3.x item 6); drop the git dependency; unpin + xugrid, or pin it as a range; support Python 3.14; drop 3.11 to match + flopy, or keep it; decide the numpy + floor (4.0 can require numpy 2). +4. Settle the open API questions that the compat layer depends on: + `output.head`, `get_package`/`modelgrid` on Model, `name` vs + `sim_name`. +5. Survey pyemu, modflow-setup and the MF6 repo's scripts for which + flopy3 mf6 internals they use, to decide what goes in compat versus + the legacy path. +6. Build the compat layer here (generated constructors and the + Simulation/Model aliases, growing out of `flopy4/mf6/legacy.py`), and + run the tier C scoreboard against MF6's autotests and the examples + from here. Neither needs flopy's repo. +7. Release `flopy4` alphas and betas from GitHub. MMP is the last beta. + +**Phase 1 (flopy `develop`, 3.x):** items 1 to 5 from "3.x work" (item 6, +the devtools release, is in Phase 0). Each is a normal PR and releasable +on its own. + +**Phase 2 (cut `4.x` from `develop`, after MMP):** + +1. `git mv flopy/mf6 flopy/mf6/legacy` (or wherever it goes), fix the + imports, rewrite tier B's imports. CI green with only the old engine. +2. Bring flopy4 in (scripted rename) as `flopy.mf6`, with its tests + (tier D). CI green. +3. Implement the shared-code protocol (from 3.x item 2) for flopy4. Run + tier A against both engines. +4. Bring the compat layer over and add flopy's own tests to the tier C + scoreboard (MF6 autotests and examples already run from Phase 0). +5. Release candidates `4.0.0rc1…` once tiers A and D are green on + `4.x`, from PyPI and conda-forge. + +Merge `develop` into `4.x` weekly. Phase 1 keeps conflicts away from +anything but `flopy/mf6`. + +**Phase 3:** 4.0 GA, the last release candidate. Cut `v3.x` from `develop` for maintenance and make +`4.x` the new `develop`. Then 3.x gets fixes only, for a fixed time +(map.md says "limited"; pick a date and announce it at 4.0). + +## Critical path + +The phases above are an order for each piece of work, not one queue. Most +of Phases 0 and 1 can run at the same time. + +### Parallel tracks + +Start all four at once: + +| track | where | items | needed by | +|---|---|---|---| +| **A. Engine correctness** | here | bare-list bug, equivalence harness (tier E), then fixing what it finds | MMP. **Longest and least predictable; start first** | +| **B. Packaging** | devtools, here | devtools release, drop the git dependency, xugrid range, Python 3.14, numpy floor | the first release candidate (GitHub installs don't need it) | +| **C. 3.x prep** | flopy `develop` | 3.x items 1 to 5 | items 1 and 2 by the branch cut; items 3 to 5 can land later and merge into `4.x` weekly | +| **D. Decisions and survey** | team | open decisions, downstream survey | decision 3 by MMP; 1, 2 and 6 by the branch cut; the rest as below | + +### Before MMP + +Tracked in `docs/dev/mmp-plan.md`. From this plan: tier E passes on the +registry corpus, decision 3 (`output.head`) is made and implemented, +since testers write code against it, and the tier C scoreboard runs from +here. + +### Before cutting `4.x` + +- [ ] MMP released, and its feedback worked through. +- [ ] Decision 2 (compat data access) made and implemented. +- [ ] 3.x item 1 merged: engine-agnostic utilities out of `flopy/mf6` + (the legacy move depends on it). +- [ ] 3.x item 2 merged: the shared-code protocol, implemented by + flopy3's mf6 classes (Phase 2 step 3 depends on it). +- [ ] Decision 1: legacy engine yes or no, and its namespace (Phase 2 + step 1 is the move). +- [ ] Decision 6: where flopy4's top-level modules go (Phase 2 step 2 + is the move-in). +- [ ] Equivalence harness running in CI here, so its results carry over. + It doesn't have to pass yet. + +### Before the first release candidate (`4.0.0rc1`) + +- [ ] modflow-devtools released, and flopy4 depends on it from PyPI. + PyPI rejects packages with direct-URL (git) dependencies, so this is a + hard blocker, not just tidying. +- [ ] xugrid pinned to a range; Python range matches flopy's (`>=3.12`, + 3.14 included). +- [ ] Tiers A and D green on `4.x`, and tier E still passing there. +- [ ] Tier C scoreboard running on `4.x` with flopy's tests added, so + the release notes can say how much of the flopy3 API the compat layer + covers. +- [ ] Downstream survey done, and its results sorted into compat vs + legacy-only. +- [ ] MF6's CI running against `4.x` with `flopy mf6 sync --developmode`, + at least as a non-blocking job. + +### Before GA + +- [ ] 3.x item 4 (deprecations) shipped in a 3.x release, ideally two + releases before 4.0, so users see the warnings first. +- [ ] Decision 7 (3.x support window) made and announced with 4.0. + +### The longest chain + +Equivalence harness → engine fixes → tier E passes → MMP → API settled → +`4.x` cut → release candidates → GA. The +other tracks are shorter and should finish inside it. If one of them +starts to hold up a checkpoint above, that's the time to add people to +it. + +## Sizing + +Rough, from the inventory above. "Small" is days, "medium" is weeks, +"large" is a month or more for one person. + +| item | size | why | +|---|---|---| +| move engine-agnostic utils out, re-export | small | imports only | +| shared-code protocol (plot, mp7, gridgen, reverse) | medium | about 10 call sites, plus tests | +| 3.x deprecations | small–medium | warnings plus docs | +| 3.x forward-compatible keywords | medium | codegen templates, and 3.x's own tests | +| devtools release, dependency cleanup, Python 3.14 | medium | xugrid and numpy floors may cascade | +| equivalence harness | medium to build; fixing what it finds is open-ended | the biggest unknown | +| legacy move and import rewrite | small | mechanical | +| flopy4 move-in | small–medium | renames and packaging, plus CI time | +| compat constructors (generated) | medium | rules known, 76% free | +| compat Simulation/Model methods | small–medium | mostly aliases | +| compat data access | unknown until decided | (a) small, (b) large | +| MF6 CI switch to sync | small (`developmode` sync exists, #522) | | +| `model_splitter` for the new engine | large | 4175 lines on old internals; legacy-only in 4.0 is fine | + +The risks are in the engine itself: whether flopy4 writes models that +run and give the same answers across the whole corpus (tier E). The code +move isn't a risk. + +## Open decisions + +1. Legacy engine in 4.x: yes or no, its namespace, and how long it stays. + (I recommend yes, `flopy.mf6.legacy`, until 5.0.) +2. Compat data access: (a) raise with a hint, or (b) views. +3. `output.head`: which side renames. +4. Whether to accept 4.x keywords in 3.x (3.x item 5). +5. numpy floor and Python range for 4.0. flopy `develop` is at Python + `>=3.12` and numpy `>=2.2`; the question is whether flopy4 needs more + than that. +6. Where flopy4's top-level modules (`uio`, `spec`, `dimensions`, ...) go: + `flopy._core`, or under `flopy.mf6` if nothing else will use them. + NFR-4 (one API for MF6 and older MODFLOW programs) says somewhere + shared. +7. How long 3.x is maintained after 4.0. From 42b4731258d9f1036cbe77503201e59109a7ea68 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 11:58:05 -0800 Subject: [PATCH 07/77] drop cross-branch notes now that the plans share a branch Co-Authored-By: Claude Opus 5.5 --- docs/dev/mmp-plan.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/dev/mmp-plan.md b/docs/dev/mmp-plan.md index 804a9485..5f2a1e25 100644 --- a/docs/dev/mmp-plan.md +++ b/docs/dev/mmp-plan.md @@ -93,8 +93,8 @@ after MMP is under "After MMP". | track | items | by MMP | detail | |---|---|---|---| | **A. Engine correctness** | bare-list bug; equivalence harness (tier E) over the registry; round-trip tests; fix what they find | everything | `flopy-integration-plan.md` "Tests", `docs/dev/test.md`. **Longest and least predictable. Start first.** | -| **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open" (branch `xarray-plan`) | -| **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md` (branch `xarray-plan`) | +| **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open" | +| **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md` | | **D. Docs** | API reference; concepts (#217); 3.x examples adapted; `migrating.md` for compat; SDD update once export phase 1 lands | all | `map.md`, `srs.md`, `sdd.md` | | **E. Compat layer** | generated `ModflowXxx` constructors and the Simulation/Model aliases, growing out of `flopy4/mf6/legacy.py`; tier C scoreboard against MF6 autotests and examples | running, with numbers | `flopy-integration-plan.md` "The compat layer" | | **F. 3.x prep** (flopy `develop`) | 3.x items 1 to 5: utilities out of `flopy/mf6`, the shared-code protocol, upstream patches, deprecations, forward-compatible keywords | nothing required; items 1 and 2 by the `4.x` cut | `flopy-integration-plan.md` "3.x work" | From 0f38319d4627ad3e6b9c8c50d3a80f473e674075 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 12:07:08 -0800 Subject: [PATCH 08/77] add flopy3 migration plan, link it from the MMP plan Co-Authored-By: Claude Opus 5.5 --- docs/dev/mmp-plan.md | 12 +- flopy3-migration-plan.md | 254 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 261 insertions(+), 5 deletions(-) create mode 100644 flopy3-migration-plan.md diff --git a/docs/dev/mmp-plan.md b/docs/dev/mmp-plan.md index 5f2a1e25..ae2efd6c 100644 --- a/docs/dev/mmp-plan.md +++ b/docs/dev/mmp-plan.md @@ -79,8 +79,9 @@ this doc. - [ ] Datetime period keys (#518) in, or explicitly deferred. - [ ] API reference in the docs, public vs internal marked (NFR-1, NFR-6). - [ ] Concepts doc (#217), and `migrating.md` covering the compat layer. -- [ ] Tier C scoreboard running here against MF6's autotests and the - examples, so the release notes can say how much flopy3 code runs. +- [ ] Migration scoreboard running here against MF6's autotests and the + examples, so the release notes can say how much flopy3 code runs + unchanged and how much after `flopy4 migrate`. - [ ] Install from GitHub works on Windows, Linux and macOS (SYS-2), with the command documented. - [ ] MMP milestone triaged: each open issue fixed, decided, or moved. @@ -96,7 +97,7 @@ after MMP is under "After MMP". | **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open" | | **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md` | | **D. Docs** | API reference; concepts (#217); 3.x examples adapted; `migrating.md` for compat; SDD update once export phase 1 lands | all | `map.md`, `srs.md`, `sdd.md` | -| **E. Compat layer** | generated `ModflowXxx` constructors and the Simulation/Model aliases, growing out of `flopy4/mf6/legacy.py`; tier C scoreboard against MF6 autotests and examples | running, with numbers | `flopy-integration-plan.md` "The compat layer" | +| **E. Migration** | classify every flopy3 difference (accept, deprecate, break); runtime aliases in flopy4 itself (positional parent, `register_ims_package`, `write_simulation`, `get_package`, `modelgrid`, `*_filerecord`, ...); the shared `Legacy` table; the `flopy4 migrate` codemod; the corpus scoreboard over MF6 autotests and examples. Generated `ModflowXxx` class names stay in flopy 4.x's compat layer | classification, aliases and scoreboard; the codemod if time allows | `flopy3-migration-plan.md`; `flopy-integration-plan.md` "The compat layer" | | **F. 3.x prep** (flopy `develop`) | 3.x items 1 to 5: utilities out of `flopy/mf6`, the shared-code protocol, upstream patches, deprecations, forward-compatible keywords | nothing required; items 1 and 2 by the `4.x` cut | `flopy-integration-plan.md` "3.x work" | | **G. Packaging** | devtools release; drop the git dependency; xugrid range; Python 3.14 and the 3.11 question; numpy floor; lazy xugrid import | nothing required (GitHub installs); all by the first release candidate | `flopy-integration-plan.md` Phase 0 item 3; `xarray-package-plan.md` (lazy xugrid, grid composition) | @@ -109,8 +110,8 @@ release candidates follow MMP. - Track A's harness finds engine bugs, and fixing them may touch Track C (period data, arrays). Keep Track C's changes small until the harness is running. -- Track E's compat constructors wait for Track B's naming decisions, or - they get regenerated. +- Track E's aliases and codemod rules wait for Track B's naming + decisions (and default file naming), or they get regenerated. - Track D's API reference waits for Track B, and the SDD update waits for export phase 1 (Track C). @@ -142,6 +143,7 @@ From the plans this doc links to, ordered by when they're needed: | Pre-release version numbers (`4.0.0bN`?) | first alpha | here | | `output.head`: which side renames (integration 3) | MMP | `flopy-integration-plan.md` | | Remaining Track B API questions | MMP | Track B | +| Default file names (`m.chd` vs `m.chd0.chd`) | the Part 1 aliases; decide early, it changes flopy4's own tests | `flopy3-migration-plan.md` | | FUNC-10 test matrix; FUNC-11 in or out of MMP | MMP | SRS | | Compat data access (integration 2) | `4.x` cut | `flopy-integration-plan.md` | | Legacy engine and its namespace (integration 1) | `4.x` cut | `flopy-integration-plan.md` | diff --git a/flopy3-migration-plan.md b/flopy3-migration-plan.md new file mode 100644 index 00000000..f2c49d69 --- /dev/null +++ b/flopy3-migration-plan.md @@ -0,0 +1,254 @@ +# flopy3 → flopy4 migration: aliases first, then a codemod + +Written 2026-11-03 against pyphoenix `develop` (`000f63c3`). Prompted by +mjreno's flopy4 round-trip PRs on MF6 (MODFLOW-ORG/modflow6#3048, #3057, +#3067). Counts of flopy3 keywords and of MF6 autotest usage are from +`flopy-integration-plan.md`, not recounted here. + +## Summary + +- **The MF6 PRs don't need flopy4 build code.** #3067 builds each model + with flopy3, loads the written input with `Simulation.load`, writes it + back with `sim.write()`, reruns mf6 and compares. That covers the + loader and writer across 334 test files. Duplicating 334 `build_models` + functions in flopy4 to also cover the constructors would double the + maintenance for little gain. +- **Shrink the gap before building the tool.** Many differences between + the flopy3 and flopy4 APIs come from implementation details (attrs' + `kw_only`, DFN field names), not from design decisions. Accepting the + flopy3 forms at runtime (Part 1) makes most flopy3 scripts run + unchanged, with warnings, and leaves the codemod much less to do. +- **The codemod applies the runtime aliases statically.** Each alias is a + table entry: the flopy3 form, the flopy4 form and a converter. The + runtime shim and the codemod (Part 2) both read the same table, so a + rewrite the codemod can't make (the value is a variable, or built in a + loop) can be left alone, and the code still runs. The codemod only + needs `# TODO(flopy4)` comments for differences that have no alias. +- **Migration becomes two steps:** first make it run (upgrade, fix + warnings as convenient), then make it idiomatic (`flopy4 migrate`). + Neither step requires the other. + +## Part 1: make the gap smaller + +Each flopy3 → flopy4 difference falls into one of three groups: + +- **Accept**: take the flopy3 form for good, with no warning. The + difference was accidental, and the flopy3 form costs nothing. +- **Deprecate**: take the flopy3 form with a `DeprecationWarning` that + names the replacement. The new form is better, but the old one is easy + to convert. +- **Break**: the meaning changed. Raise an error with a hint, or document + the change. The codemod flags these and doesn't rewrite them. + +### Constructors + +| flopy3 | flopy4 today | proposal | why | +|---|---|---|---| +| parent positional: `ModflowGwfdis(gwf, ...)` | `parent=` keyword only | **accept** | `kw_only=True` exists so attrs allows optional base fields before required subclass fields (`component.py:232`), not to stop positional use. `LegacyKwargs.__call__` can move one positional argument to `parent=` before attrs sees it. `Dis(gwf, nlay=3)` reads fine. | +| bare list `stress_period_data=[...]` | raises `'list' object has no attribute 'items'` | **accept (bug fix)** | `migrating.md` already says a list means period 0. | +| `*_filerecord` / `*printrecord` (106 keywords), e.g. `head_filerecord="m.hds"` or `[("m.hds",)]` | `head_file=` | **deprecate** | `head_file` is better. Convert each flopy3 value shape (a string, a 1-tuple, a list of 1-tuples) to a path. These are generated `Legacy` entries like the period-data ones. | +| `perioddata`, `steady_state`/`transient`, `saverecord`/`printrecord`, ... | `stress_period_data` | **deprecate** (done) | Already `LegacyKwargs` (`legacy.py`, `_LEGACY_PERIOD_KWARGS`). | +| `timeseries=`, `timearrayseries=`, `observations=` (66 keywords) | `ts` / `tas` / `obs` children | **deprecate** | flopy3 accepts a dict (`{"filename": ..., "timeseries": ..., ...}`) or a file name. The common shapes convert. Raise on any other shape. | +| `maxbound=` (12) | derived from the data | **accept, check it** | Drop it if it agrees with the data, raise if it doesn't. A warning only adds noise: the value is redundant, not wrong. | +| `pname=` | `name=` | **accept** | A synonym. | +| `parent_*`, `loading_package` (12) | — | **accept, ignore** | flopy3 plumbing. Ignore with no warning, or with a warning if it was set to anything but the default. | +| `dev_*` options (36) | only with `developmode` | **break, clearly** | Raise an error saying "regenerate with `flopy4 mf6 sync --developmode`". MF6 is the main user, and its CI now has the switch. | +| nam `packages=`, IMS `rcloserecord`, TS records, other leftovers (33) | various | **case by case** | Classify in step 1 below. Most of them are likely to be "deprecate". | + +### Simulation and Model + +| flopy3 | flopy4 today | proposal | why | +|---|---|---|---| +| `MFSimulation(sim_name=, sim_ws=, exe_name=)` | `Simulation(name=, workspace=)` | **deprecate** `sim_name`/`sim_ws`; **accept** `exe_name` | `exe_name` is useful: store it as the default executable for `run()`, instead of only accepting `run(exe=)`. | +| `ModflowGwf(sim, modelname=, model_nam_file=)` | `Gwf(parent=sim, name=)` | **deprecate** `modelname` | | +| `MFSimulation.load(sim_ws=...)` | `Simulation.load(path_to_mfsim.nam)` | **accept a directory** | If the path is a directory, look for `mfsim.nam` in it. That's a good change regardless of migration. Keep `sim_ws=` as a deprecated keyword. | +| `write_simulation()` / `run_simulation()` | `write()` / `run()` returning `None` | **deprecate** both | `run_simulation` returns `(success, buff)` as in flopy3, and `run` keeps its own return type. | +| `get_package(name or type)`, `get_model(name)`, `get_package_list()` | `gwf["chd0"]` only | **accept, add to core** | They're useful lookups, not leftovers. `Flopy3Model` (`flopy4/mf6/adapters.py:186`) already has them, so move the logic into `ModelMethods` and `SimulationMethods`. | +| `gwf.modelgrid` | `gwf.grid` (`model_methods.py:11`) | **accept** `modelgrid` | It's the same object. Many plotting and export functions in flopy are written against `modelgrid`. | +| `sim.register_ims_package(ims, [names])` | `ims.models = [...]` | **deprecate** | 340 MF6 test files use it. The method is one line. | +| default file names: `m.chd`, then `m_0.chd`, ... | `m.chd0.chd`, `m.chd1.chd`; `s.ims0.ims` | **reconsider** | Leaving the first instance unsuffixed (`m.chd`, then `m.chd1.chd`) matches flopy3 in the common one-instance case, and it's tidier. Tests and scripts that reference files by name (observation and budget checks) then keep working. This changes file names in flopy4's own tests, so decide early. | + +### Data access and output: decide, don't alias + +These are the "break" group. Aliases here would hide a real change in +meaning. + +- **`npf.k.array` / `.get_data()` / `.set_data()`.** `flopy-integration-plan.md` + Decision 2 leans toward (a): `npf.k` is the value itself, and + `.array` raises with a hint. That group keeps the rule "same result or + a clear error". One cheap middle ground: `get_data()` with no + arguments on an array value could return the array, but numpy arrays + can't take new methods, so it would need a wrapper. Don't do this. +- **`gwf.output.head()`.** flopy4's `output.head` is a `DataArray` + property (`gwf_methods.py:31`); flopy3's is a method returning + `HeadFile`. Same name, different thing. Pick one before release + (`flopy-integration-plan.md` suggests options). If `head` stays a + property, calling it fails with `'DataArray' object is not callable`, + which is confusing. A wrapper that raises with a migration hint when + called would make that error clearer. + +### What the changes buy + +After Part 1, a typical MF6 autotest `build_models` (constructors, +`register_ims_package`, `write_simulation`) should run against flopy4 +with only `ModflowXxx` → `xxx.Xxx` class names changed. The class names +are the one difference that stays. Generated `ModflowXxx` aliases belong +to the flopy 4.x compat layer (`flopy-integration-plan.md`), not to +flopy4 itself, because they'd double the namespace. + +## Part 2: the codemod + +### Interface + +- Library: `flopy4.migrate.transform(source: str, *, path=None) -> Result`, + where `Result` holds `code: str` and `diagnostics: list[Diagnostic]` + (a line, a rule id, a message, and whether the code was rewritten or + only flagged). +- Script: `flopy4 migrate PATH... [--diff | --write] [--select RULES] + [--check]` through `flopy4/cli.py`. `--check` exits non-zero if + anything would change (for CI). The default is `--diff`. The script + never writes files unless asked. +- Built on libcst, so comments and formatting outside the rewritten + nodes stay as they were. libcst becomes an optional dependency + (a `migrate` extra). + +### Rules + +Each rule is one transformation. Order matters only where noted. + +1. **Imports.** `import flopy` and `flopy.mf6.X` → `from flopy4.mf6 import + gwf, gwt, ...` with the modules used. Leave every non-mf6 use of + `flopy` alone (`flopy.utils`, `flopy.plot`, `flopy.discretization`). +2. **Class names.** `ModflowGwfdis` → `gwf.Dis`, `MFSimulation` → + `Simulation`, `ModflowTdis` → `Tdis`, and so on. The table is + generated from the DFNs: flopy3's name is `Modflow` + component name + with the hyphen removed. Use module-qualified names, because `Dis` + and others repeat across model types. +3. **Parent keyword.** A positional first argument → `parent=`. Optional + once Part 1 accepts a positional parent. Keep it as a style rule, + off by default. +4. **Keyword renames** with no change to the value: from the generated + `Legacy` tables (`*_filerecord` → `*_file` where the value is a + literal string, `sim_ws` → `workspace`, `modelname` → `name`, ...). +5. **Keyword reshapes.** STO's `steady_state`/`transient`, OC's + `saverecord`/`printrecord`, filerecord tuples. Rewrite when every + value involved is a literal (a dict, list or tuple of constants). + Otherwise leave it alone and add an info diagnostic: the runtime + alias handles it. +6. **Method renames.** `write_simulation()` → `write()`; `run_simulation()` + → `run()`, but only when the return value is unused. If it is used + (`success, buff = sim.run_simulation()`), leave it and flag it. +7. **Flags with no rewrite**, for the "break" group: `.array`, + `.get_data`, `.set_data`, `output.head()` / `.budget()`, + `simulation_data`, `dev_*` keywords, `model_splitter`. These get a + `# TODO(flopy4): ...` comment linking to `docs/migrating.md`, or only + a diagnostic with `--no-comments`. + +### Knowing what a name refers to + +Rules 6 and 7 need to know that `sim` is a simulation and `gwf` a model. +libcst's scope provider plus a simple per-function pass covers the +common cases: track assignments from constructor calls (`gwf = +ModflowGwf(...)`) and from `MFSimulation.load(...)`, `sim.get_model(...)`. +Don't do cross-function type inference. When the type of a name is +unknown, match method names that only exist on one kind of object +(`write_simulation` is always a simulation), and flag the rest instead +of guessing. + +### Generated tables, one source of truth + +Extend `_legacy_kwargs` in `flopy4/mf6/utils/codegen/make.py` so that +`sync` writes one data module (say `flopy4/mf6/_legacy_table.py`) with, +for each component: flopy3's class name, keyword renames, and named +converters. Both `LegacyKwargs` (runtime) and the codemod read it. A test +checks that every `Legacy` entry has a codemod rule, or is marked +runtime-only on purpose. + +### Testing + +- **Unit:** each rule gets before/after source snippets, plus an + idempotence check: running the codemod on its own output changes + nothing. +- **Corpus:** run the codemod on each MF6 autotest's `build_models` + (local modflow6 checkout, not vendored), then run the result and + compare heads and budgets against the flopy3 build, using the existing + framework comparison. Report the share that run unchanged, run after + the codemod, and remain flagged. Use it as a scoreboard (xfail + manifest), not a gate, as with the compat tier in + `flopy-integration-plan.md`. +- **Examples:** the 76 modflow6-examples scripts as a second corpus. They + use more `.array` / `.get_data()`, so expect more flags. + +## Steps + +1. **Classify every difference.** Generate the full keyword diff (all + 1101 keywords across the 100 shared classes) and the method diff, and + label each accept / deprecate / break. Resolve the 33 "other" + keywords. Write the result into `docs/migrating.md` as the reference + table. *No code changes.* +2. **Decisions that block Part 1:** default file naming, `output.head` + (property or method), data access (Decision 2). Step 3 can start + without them, but its scoreboard numbers inform them. +3. **Bug fix:** bare-list `stress_period_data` (independent, do first). +4. **Part 1 runtime aliases**, in order of MF6 autotest usage: + positional parent, `register_ims_package`, `write_simulation` / + `run_simulation`, `Simulation.load` from a directory, `get_model` / + `get_package`, `modelgrid`, `sim_ws` / `sim_name` / `modelname` / + `exe_name`, `*_filerecord`, `maxbound`, `ts` / `obs` keywords. +5. **Generated table** shared by runtime and codemod. +6. **Codemod rules 1, 2, 4, 6** (mechanical renames), CLI and library + entry points, unit tests. +7. **Rules 5 and 7** (reshapes and flags), name tracking. +8. **Corpus scoreboard** over MF6 autotests, then modflow6-examples. + Runs in flopy4's CI against a modflow6 checkout; nothing changes in + MF6's repo. Doubles as readiness tracking for MF6's switch during the + 4.x release candidates (see "MF6's test suite"). +9. **Docs:** a "Migrating" page that leads with "your code probably + runs; here's what the warnings mean; `flopy4 migrate` fixes them". + +Steps 3–4 alone already make the MF6 test corpus mostly runnable against +flopy4. If that turns out to be enough for MF6's needs, the codemod +becomes a convenience for users, not a prerequisite. + +## Rough effort + +| step | size | +|---|---| +| 1. classification | 1–2 days, mostly generated | +| 3. bare-list fix | hours | +| 4. runtime aliases | ~1 week, mostly generated `Legacy` entries plus a handful of methods | +| 5–6. table and mechanical codemod | ~1 week | +| 7. reshapes, flags, name tracking | ~1 week | +| 8. corpus scoreboard | a few days, then ongoing tail work | + +## Open questions + +- Should "accept" aliases stay permanently, or warn eventually? This + plan says permanently for positional parent, `get_package` / + `get_model`, `modelgrid`, `exe_name` and loading from a directory; + they're good API, not leftovers. +- Should the codemod live in flopy4 or in flopy 4.x next to the compat + layer? In flopy4 for now; it moves with everything else at the merge. +- When a flopy4 change breaks MF6's optional `fp4` jobs, who fixes it, + and how fast? Still open. It matters more once MF6's build code moves + to flopy4. + +## MF6's test suite + +Settled 2026-11-03: + +- The round-trip PRs (#3048, #3057, #3067) are worth having for both + sides: flopy4 gets IO coverage across MF6's corpus, MF6 gets its NetCDF + input tested. Keep them as they are. +- MF6 will switch its tests' build code to flopy4 eventually: probably + while flopy 4.x is in release candidates, after MMP and before GA. Not + before then, because the API (including the Part 1 changes here) is + still moving. +- Until then, constructor coverage stays on flopy4's side: step 8 runs + the codemod over a modflow6 checkout in flopy4's CI, and MF6's + `build_models` stay flopy3. +- When MF6 switches, step 8's scoreboard shows how much of the suite the + codemod converts cleanly, and the conversion uses the codemod, not + hand edits. +- For teaching the new style, port a dozen modflow6-examples by hand + rather than adding flopy4 builds to the autotests. From 11560725714af38a4192302bf268c5216a0a7638 Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 12:10:52 -0800 Subject: [PATCH 09/77] plan MMP data-shape decisions from the layout/logical rule layout-logical-decisions.md states one rule from the #287 work (the object model holds input as given, logical forms are views, dynamic derivations are never materialised silently) and applies it to the open MMP issues it bears on: #229, #523, #345, #226 and #259, with the MF6 source checked for the UZF and SFR claims. It also proposes an Options section for the representation spec and separates round-trip from cross-engine equivalence. mmp-plan.md links it from Tracks B and C, notes that #226, #259 and binding-graph navigation share one attach operation, and adds the rule to the decisions table. Co-Authored-By: Claude Opus 5.5 --- docs/dev/mmp-plan.md | 13 +- layout-logical-decisions.md | 245 ++++++++++++++++++++++++++++++++++++ 2 files changed, 256 insertions(+), 2 deletions(-) create mode 100644 layout-logical-decisions.md diff --git a/docs/dev/mmp-plan.md b/docs/dev/mmp-plan.md index ae2efd6c..7e5f2d53 100644 --- a/docs/dev/mmp-plan.md +++ b/docs/dev/mmp-plan.md @@ -94,8 +94,8 @@ after MMP is under "After MMP". | track | items | by MMP | detail | |---|---|---|---| | **A. Engine correctness** | bare-list bug; equivalence harness (tier E) over the registry; round-trip tests; fix what they find | everything | `flopy-integration-plan.md` "Tests", `docs/dev/test.md`. **Longest and least predictable. Start first.** | -| **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open" | -| **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md` | +| **B. API decisions** | `output.head` (integration decision 3); `get_package`/`modelgrid` on Model; `name` vs `sim_name`; accessor naming and `flatten()`/`node`; component naming (#258); parent/child init (#259); unified output loading (#342); the data-shape questions, decided together by one rule: STO types (#229), UZF placeholders (#523), SFR `connectiondata` (#345), packages without a grid (#226) | decided and implemented | `flopy-integration-plan.md` "Open decisions"; `xarray-package-plan.md` "Open"; `layout-logical-decisions.md` for #226, #229, #259, #345, #523 | +| **C. Data and xarray** | `to-xarray-constants` fix; datetime period keys (#518, #226); lossless `to_xarray`/`from_xarray` and `dense()` (export phase 1), including an Options section in the representation spec (logical options as data, layout options as encoding, output options off the data) and opt-in views for dynamic derivations (`AUXMULTNAME`, TAS); deferred external arrays (#516) | the fix, #518 and export phase 1; #516 for FUNC-8 | `xarray-export-plan.md`, `datetime-periods-plan.md`, `docs/dev/xarray-representation.md`, `layout-logical-decisions.md` "Track C" | | **D. Docs** | API reference; concepts (#217); 3.x examples adapted; `migrating.md` for compat; SDD update once export phase 1 lands | all | `map.md`, `srs.md`, `sdd.md` | | **E. Migration** | classify every flopy3 difference (accept, deprecate, break); runtime aliases in flopy4 itself (positional parent, `register_ims_package`, `write_simulation`, `get_package`, `modelgrid`, `*_filerecord`, ...); the shared `Legacy` table; the `flopy4 migrate` codemod; the corpus scoreboard over MF6 autotests and examples. Generated `ModflowXxx` class names stay in flopy 4.x's compat layer | classification, aliases and scoreboard; the codemod if time allows | `flopy3-migration-plan.md`; `flopy-integration-plan.md` "The compat layer" | | **F. 3.x prep** (flopy `develop`) | 3.x items 1 to 5: utilities out of `flopy/mf6`, the shared-code protocol, upstream patches, deprecations, forward-compatible keywords | nothing required; items 1 and 2 by the `4.x` cut | `flopy-integration-plan.md` "3.x work" | @@ -110,6 +110,14 @@ release candidates follow MMP. - Track A's harness finds engine bugs, and fixing them may touch Track C (period data, arrays). Keep Track C's changes small until the harness is running. +- #226 (packages without a grid), #259 (parent/child init) and #287's + binding-graph navigation (`Spc.target`, `Model.flow_model`, ...) need + the same attach operation. Design them together, or the relation ends + up held in two places. +- Track A's two comparisons check different things: round-trip + (FUNC-7) preserves layout as given, while the flopy3/flopy4 file + comparison is logical, normalized by the option labels + (`layout-logical-decisions.md` "Track A"). - Track E's aliases and codemod rules wait for Track B's naming decisions (and default file naming), or they get regenerated. - Track D's API reference waits for Track B, and the SDD update waits @@ -143,6 +151,7 @@ From the plans this doc links to, ordered by when they're needed: | Pre-release version numbers (`4.0.0bN`?) | first alpha | here | | `output.head`: which side renames (integration 3) | MMP | `flopy-integration-plan.md` | | Remaining Track B API questions | MMP | Track B | +| The layout/logical rule, and with it #226, #229, #259, #345, #523 | MMP; early, since Track B's other data questions follow from it | `layout-logical-decisions.md` | | Default file names (`m.chd` vs `m.chd0.chd`) | the Part 1 aliases; decide early, it changes flopy4's own tests | `flopy3-migration-plan.md` | | FUNC-10 test matrix; FUNC-11 in or out of MMP | MMP | SRS | | Compat data access (integration 2) | `4.x` cut | `flopy-integration-plan.md` | diff --git a/layout-logical-decisions.md b/layout-logical-decisions.md new file mode 100644 index 00000000..6919c51f --- /dev/null +++ b/layout-logical-decisions.md @@ -0,0 +1,245 @@ +# Layout vs. logical: a rule for the MMP data questions + +Written 2026-11-03 against `develop` (`8d25e4b7`). Status: **proposal.** +Several open MMP issues are about the same thing: how far flopy4's object +model should mirror MF6's file format, and where it should present the +underlying data instead. This doc states one rule for that, from the #287 +work, and applies it to each issue. It feeds Track B and Track C of +`docs/dev/mmp-plan.md`. + +The research behind the rule is in the working notes for discussion #287 +(`option-classification-results.md`, `constraint-audit-results.md`, +`discussion-287-directions.md`; not on this branch yet). The parts this +doc needs are summarized below. + +## The rule + +### Three levels + +MF6 input can be described at three levels: + +- **Layout**: how the file writes things down. Blocks, keywords, arrays + vs. lists, `CONSTANT`/`INTERNAL`, `LAYERED`, unit factors. +- **Logical**: what the data is. Tables, columns, keys, and the settings + that change what the model computes. +- **Runtime**: MF6's memory. Its own layout, with solver artefacts. + +### Option labels + +Each of the 224 option names in the DFNs got a label by hand: + +| label | count | examples | +|---|---|---| +| logical | 121 | `MOVER`, `SIMULATE_ET`, `AUXMULTNAME`, `SURF_RATE_SPECIFIED` | +| output | 82 | `PRINT_INPUT`, `SAVE_FLOWS`, `*_FILEOUT` | +| layout | 20 | `READASARRAYS`, `STORAGECOEFFICIENT`, `K22OVERK`, the unit factors | +| mixed | 1 | CSUB `HEAD_BASED` | + +The split is clean: no layout option is triggered by, or triggers, a +logical constraint. The only crossings are conditions on where a layout +conversion is defined, e.g. RCHA can't express two rates in one column. + +### Derivation + +Some options declare a rule, "this quantity is computed from those". Each +such option is either: + +- **static**: every input is fixed once written, so storing the result + loses nothing (`STORAGECOEFFICIENT`, `K22OVERK`, the unit factors); +- **dynamic**: an input can change during the run (a time series, TVK, + model state), and MF6 re-evaluates the rule each time it's used + (`AUXMULTNAME`, `AUXDEPTHNAME`, TS/TAS, `LOCAL_Z`). + +No layout option is dynamic. 14 of the 20 layout options are static +derivations. + +### What it means for flopy4 + +1. **The object model holds input as given.** Layout choices stay in it: + `Rcha` is its own class, `STORAGECOEFFICIENT` is an option, a constant + stays a constant. This is what round-trip fidelity (FUNC-7) needs, and + what flopy3 users expect. +2. **Logical forms are views.** Dense arrays, normalized tables, decoded + values and conversions between layouts are computed from what's held, + on request. They don't replace it. +3. **Dynamic derivations are never materialised silently.** A view that + multiplies out `AUXMULTNAME` or evaluates a TAS says so in its name or + arguments. Materialising one changes behaviour if an input later + changes. +4. **Constraints follow their level.** A column that only means something + under a logical option is absent without it, in the object model too. + A column the file format requires positionally is the writer's + problem. + +## The issues + +### #229: split STO by storage coefficient vs. specific storage? + +Deltares suggests (a) two packages, or (b) separate `sc`/`ss` parameters. + +- **Rule:** `STORAGECOEFFICIENT` is a layout option, a static derivation. + The `ss` array holds SC instead of Ss, and MF6 converts with the cell + thickness, which is fixed (gwf-sto: the keyword "indicates that the SS + array is read as storage coefficient rather than specific storage"). + One quantity, two encodings. +- **Decision:** don't split the package (a). Two package types would + turn an encoding into a type distinction, and every consumer of STO + would have to handle both. Option (b) is fine as constructor sugar: + `Sto(sc=...)` sets `ss` and `storagecoefficient=True`. Passing both is + an error. +- **Migration:** accept. flopy3's form (`ss=` plus the flag) is unchanged. +- **MF6 side:** the issue's comment suggests raising it with MF6. There's + no logical reason to split there either. + +### #523: UZF placeholder values + +UZF's period rows always have all seven value columns, but `pet`, +`extdp` and `extwc` only mean something with `SIMULATE_ET`, and `ha`, +`hroot`, `rootact` only with `UNSAT_ETAE`. Users write placeholders. + +- **Verified in MF6** (`gwf-uzf.f90`): `SIMULATE_ET` sets `ietflag = 1` + (line 487) and `UNSAT_ETAE` sets it to 2 (line 529). `uzf_rp` uses the + ET columns only when `ietflag /= 0`, and the ETAE ones only when + `ietflag == 2` (lines 691–693, 712–719, 764–768). +- **Rule:** both options are logical, so the columns exist only when the + option is set. The fixed table is a layout. The keystring rows of the + other advanced packages are another layout of the same relation: + settings per feature per period, carried forward. +- **Decision:** make the six columns depend on their options, using the + conditional fields from #473. Users don't pass them without the + option, and get an error if they do. The writer writes a filler where + the format needs one: the ET columns sit in the middle of the row, so + without `SIMULATE_ET` the writer still has to put something there. + Loading drops whatever filler the file has. No keystring redesign is + needed, in flopy4 or MF6. +- **Migration:** deprecate passing the placeholders. Warn, and ignore + them. +- **Check first:** whether the trailing `ha`/`hroot`/`rootact` can be + left off the line when `UNSAT_ETAE` is off, or whether MF6's reader + needs them. + +### #345: normalize SFR `connectiondata`? + +SFR writes one row per reach, with an inline array `ic` of length +`ncon(ifno)` from `packagedata`. LAK, MAW and UZF write one row per +connection. The issue suggests changing MF6's format to match. + +- **Verified in MF6:** the sign of `ic` encodes the direction (DFN: a + negative value means the connected reach joins at the downstream end). + Every connection has to be listed from both ends, and MF6 checks it + (`sfr_check_connections`, `gwf-sfr.f90:4416–4434`: "Reach n is + connected to reach m but reach m is not connected to reach n"). +- **Rule:** logically this is an edge table: one row per + upstream→downstream pair. The file stores each edge twice, with a sign + for direction, and `ncon` repeats a count of the rows. Both are layout + redundancy that MF6 has to check. +- **Decision:** leave MF6's format alone. flopy4 holds `connectiondata` + as given and offers a view: + - a long table, one row per (reach, connection), like LAK's; and + - optionally an edge table, one row per pair, with `ncon` and the + reciprocity check derived from it. + + Whether a user can *build* SFR from the edge table, with flopy4 writing + both directions and `ncon`, is a later, additive feature. +- **Migration:** none. The given form is unchanged. + +### #226: packages without grid or time + +`Ic(strt=0.0)` without a model now works, and stays a scalar (checked on +`develop`; the traceback in the issue is from xattree). What's left is +whether and when to expand it. + +- **Rule:** the grid is a parameter supplied by the binding: a package + gets its dims from the model it's attached to. A constant is a valid + value as given, and `CONSTANT` is how the file encodes it. The issue's + comment asks whether the user meant "write `CONSTANT`" or "my array is + constant". Under the rule both readings hold: the value is held as a + constant, and the writer picks `CONSTANT` because that's what's held. +- **Decision:** + - Keep the value as given. Don't expand on attach. + - When the package is bound to a model with a grid, give it a view with + the grid's shape: the same expansion `dense()` does, named + consistently with it (Track C). + - Check shapes on attach, so an array that doesn't fit the grid fails + there, not at write. + - Leave `WriteContext` for overriding the encoding. +- **Shares machinery with #259 and #287 step 4** (navigation over the + binding graph). + +### #259: parent/child initialization + +Top-down (`Dis(parent=gwf)`) and bottom-up (`Gwf(dis=dis)`) are both +supported, and the issue leans towards encouraging bottom-up. + +- **Rule:** the parent/child relation is a binding row, as in the name + file. Both styles create the same row, so neither is more correct. + What matters is that the relation is held once, and both sides are + views of it. +- **Decision:** keep both styles. Implement attach (`gwf.dis = ...`, + `Dis(parent=gwf)`, `Gwf(dis=...)`) as one operation over the binding + graph that updates both sides and runs #226's shape check. Design it + together with #287 step 4 (`Spc.target`, `Mvr.models`, `Obs.target`, + `Model.flow_model`), which needs the same graph. +- **Migration:** accept. flopy3's top-down style keeps working. + +### Related: #240, #243 + +Fill values and time representation are representation questions. +Settle them in export phase 1 (Track C), below. + +## Track C: the representation + +`docs/dev/xarray-representation.md` already follows the rule for period +data: stored as given, `dense()` a view, TS/TAS values kept as named +references. It doesn't yet say how options appear. Add an Options section +in export phase 1: + +| option label | in the representation | +|---|---| +| logical | data or attrs on the package node | +| layout | an encoding record, not data. xarray's `encoding` is the natural home, and the static derivations are the same kind of thing as CF's `scale_factor`/`add_offset`. But xarray drops `encoding` on most operations, so it may need to be attrs with a reserved prefix. Decide in phase 1. | +| output | not on the package's data. A separate node, or left out. | +| dynamic derivation | the rule, not its result: the aux variable carries its role (e.g. a multiplier). `dense()` doesn't apply it unless asked. | + +`tas-interpolation-plan.md`'s step view is a dynamic-derivation view +under this rule, so it should be opt-in, and `from_xarray` should never +read its output back as input. + +This is time-sensitive. The representation reaches testers at MMP, and +the xarray package plan's extraction criterion is a spec that has been +stable for a release or two. + +## Track A: two kinds of equivalence + +- **Round-trip** (FUNC-7, load → write → load): layout-preserving. A + round trip that turns `READASARRAYS` into list input, or a `CONSTANT` + into a full array, is a bug even if MF6 gives the same heads. +- **Cross-engine** (tier E's comparison of files written by flopy3 and + flopy4): logical. Differences in layout and output options are + expected. The option labels are the normalizer's rule table, rather + than a hand-tuned diff filter. + +## Not before MMP + +The rest of the #287 work is off the gate. None of it changes what +testers write: + +- the DDL slice and parameter syntax (#287 steps 2–3); +- conversions between layouts (RCH ↔ RCHA by densify and sparsify, step + 5), which are additive; +- logical annotations in the DFN schema, in devtools 2.0 if ready before + the release candidates, otherwise in a later minor version; +- `source`/derivation metadata for modflowapi's memory catalog; +- the MF6 NetCDF format conversation. + +## Open + +- #523: whether MF6 needs the trailing ETAE columns when `UNSAT_ETAE` is + off. +- #345: whether the edge view is worth having before MMP, or only the + long table. +- #226: the name of the expanded view, decided with the accessor naming + in Track B. +- Track C: `encoding` vs. prefixed attrs for layout options. +- Whether to post #287 before these decisions, so the Deltares + developers who raised #226, #229 and #259 can weigh in on the rule. From 5c9de7431647b143b77560691913d3dac9b1ff8e Mon Sep 17 00:00:00 2001 From: wpbonelli Date: Tue, 3 Nov 2026 12:11:54 -0800 Subject: [PATCH 10/77] add the #287 working notes, classification and constraint audit The research behind layout-logical-decisions.md: the #287 directions and comment draft, the relational spec draft, the polymorphism notes, the option classification and constraint audit plans and results, and the scripts and data they came from (option-classification/, constraint-audit/). Co-Authored-By: Claude Opus 5.5 --- constraint-audit-plan.md | 179 ++ constraint-audit-results.md | 350 +++ constraint-audit/dfn-constraints.csv | 2246 ++++++++++++++++++++ constraint-audit/dfn_pass.py | 167 ++ constraint-audit/mf6-checks.csv | 1280 +++++++++++ constraint-audit/mf6-sample.csv | 167 ++ constraint-audit/mf6_sites.py | 184 ++ constraint-audit/sample.py | 249 +++ discussion-287-comment.md | 179 ++ discussion-287-directions.md | 355 ++++ discussion-relational-spec-draft.md | 49 + layout-logical-decisions.md | 5 +- option-classification-plan.md | 243 +++ option-classification-results.md | 241 +++ option-classification/inventory.py | 138 ++ option-classification/label.py | 260 +++ option-classification/options-labelled.csv | 225 ++ option-classification/options-labelled.md | 246 +++ option-classification/options.csv | 225 ++ option-classification/options.md | 230 ++ schema-polymorphism-notes.md | 119 ++ 21 files changed, 7335 insertions(+), 2 deletions(-) create mode 100644 constraint-audit-plan.md create mode 100644 constraint-audit-results.md create mode 100644 constraint-audit/dfn-constraints.csv create mode 100644 constraint-audit/dfn_pass.py create mode 100644 constraint-audit/mf6-checks.csv create mode 100644 constraint-audit/mf6-sample.csv create mode 100644 constraint-audit/mf6_sites.py create mode 100644 constraint-audit/sample.py create mode 100644 discussion-287-comment.md create mode 100644 discussion-287-directions.md create mode 100644 discussion-relational-spec-draft.md create mode 100644 option-classification-plan.md create mode 100644 option-classification-results.md create mode 100644 option-classification/inventory.py create mode 100644 option-classification/label.py create mode 100644 option-classification/options-labelled.csv create mode 100644 option-classification/options-labelled.md create mode 100644 option-classification/options.csv create mode 100644 option-classification/options.md create mode 100644 schema-polymorphism-notes.md diff --git a/constraint-audit-plan.md b/constraint-audit-plan.md new file mode 100644 index 00000000..b66870a5 --- /dev/null +++ b/constraint-audit-plan.md @@ -0,0 +1,179 @@ +# Constraint audit: plan + +Working notes, 2026-11-03. Follows `option-classification-results.md`. Serves as +the requirements pass for step 2 (DDL slice) in `discussion-287-directions.md`. + +## Goal + +List the kinds of constraint a valid MF6 simulation must satisfy, and for each +kind record: + +- **where it lives:** the DFN, MF6 Fortran, or both; +- **which level it belongs to:** layout, logical, or runtime; +- **whether the DFN can express it today**, and whether a DDL-style logical + spec could. + +This is a taxonomy with coverage numbers, not a full list of every check. +Instances are sampled until new ones stop producing new kinds. + +## Why + +- **The spec is thin on constraints.** dev3 snapshots have 50 `when` and 32 + `valid` keys. MF6 has 1,594 `call store_error` sites in 174 files (rough + count; it includes `store_error_filename` and repeats). Most of the input + contract exists only in Fortran. +- **It tests the layering a second time.** Each constraint should belong to + one level. A constraint that spans levels is the constraint-level + counterpart of a mixed option, e.g. a `when` that tests an encoding + option. +- **It sets the requirements for step 2.** The kinds map onto DDL constructs + (CHECK, NOT NULL, PRIMARY KEY, FOREIGN KEY, multi-table assertions). What + doesn't map is the interesting result. +- **It tests the `when` = binding hunch.** If conditional presence and + component co-occurrence have the same form, `when` and binding parameters + may be one construct with two sources (component options vs. context). +- **Practical uses:** a source for generated flopy4 validation, and a + concrete DFN gap list for MF6 developers. + +## Taxonomy (provisional, to revise from the evidence) + +Ordered roughly from local to global. + +| # | Kind | Examples | +|---|---|---| +| K1 | lexical / format | keywords, record syntax, block order, file-reference forms, `OPEN/CLOSE` | +| K2 | type and domain | enums, `K > 0`, `0 ≤ SY ≤ 1`, integer vs. real | +| K3 | shape and dimension | arrays sized by grid dims, list rows ≤ `MAXBOUND`, `NCVERT` = vertex count | +| K4 | conditional presence | aux columns = `AUXILIARY` names; boundname column only with `BOUNDNAMES`; `when` | +| K5 | option coherence | options that exclude or require each other | +| K6 | keys and uniqueness | unique package names; duplicate cellids allowed or not (varies by package) | +| K7 | referential | cellid in grid and `IDOMAIN > 0`; boundnames, TS names, MVR packages, exchange models exist | +| K8 | co-occurrence and cardinality | one DIS per model; GWF needs NPF and IC; TVK needs NPF; GWT needs FMI or a GWF-GWT exchange; MVR needs `MOVER`; each model in one solution | +| K9 | temporal | period numbers ascending and ≤ `NPER`; TS times cover the run; ATS periods valid | +| K10 | structural / physical | `TOP > BOT`; DISU `JA` symmetric, `IHC` consistent; SFR/LAK connectivity; vertex geometry | + +Expect K10 to split: some checks are model content (`TOP > BOT`), some are +numerics (thin-cell warnings), and some guard MF6's own assumptions. + +## Attributes recorded per kind (and per sampled instance) + +- **level:** layout / logical / runtime-only. "Runtime-only" means it + depends on simulated state or numerics, not on input alone. +- **arity:** one field, one row, one table, one component, several + components, the whole simulation. +- **config/data:** whether it constrains configuration (options, + dimensions), data (arrays, rows), or relates the two (e.g. `MAXBOUND` + vs. row count). This is the second axis from the earlier discussion. +- **trigger:** unconditional, or conditional on an option, a dimension, or + another component's presence. +- **where:** DFN key (`shape`, `when`, `valid`, `optional`, file + `component`, `[observations]`), MF6 file:line, mf6io prose, or a + combination. +- **DFN-expressible today:** yes / partly / no. +- **DDL-expressible:** which construct, or "no" with a reason. +- **severity:** error vs. warning (MF6 has `store_warning` too). +- **checked when:** IDM load, `df`/`ar`, `rp` (per period), or `ad` (per + step). Late checks suggest constraints that depend on data not available + at load. + +## Sources + +1. **DFN snapshots** (devtools `autotest/dfns/__snapshots__/v2.0.0.dev3/*.toml`): + `shape`, `when`, `valid`, `optional`, `default`, file `component`, + `[observations]` arms, and the block structure itself. +2. **MF6 source** (`~/dev/modflow6`, record the commit): `store_error`, + `store_warning` and `ustop` sites, plus the error messages around them. + Generic IDM checks (`src/Utilities/Idm/`) are mostly K1–K3 and apply to + everything. Package checks are where K4–K10 live. +3. **mf6io docs** (`doc/mf6io`): prose rules ("must", "only if", "cannot") + that neither of the above encodes. Used to catch rules that are enforced + nowhere. +4. **The option labels** (`option-classification/options-labelled.csv`): + the `schema` and `binding` tags should line up with K4 and K7/K8 + triggers. + +## Passes + +### Pass 1: DFN, mechanical + +Script over the snapshots. One row per constraint-bearing key. Columns: +component, block, field, key, value, and inferred kind (shape → K3, +`when` → K4, `valid` → K2, file `component` → K8, ...). Then tally by kind +and component. Also list each `when` condition and what it tests: a `schema` +option, an encoding option, or something else. That's the first layering +check. + +Output: `constraint-audit/dfn-constraints.csv` and a short summary. + +### Pass 2: MF6 checks, sampled + +1. Extract all `store_error`/`store_warning` sites with file, line, + enclosing routine and the nearest message literal. Messages are usually + built a few lines above the call with `write(errmsg, ...)`. Script it + and accept some misses. +2. Cluster by normalised message ("out of range", "not found", "must be + greater than", "duplicate", ...). Counts per cluster show where the bulk + is. +3. Read by component family: all generic IDM checks, then one or two + packages per family (GWF: DIS/DISU, NPF, SFR, MVR; GWT: SSM, FMI; PRT: + PRP; SWF/CHF; exchanges; IMS; TDIS/ATS). Label each read instance with + a kind and the attributes above. +4. Stop a family when three packages in a row add no new kind or attribute + combination. Extend if one does. + +Output: `constraint-audit/mf6-checks.csv` (all sites, clustered) and +`constraint-audit/mf6-sample.csv` (labelled sample). + +### Pass 3: docs, targeted + +Grep mf6io for "must", "only", "cannot", "required", "not allowed". For +families read in pass 2, check each rule against the DFN and Fortran rows. +Record rules found only in prose. + +### Pass 4: classify and cross-check + +Merge into one table of kinds with counts, levels, coverage, and DDL +mapping. Run the checks below. + +## Checks and expected findings + +Fixed before looking, so the results can surprise us: + +- **Layering:** K1 is layout. K2–K9 are logical. K10 is mostly logical, + some runtime. A logical-kind constraint whose trigger is a layout option + is a leak. So is a layout check that depends on another component's + data, except for the cross-component and computed lenses we already know + about. +- **`when` vs. binding:** K4 triggers should be `schema` options. K8 + triggers should be `binding` options or context. If K4 and K8 instances + can be written in one form, record that as evidence for the unified + construct. +- **Coverage:** most of K1–K3 should be in the DFN, and most of K5–K10 + should not. Report the fraction per kind. +- **DDL fit:** K2 → CHECK, K4 → NOT NULL/conditional, K6 → PRIMARY KEY, + K7 → FOREIGN KEY, K8 → multi-table assertions or binding rules, K3 → + needs dimensions as first-class values. K9 and K10 are the likeliest + misfits. + +## Scope + +- **In:** input constraints for all model types, exchanges, solutions and + utilities in dev3. +- **Out:** convergence and other runtime numerics failures, I/O errors + (file not found, unit conflicts), internal assertions that guard + programming errors rather than input. Count these in a "not input" + cluster so the totals reconcile, but don't classify them. +- **No edits** to devtools, modflow6 or modflowapi. Gaps go in a list. + Commit nothing. Post nothing to GitHub. + +## Done when + +- the kinds table has counts, level, DFN coverage and DDL mapping for each + kind; +- the `when` triggers are classified, with a verdict on `when` vs. binding; +- the layering check has a written result, listing any leaks; +- there's a gap list: rules that are only in Fortran, only in prose, or + only in the DFN; +- findings to carry into step 2 are listed; +- `discussion-287-directions.md` gets a one-line pointer. Check for the vim + swap file first. diff --git a/constraint-audit-results.md b/constraint-audit-results.md new file mode 100644 index 00000000..413a44af --- /dev/null +++ b/constraint-audit-results.md @@ -0,0 +1,350 @@ +# Constraint audit: results + +Working notes, 2026-11-03. Carries out `constraint-audit-plan.md`. MF6 at +`8aad7bc4a5`, DFNs from the devtools dev3 snapshots. + +Data and scripts are in `constraint-audit/` (untracked): + +| File | What | +|---|---| +| `dfn_pass.py` → `dfn-constraints.csv` | pass 1: 2,245 constraint-bearing DFN keys | +| `mf6_sites.py` → `mf6-checks.csv` | pass 2: all 1,279 `store_error`/`store_warning` sites with message and regex cluster | +| `sample.py` → `mf6-sample.csv` | passes 2–3: 166 hand-labelled instances (159 MF6 sites, 7 rules found only in mf6io prose) | + +## Summary + +- **The layering holds for constraints too.** + - Every constraint triggered by a layout option is either a layout + constraint or a condition on where a partial lens is defined. + - There are no exceptions: the one apparent case (`AUXMULTNAME`) turned + out to be a logical option once the derivation axis was added. + - Logical constraints are triggered by logical options, dimensions, the + presence of other components, or the binding. +- **`when` and binding are one construct, with a wider scope than `when` + has today.** + - Conditional presence (K4) and co-occurrence (K8) have the same form: + *predicate over context ⇒ element present or absent*. + - What differs is where the predicate's variables live: the component's + own options and dimensions, sibling components, or the binding row. + - Three cases cross between the two: + - `MODELNAMES`, an option MF6 requires to equal the binding context; + - `FLOW_PACKAGE_NAME`, an option that overrides a default supplied by + the binding; + - `MOVER`, an option that gates a role other components refer to. + - K8 also has a second half that isn't presence: consistency between + bound components (same grid, same solution, mirrored exchange tables). + Those are join assertions. +- **Most of the input contract is in Fortran only.** + - Of the sample, the DFN expresses 20 instances fully and 26 partly, out + of 166. + - It covers grammar, shapes, required fields, and keys and references + within one file. + - It covers almost none of the following: + - numeric domains (K2: 1 of 27); + - option coherence (K5: 0 of 14); + - co-occurrence (K8: 1 of 19); + - temporal rules (K9: 0 of 7); + - structural rules (K10: 0 of 18). +- **A DDL covers most of what's logical.** + - CHECK, FK, PK/UNIQUE and NOT NULL take 94 of the 166. + - Multi-table assertions take 39. Standard SQL has `CREATE ASSERTION`, + but no engine implements it, so in practice these are queries. + - 31 have no DDL form: + - 19 of those are layout, as they should be; + - 7 are run-state, numerics, or external-file and build checks; + - the 5 genuine logical misfits are monotone row order (2), graph + acyclicity (1) and geometric predicates (2). + +## The kinds + +Counts are from the hand-labelled sample. I picked the sample for coverage, +so the proportions describe the sample, not MF6 as a whole. + +DFN column: expressible in the DFN today (yes / partly / no). DDL column: +the construct a logical spec would use. + +| Kind | n | Level | DFN y/p/n | Typical arity | Config/data | DDL | +|---|---|---|---|---|---|---| +| K1 lexical / format | 10 | layout | 3/1/6 | row, field | data | none (grammar) | +| L lens domain (new) | 6 | layout | 0/0/6 | component, multi | config | none (encoding domain) | +| K2 type and domain | 27 | logical | 1/3/23 | field (22), row (4) | data 18, config 9 | CHECK | +| K3 shape and dimension | 10 | logical | 2/5/3 | table | **cfg-data 7** | CHECK(count = dim) | +| K4 conditional presence | 19 | logical (1 layout) | 5/1/13 | component | data | NOT NULL / conditional | +| K5 option coherence | 14 | logical 9, layout 4, output 1 | 0/2/12 | component | **config 13** | CHECK on the options row | +| K6 keys and uniqueness | 12 | logical | 3/1/8 | table, multi | data | PK / UNIQUE | +| K7 referential | 21 | logical | 5/12/4 | multi (11) | data 13, config 7 | FK | +| K8 co-occurrence | 19 | logical | 1/0/18 | multi (18) | **config 16** | assertion | +| K9 temporal | 7 | logical | 0/1/6 | multi, field | data | FK(period), ordering | +| K10 structural / physical | 18 | logical | 0/0/18 | multi (9), row | **data 18** | assertion (join on grid) | +| not input | 3 | runtime | — | — | — | — | + +Revisions to the provisional taxonomy: + +- **New kind L, lens domain.** These constraints say where a layout's + partial lens is defined: + - `READASARRAYS` excludes `SURF_RATE_SPECIFIED` and `NSEG > 1` + (`gwf-evt.f90:202, 291`); + - TAS isn't available for DISU (`BoundaryPackage.f90:1252`); + - EVTA requires a non-DISU grid (prose); + - layered NetCDF isn't supported for DIS int2d; + - binary list input isn't supported for text columns. + + They look like K5 option coherence, but they're the partiality conditions + of section 1 of the directions doc, which MF6 enforces. They belong with + the lens, not with the logical constraints. Step 5's conversion code + should report exactly these. +- **K8 splits in two.** + - **K8a, co-occurrence:** if A is present, B must be (or must not be). + MVR needs MVT, LKT needs LAK, CSUB requires `SS = 0` in STO, XT3D + excludes GNC, and OLF needs DFW. + - **K8b, consistency between bound components:** + - the GWF and GWT grids are identical (`exg-gwfgwt.f90:231`); + - exchanged models share a solution; + - GWT-GWT exchange data mirrors GWF-GWF (`exg-gwfgwt.f90:447`); + - the number of exchanges equals the number of GNCs. + + K8a has the `when` form. K8b doesn't: those are equality joins. +- **K10 is mostly joins against the grid**, not a separate physics layer. + Examples: + - `TOP > BOT`; + - SURFDEP ≤ cell thickness; + - lake `belev` ≥ cell bottom; + - UZF area ≤ cell area; + - release point in cell; + - MAW screen vs. cell. + + The remainder are graph properties of one table: SFR reciprocity, upstream + fractions summing to 1, acyclicity. The grid is the table every + structural constraint joins against, which fits "grid as parameter" + (directions doc §3). Only 1 of 18 is runtime (time step too small). +- **K1 splits into grammar, which the DFN covers, and ordering conventions, + which it doesn't.** The ordering conventions: + - period numbers increasing; + - solution-group blocks consecutive; + - ICELL2D in sequence; + - CXS rows grouped by section; + - DISU `JA` with self first. + + All of these are layout. + +## Layering check + +Constraints sorted by what triggers them (sample): + +| Trigger | Layout constraints | Logical constraints | +|---|---|---| +| unconditional | 11 | 64 | +| a dimension | — | 17 | +| a logical option | — | 17 | +| a data discriminator in the row (`obstype`, `ctype`, `condeqn`) | — | 4 | +| another component's presence | — | 12 | +| the binding | 4 (lens domain) | 21 | +| a layout option | 7 | **0** | +| an output option | — | 2 | + +- **Layout option → logical constraint: none.** + - The one candidate was `BoundaryPackage.f90:1331`: `AUXMULTNAME` must + name an aux column. + - `AUXMULTNAME` has since been relabelled logical. It declares a dynamic + derivation (`option-classification-results.md`, "The derivation axis"). + - The constraint was never evidence either way: any option that names a + column carries such a reference. +- **Layout constraints triggered by layout options:** + - `K22OVERK` requires K22; + - `FLOW_REDUCTION_LENGTH` requires `AUTO_FLOW_REDUCE`; + - `UNIT_CONVERSION` excludes `LENGTH_`/`TIME_CONVERSION`; + - `MODELNAMES` ⇔ exchange binding; + - the lens-domain cases. + + These are coherence rules within the layout layer, as expected. +- **Output options with logical prerequisites: a small new finding.** + - DFW `SAVE_VELOCITY` requires `ANGLDEGX` in the discretization + (`swf-dfw.f90:1350`). + - DISU's `VERTICAL_OFFSET_TOLERANCE` parameterizes a K10 check. + + A run directive can't change results, but it can need logical data to + carry out. That's harmless for the split. Note it for the output/"run + directive" category: some directives have preconditions. + +Verdict: **the layering holds at the constraint level.** Each constraint +belongs to one level. The only constraints that look like they cross +levels are lens-domain conditions, and those belong to the lens. + +## `when` vs. binding + +The plan's hypothesis was that K4 is triggered by schema options and K8 by +binding options or context, and that both can be written in one form. + +**What the DFN uses `when` for today:** 50 uses, testing only 3 options. + +| Option | Uses | Label | +|---|---|---| +| `boundnames` | 46 | logical/schema | +| `modelnames` | 3 | layout | +| `surf_rate_specified` | 1 | logical/schema | + +**What MF6 actually conditions on, across K4 and K8a:** + +| Predicate over | Examples | Expressible as `when` today? | +|---|---|---| +| own options | sorption → `BULK_DENSITY`, `DISTCOEF`, `SP2`; decay → `DECAY`, `DECAY_SORBED`; `NON_VERTICAL_WELLS` → ANGLEDATA block; `BOUNDNAMES` → boundname | yes for fields; there's no block-level `when` | +| own dimensions | `NTABLES > 0` → TABLES block; `NRELEASETIMES > 0` → RELEASETIMES block | no (but these are K3 counts: `count(rows) = dim`) | +| own data, same row | `obstype` → ID2 required / boundname forbidden; `ctype` → `connlen > 0`; `cprior` → divflow range | union arms, already modelled as discriminated rows | +| other fields present | ANGLE2 → ANGLE1; K22 → ANGLDEGX in DIS | no | +| sibling components | GWF MVR active → MVT; CSUB → STO `SS = 0`; XT3D → no GNC; LKT → LAK | no | +| binding row | exchange-bound MVR → `MODELNAMES`; SSM `pname` → flow package; SFT name defaults to SFR name | no | + +The form is the same throughout: `when = `, with the predicate's +names resolved in a scope chain: + +1. the component's own options and dimensions; +2. the model's components; +3. the binding's parameters. + +This fits H3 (directions doc §3): a binding parameter is a name in the outer +scope. `MODELNAMES` shows it most clearly. The option exists only so the +file can state its binding context, and MF6 rejects it whenever it +disagrees with the binding (`gwf-mvr.f90:785-798`). With a binding +parameter, the option goes away: the mover's `mname` column is filled from +the binding when it isn't written. + +**Verdict: supported, for K4 and K8a.** K8b (consistency between bound +components) doesn't have this form and needs join assertions. For step 3, +`[params]` should give binding parameters names that `when` can test, +rather than adding a second conditional construct. + +## Coverage + +DFN constraint keys (pass 1, all 146 components): + +| Kind | DFN keys | Rows | +|---|---|---| +| K1 | `index`, `netcdf`, `layered`, `case_sensitive`, `write_if_empty` (+ the grammar itself) | 323 | +| K2 | `valid` (32, all enums), `allow_zero` (2); no numeric ranges | 34 | +| K3 | `shape` 348: grid 212, `<=dim` 56, aux 44, local 36; `dims` 253 | 601 | +| K4 | required (no `optional`) 332, `when` 50 | 382 | +| K6 | `pk` 55 | 55 | +| K7 | `fk` 58 (9 cross-component), `time_series` 173, `model` 13, `dynamic` 7 | 251 | +| K8 | `parent` 145, `multi` 145, `component` 136, `component_ftype` 10 | 436 | +| K9 | `fill_forward` 70 | 70 | +| K5, K10 | — | 0 | +| (fill) | `default` 93: absence semantics, not a constraint | 93 | + +MF6 has 1,279 sites: 1,203 errors and 76 warnings. Another 398 +`store_error_filename`/`_unit` calls only add a file name to an error and +aren't counted as sites. Regex clusters estimate ~13% are not input: +I/O, programmer errors, numerics. + +Of the sampled sites, MF6 checks 118 at load and 29 at allocate-and-read +(`ar`). It checks 14 at read-and-prepare (`rp`), once per period, and 5 at +advance (`ad`), once per time step. Late checks are almost all +cross-component (K7, K8, K10 joins), plus time-series coverage, which fails +only during integration at `ad` (`TimeSeries.f90:552-601`). A spec-level +validator has the whole simulation at load and could run all of them up +front. + +## Gap list + +Report only. No edits to devtools or modflow6. + +**Only in Fortran, and expressible in the DFN as it stands:** + +- **Field `when` on the MST arrays:** + - `bulk_density`, `distcoef` and `sp2` under `sorption`; + - `decay` and `decay_sorbed` under the decay options; + - `sp2` only for Freundlich or Langmuir, which needs a value test, not + just presence. + + The same applies to the GWE counterparts. The `schema` tags from the + option classification name exactly these options. +- `pk` on the SSM SOURCES `pname` (`tsp-ssm.f90:820`), MVR PACKAGES + (`gwf-mvr.f90:917`), and the LAK/MAW/SFR "specified more than once" checks + where they aren't already marked. +- `valid` for string enums that are checked in Fortran but not declared: + - SFR `cprior`; + - LAK `claktype` and `couttype`; + - MAW `condeqn`; + - IMS `complexity`, `under_relaxation`, `no_ptc_option` and + `rclose_option`. + + Their descriptions list the values, but there's no `valid` key. 32 `valid` + keys exist, mostly on OC `rtype`, the advection scheme and sorption. + +**Only in Fortran, needing new DFN syntax:** + +- **Numeric domains** (K2). This is the biggest gap by count: the "sign/value" + regex cluster alone has 165 sites. A `range` or `min`/`max` key, with + open/closed bounds, would cover `> 0`, `≥ 0` and `[0, 1]`. Row-level + comparisons need an expression: `THTS > THTR`, `DTMIN < DTMAX`, + `telev ≥ belev`. +- **Option coherence** (K5): a `conflicts`/`requires` key on options. NPF + alone has 7 such rules (`gwf-npf.f90:1715-1757`). +- **Block-level `when`:** `ANGLEDATA` under `NON_VERTICAL_WELLS`, and + `DIVERSIONS` only when some reach has `ndv > 0`. +- **Co-occurrence and requiredness of components** (K8a). `multi = false` + says at most one, but nothing says at least one. Examples: DIS is required + in every model, IC and NPF in GWF, DFW in OLF. Nothing expresses MVR ⇒ MVT + or LKT ⇒ LAK. +- **Cross-component dimensions:** SPC `MAXBOUND` must cover the flow + package's bounds (`TspSpc.f90:347`). +- **Key kinds on roles:** the SPC↔SPCA pairing is enforced in Fortran + (`TspSpc.f90:360-380`) and only implied by DFN naming. This is H4 from + the directions doc, now confirmed against MF6. + +**Only in prose (mf6io, hand-written; 209 rule-like sentences, 7 checked):** + +- DISV vertices in clockwise order (`gwf/disv.tex`): not enforced. +- A TAS file can be associated with only one package (`gwf/tsi.tex`): no + check found. +- LKT needs a corresponding LAK, and the advanced transport packages take + their flow package's name by default (`gwt/sft.tex`, `lkt.tex`). These + are enforced indirectly, through the name lookup (`gwt-sft.f90:206`). +- PRT/GWT through FMI need particular GWF output (`SAVE BUDGET ALL`, and + heads for all layers). FMI checks part of this (`tsp-fmi.f90:676-688`). + The rest is a constraint on another run's output. + +**Only in the DFN:** none found in the sample. The DFN's `pk`/`fk`/`shape` +rules all have Fortran counterparts where I looked. I didn't check this +systematically. + +**Catalog-adjacent:** time-series coverage is checked only at `ad`. A +spec could state it as `TS times ⊇ simulation span`, unless +`extendToEndOfSimulation` applies, and check it up front. + +## For step 2 (DDL slice) + +1. Put **dimensions in as first-class values**. K3 is mostly + `count(rows) = dim` or `≤ dim` (7 of 10 relate configuration to data). + The question in the directions doc about whether `MAXBOUND` is layout or + identity becomes concrete: `MAXBOUND` is a CHECK bound on a table's row + count. Bound identity comes from the row key (`<=maxbound` as a PK + domain), not from the dimension. +2. **The grid is a table everyone joins against.** K10 assertions join + package rows to cell geometry. The DDL slice should include the grid as + a relation (cell → top, bot, area, idomain). Model it as a parameter, + not per package. +3. **Use a scope chain for `when`** (own options → model → binding + parameters), and drop `MODELNAMES` in favour of a binding parameter. +4. **Use CHECK on the options row for K5.** A component's options form a + one-row table, so option coherence is a row-level CHECK. This confirms + the config/data axis: configuration constraints are schema-level and + can be checked without data. +5. **Expect the misfits:** monotone row order (LAK table stages, PRP release + times, PXDP), SFR acyclicity, and geometric predicates (point in cell, + clockwise). These need functions or recursive queries. Keep them as + named checks, the way rules work today. +6. **Lens-domain constraints (L) go with the lens**, not in the logical + schema. They're the step 5 partiality conditions. + +## Limits + +- The regex clusters agree with hand labels only 31% of the time (49 of + 158). Don't read the cluster counts as kind counts. They're message + shapes ("must be > 0" covers K2 and K10). +- The sample is purposive: families were chosen in advance, and within + each family I chose instances for variety. A random sample of ~50 would + give per-kind proportions for MF6 as a whole, if those are needed. +- Of the mf6io prose, I checked 7 of 209 hand-written rule sentences in + detail. The 799 rule sentences in the generated `mf6ivar` duplicate DFN + descriptions. They're the "only in description text" category, and I + didn't parse them. diff --git a/constraint-audit/dfn-constraints.csv b/constraint-audit/dfn-constraints.csv new file mode 100644 index 00000000..aee8ba1d --- /dev/null +++ b/constraint-audit/dfn-constraints.csv @@ -0,0 +1,2246 @@ +component,block,path,key,value,kind,level,cfgdata,trigger,note +chf-cdb,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-cdb,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-cdb,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-cdb,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-cdb,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-cdb,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-cdb,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-cdb,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-cdb,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-cdb,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-cdb,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +chf-cdb,period,period.stress_period_data.idcxs,fk,chf-cxs.packagedata.idcxs,K7,logical,data,,cross-component +chf-cdb,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-cdb,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-cdb,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-cdb,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +chf-chd,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-chd,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-chd,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-chd,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-chd,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-chd,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +chf-chd,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-chd,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-chd,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-chd,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-chd,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-chd,period,period.stress_period_data.head,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-chd,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-chd,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-chd,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +chf-cxs,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-cxs,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-cxs,,dims.nsections,dims,nsections,K3,logical,config,,local dimension +chf-cxs,,dims.npoints,dims,npoints,K3,logical,config,,local dimension +chf-cxs,dimensions,dimensions.nsections,required,True,K4,logical,config,unconditional, +chf-cxs,dimensions,dimensions.npoints,required,True,K4,logical,config,unconditional, +chf-cxs,packagedata,packagedata.packagedata,shape,['nsections'],K3,logical,data,,local +chf-cxs,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +chf-cxs,packagedata,packagedata.packagedata.idcxs,index,True,K1,layout,data,,1-based index +chf-cxs,packagedata,packagedata.packagedata.idcxs,pk,True,K6,logical,data,, +chf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,shape,['npoints'],K3,logical,data,,local +chf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,required,True,K4,logical,data,unconditional, +chf-dfw,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-dfw,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-dfw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-dfw,griddata,griddata.manningsn,shape,['nodes'],K3,logical,data<-config,,grid +chf-dfw,griddata,griddata.idcxs,index,True,K1,layout,data,,1-based index +chf-dfw,griddata,griddata.idcxs,shape,['nodes'],K3,logical,data<-config,,grid +chf-dfw,griddata,griddata.idcxs,fk,chf-cxs.packagedata.idcxs,K7,logical,data,,cross-component +chf-disv1d,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-disv1d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-disv1d,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +chf-disv1d,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +chf-disv1d,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +chf-disv1d,options,options.crs,case_sensitive,True,K1,layout,config,, +chf-disv1d,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +chf-disv1d,griddata,griddata.width,shape,['nodes'],K3,logical,data<-config,,grid +chf-disv1d,griddata,griddata.bottom,shape,['nodes'],K3,logical,data<-config,,grid +chf-disv1d,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +chf-disv1d,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +chf-disv1d,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +chf-disv1d,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +chf-disv1d,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +chf-disv1d,cell1d,cell1d.cell1d,shape,['nodes'],K3,logical,data<-config,,grid +chf-disv1d,cell1d,cell1d.cell1d,required,True,K4,logical,data,unconditional, +chf-disv1d,cell1d,cell1d.cell1d.icell1d,index,True,K1,layout,data,,1-based index +chf-disv1d,cell1d,cell1d.cell1d.icell1d,pk,True,K6,logical,data,, +chf-disv1d,cell1d,cell1d.cell1d.icvert,index,True,K1,layout,data,,1-based index +chf-disv1d,cell1d,cell1d.cell1d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +chf-evp,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-evp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-evp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-evp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-evp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-evp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +chf-evp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-evp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-evp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-evp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-evp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-evp,period,period.stress_period_data.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-evp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-evp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-evp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +chf-flw,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-flw,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-flw,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-flw,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-flw,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-flw,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +chf-flw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-flw,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-flw,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-flw,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-flw,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-flw,period,period.stress_period_data.q,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-flw,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-flw,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-flw,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +chf-ic,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +chf-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +chf-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +chf-ic,griddata,griddata.strt,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +chf-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +chf-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-nam,options,options.list,case_sensitive,True,K1,layout,config,, +chf-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +chf-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +chf-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +chf-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +chf-oc,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-oc,period,period.stress_period_data.saverecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +chf-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +chf-oc,period,period.stress_period_data.printrecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +chf-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +chf-pcp,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-pcp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-pcp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-pcp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-pcp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-pcp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +chf-pcp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-pcp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-pcp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-pcp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-pcp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-pcp,period,period.stress_period_data.precipitation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-pcp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-pcp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-pcp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +chf-sto,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-sto,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +chf-sto,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-sto,period,period.stress_period_data,valid,"['steady-state', 'transient']",K2,logical,data,,enum domain +chf-zdg,,,parent,chf-nam,K8,logical,config,binding,may be bound only under this parent type +chf-zdg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +chf-zdg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +chf-zdg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +chf-zdg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +chf-zdg,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +chf-zdg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +chf-zdg,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +chf-zdg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +chf-zdg,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +chf-zdg,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +chf-zdg,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +chf-zdg,period,period.stress_period_data.idcxs,fk,chf-cxs.packagedata.idcxs,K7,logical,data,,cross-component +chf-zdg,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-zdg,period,period.stress_period_data.slope,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-zdg,period,period.stress_period_data.rough,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-zdg,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +chf-zdg,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +chf-zdg,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +exg-chfgwf,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-chfgwf,,,multi,True,K8,logical,config,binding,cardinality per parent: many +exg-chfgwf,,dims.nexg,dims,nexg,K3,logical,config,,local dimension +exg-chfgwf,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +exg-chfgwf,dimensions,dimensions.nexg,required,True,K4,logical,config,unconditional, +exg-chfgwf,exchangedata,exchangedata.exchangedata,shape,['nexg'],K3,logical,data,,local +exg-chfgwf,exchangedata,exchangedata.exchangedata,required,True,K4,logical,data,unconditional, +exg-chfgwf,exchangedata,exchangedata.exchangedata.cellidm1,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-chfgwf,exchangedata,exchangedata.exchangedata.cellidm2,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwegwe,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwegwe,,,multi,True,K8,logical,config,binding,cardinality per parent: many +exg-gwegwe,,dims.nexg,dims,nexg,K3,logical,config,,local dimension +exg-gwegwe,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwegwe,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwegwe,options,options.gwfmodelname1,required,True,K4,logical,config,unconditional, +exg-gwegwe,options,options.gwfmodelname2,required,True,K4,logical,config,unconditional, +exg-gwegwe,options,options.adv_scheme,valid,"['upstream', 'central', 'tvd']",K2,logical,config,,enum domain +exg-gwegwe,options,options.mve_filerecord.mve6_filename,component,gwe-mve,K8,logical,config,binding,child component type +exg-gwegwe,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +exg-gwegwe,dimensions,dimensions.nexg,required,True,K4,logical,config,unconditional, +exg-gwegwe,exchangedata,exchangedata.exchangedata,shape,['nexg'],K3,logical,data,,local +exg-gwegwe,exchangedata,exchangedata.exchangedata,required,True,K4,logical,data,unconditional, +exg-gwegwe,exchangedata,exchangedata.exchangedata.cellidm1,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwegwe,exchangedata,exchangedata.exchangedata.cellidm2,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwegwe,exchangedata,exchangedata.exchangedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +exg-gwegwe,exchangedata,exchangedata.exchangedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +exg-gwfgwe,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwfgwe,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +exg-gwfgwf,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwfgwf,,,multi,True,K8,logical,config,binding,cardinality per parent: many +exg-gwfgwf,,dims.nexg,dims,nexg,K3,logical,config,,local dimension +exg-gwfgwf,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwfgwf,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwfgwf,options,options.cell_averaging,valid,"['harmonic', 'logarithmic', 'amt-lmk']",K2,logical,config,,enum domain +exg-gwfgwf,options,options.gnc_filerecord.gnc6_filename,component,gwf-gnc,K8,logical,config,binding,child component type +exg-gwfgwf,options,options.mvr_filerecord.mvr6_filename,component,gwf-mvr,K8,logical,config,binding,child component type +exg-gwfgwf,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +exg-gwfgwf,dimensions,dimensions.nexg,required,True,K4,logical,config,unconditional, +exg-gwfgwf,exchangedata,exchangedata.exchangedata,shape,['nexg'],K3,logical,data,,local +exg-gwfgwf,exchangedata,exchangedata.exchangedata,required,True,K4,logical,data,unconditional, +exg-gwfgwf,exchangedata,exchangedata.exchangedata.cellidm1,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwfgwf,exchangedata,exchangedata.exchangedata.cellidm2,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwfgwf,exchangedata,exchangedata.exchangedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +exg-gwfgwf,exchangedata,exchangedata.exchangedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +exg-gwfgwt,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwfgwt,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +exg-gwfprt,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwfprt,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +exg-gwtgwt,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-gwtgwt,,,multi,True,K8,logical,config,binding,cardinality per parent: many +exg-gwtgwt,,dims.nexg,dims,nexg,K3,logical,config,,local dimension +exg-gwtgwt,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwtgwt,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +exg-gwtgwt,options,options.gwfmodelname1,required,True,K4,logical,config,unconditional, +exg-gwtgwt,options,options.gwfmodelname2,required,True,K4,logical,config,unconditional, +exg-gwtgwt,options,options.adv_scheme,valid,"['upstream', 'central', 'tvd']",K2,logical,config,,enum domain +exg-gwtgwt,options,options.mvt_filerecord.mvt6_filename,component,gwt-mvt,K8,logical,config,binding,child component type +exg-gwtgwt,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +exg-gwtgwt,dimensions,dimensions.nexg,required,True,K4,logical,config,unconditional, +exg-gwtgwt,exchangedata,exchangedata.exchangedata,shape,['nexg'],K3,logical,data,,local +exg-gwtgwt,exchangedata,exchangedata.exchangedata,required,True,K4,logical,data,unconditional, +exg-gwtgwt,exchangedata,exchangedata.exchangedata.cellidm1,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwtgwt,exchangedata,exchangedata.exchangedata.cellidm2,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-gwtgwt,exchangedata,exchangedata.exchangedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +exg-gwtgwt,exchangedata,exchangedata.exchangedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +exg-olfgwf,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +exg-olfgwf,,,multi,True,K8,logical,config,binding,cardinality per parent: many +exg-olfgwf,,dims.nexg,dims,nexg,K3,logical,config,,local dimension +exg-olfgwf,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +exg-olfgwf,dimensions,dimensions.nexg,required,True,K4,logical,config,unconditional, +exg-olfgwf,exchangedata,exchangedata.exchangedata,shape,['nexg'],K3,logical,data,,local +exg-olfgwf,exchangedata,exchangedata.exchangedata,required,True,K4,logical,data,unconditional, +exg-olfgwf,exchangedata,exchangedata.exchangedata.cellidm1,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +exg-olfgwf,exchangedata,exchangedata.exchangedata.cellidm2,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +gwe-adv,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-adv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-adv,options,options.scheme,valid,"['central', 'upstream', 'tvd']",K2,logical,config,,enum domain +gwe-cnd,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-cnd,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-cnd,griddata,griddata.alh,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.alh,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.alh,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.alv,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.alv,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.alv,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.ath1,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ath1,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ath1,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.ath2,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ath2,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ath2,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.atv,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.atv,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.atv,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.ktw,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ktw,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.ktw,shape,['nodes'],K3,logical,data<-config,,grid +gwe-cnd,griddata,griddata.kts,layered,True,K1,layout,data,, +gwe-cnd,griddata,griddata.kts,netcdf,True,K1,layout,data,, +gwe-cnd,griddata,griddata.kts,shape,['nodes'],K3,logical,data<-config,,grid +gwe-ctp,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-ctp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-ctp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwe-ctp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-ctp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-ctp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-ctp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-ctp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwe-ctp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-ctp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwe-ctp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-ctp,period,period.stress_period_data.temp,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-ctp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-ctp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-ctp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-dis,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-dis,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-dis,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwe-dis,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +gwe-dis,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +gwe-dis,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwe-dis,,dims.nodes,dims,nlay * nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwe-dis,,dims.ncelldim,dims,3,K3,logical,config,binding:model,cross-component dimension +gwe-dis,options,options.crs,case_sensitive,True,K1,layout,config,, +gwe-dis,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwe-dis,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwe-dis,dimensions,dimensions.nlay,default,1,fill,logical,config,,"absence semantics, not a constraint" +gwe-dis,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +gwe-dis,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwe-dis,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +gwe-dis,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwe-dis,griddata,griddata.delr,netcdf,True,K1,layout,data,, +gwe-dis,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +gwe-dis,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +gwe-dis,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwe-dis,griddata,griddata.delc,netcdf,True,K1,layout,data,, +gwe-dis,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +gwe-dis,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +gwe-dis,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwe-dis,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwe-dis,griddata,griddata.top,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +gwe-dis,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwe-dis,griddata,griddata.top,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwe-dis,griddata,griddata.botm,layered,True,K1,layout,data,, +gwe-dis,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwe-dis,griddata,griddata.botm,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwe-dis,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwe-dis,griddata,griddata.botm,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwe-dis,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwe-dis,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwe-dis,griddata,griddata.idomain,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwe-disu,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-disu,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-disu,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +gwe-disu,,dims.nja,dims,nja,K3,logical,config,binding:model,cross-component dimension +gwe-disu,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwe-disu,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +gwe-disu,,dims.njas,dims,(nja - nodes) / 2,K3,logical,config,,local dimension +gwe-disu,options,options.vertical_offset_tolerance,default,0.0,fill,logical,config,,"absence semantics, not a constraint" +gwe-disu,options,options.crs,case_sensitive,True,K1,layout,config,, +gwe-disu,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +gwe-disu,dimensions,dimensions.nja,required,True,K4,logical,config,unconditional, +gwe-disu,griddata,griddata.top,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwe-disu,griddata,griddata.bot,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,griddata,griddata.bot,required,True,K4,logical,data,unconditional, +gwe-disu,griddata,griddata.area,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,griddata,griddata.area,required,True,K4,logical,data,unconditional, +gwe-disu,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.iac,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.iac,required,True,K4,logical,data,unconditional, +gwe-disu,connectiondata,connectiondata.ja,index,True,K1,layout,data,,1-based index +gwe-disu,connectiondata,connectiondata.ja,shape,['nja'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.ja,required,True,K4,logical,data,unconditional, +gwe-disu,connectiondata,connectiondata.ihc,shape,['nja'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.ihc,required,True,K4,logical,data,unconditional, +gwe-disu,connectiondata,connectiondata.cl12,shape,['nja'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.cl12,required,True,K4,logical,data,unconditional, +gwe-disu,connectiondata,connectiondata.hwva,shape,['nja'],K3,logical,data<-config,,grid +gwe-disu,connectiondata,connectiondata.hwva,required,True,K4,logical,data,unconditional, +gwe-disu,connectiondata,connectiondata.angldegx,shape,['nja'],K3,logical,data<-config,,grid +gwe-disu,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwe-disu,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +gwe-disu,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwe-disu,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwe-disu,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +gwe-disu,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +gwe-disu,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwe-disu,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwe-disu,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwe-disu,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwe-disv,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-disv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-disv,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwe-disv,,dims.ncpl,dims,ncpl,K3,logical,config,binding:model,cross-component dimension +gwe-disv,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwe-disv,,dims.nodes,dims,nlay * ncpl,K3,logical,config,binding:model,cross-component dimension +gwe-disv,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +gwe-disv,options,options.crs,case_sensitive,True,K1,layout,config,, +gwe-disv,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwe-disv,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwe-disv,dimensions,dimensions.ncpl,required,True,K4,logical,config,unconditional, +gwe-disv,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +gwe-disv,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwe-disv,griddata,griddata.top,shape,['ncpl'],K3,logical,data<-config,,grid +gwe-disv,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwe-disv,griddata,griddata.botm,layered,True,K1,layout,data,, +gwe-disv,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwe-disv,griddata,griddata.botm,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwe-disv,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwe-disv,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwe-disv,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwe-disv,griddata,griddata.idomain,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwe-disv,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwe-disv,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +gwe-disv,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwe-disv,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwe-disv,cell2d,cell2d.cell2d,shape,['ncpl'],K3,logical,data<-config,,grid +gwe-disv,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +gwe-disv,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwe-disv,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwe-disv,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwe-disv,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwe-esl,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-esl,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-esl,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwe-esl,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-esl,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-esl,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-esl,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-esl,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwe-esl,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-esl,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwe-esl,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-esl,period,period.stress_period_data.senerrate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-esl,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-esl,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-esl,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-est,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-est,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-est,options,options.density_water,default,1000.0,fill,logical,config,,"absence semantics, not a constraint" +gwe-est,options,options.heat_capacity_water,default,4184.0,fill,logical,config,,"absence semantics, not a constraint" +gwe-est,options,options.latent_heat_vaporization,default,2453500.0,fill,logical,config,,"absence semantics, not a constraint" +gwe-est,griddata,griddata.porosity,layered,True,K1,layout,data,, +gwe-est,griddata,griddata.porosity,shape,['nodes'],K3,logical,data<-config,,grid +gwe-est,griddata,griddata.porosity,required,True,K4,logical,data,unconditional, +gwe-est,griddata,griddata.decay_water,layered,True,K1,layout,data,, +gwe-est,griddata,griddata.decay_water,shape,['nodes'],K3,logical,data<-config,,grid +gwe-est,griddata,griddata.decay_solid,layered,True,K1,layout,data,, +gwe-est,griddata,griddata.decay_solid,shape,['nodes'],K3,logical,data<-config,,grid +gwe-est,griddata,griddata.heat_capacity_solid,layered,True,K1,layout,data,, +gwe-est,griddata,griddata.heat_capacity_solid,shape,['nodes'],K3,logical,data<-config,,grid +gwe-est,griddata,griddata.heat_capacity_solid,required,True,K4,logical,data,unconditional, +gwe-est,griddata,griddata.density_solid,layered,True,K1,layout,data,, +gwe-est,griddata,griddata.density_solid,shape,['nodes'],K3,logical,data<-config,,grid +gwe-est,griddata,griddata.density_solid,required,True,K4,logical,data,unconditional, +gwe-fmi,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-fmi,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-ic,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +gwe-ic,griddata,griddata.strt,netcdf,True,K1,layout,data,, +gwe-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +gwe-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +gwe-ic,griddata,griddata.strt,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwe-lke,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-lke,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-lke,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-lke,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-lke,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-lke,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwe-lke,packagedata,packagedata.packagedata.lakeno,index,True,K1,layout,data,,1-based index +gwe-lke,packagedata,packagedata.packagedata.lakeno,pk,True,K6,logical,data,, +gwe-lke,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-lke,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-lke,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-lke,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-lke,period,period.stress_period_data.lakeno,index,True,K1,layout,data,,1-based index +gwe-lke,period,period.stress_period_data.lakeno,fk,packagedata.lakeno,K7,logical,data,,intra-component +gwe-lke,period,period.stress_period_data.laksetting.temperature,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,period,period.stress_period_data.laksetting.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,period,period.stress_period_data.laksetting.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,period,period.stress_period_data.laksetting.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,period,period.stress_period_data.laksetting.ext-inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-lke,period,period.stress_period_data.laksetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-mve,,,parent,"['exg-gwegwe', 'gwe-nam']",K8,logical,config,binding,may be bound only under this parent type +gwe-mve,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-mwe,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-mwe,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-mwe,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-mwe,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-mwe,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-mwe,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwe-mwe,packagedata,packagedata.packagedata.mawno,index,True,K1,layout,data,,1-based index +gwe-mwe,packagedata,packagedata.packagedata.mawno,pk,True,K6,logical,data,, +gwe-mwe,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-mwe,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-mwe,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-mwe,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-mwe,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-mwe,period,period.stress_period_data.mawno,index,True,K1,layout,data,,1-based index +gwe-mwe,period,period.stress_period_data.mawno,fk,packagedata.mawno,K7,logical,data,,intra-component +gwe-mwe,period,period.stress_period_data.mwesetting.temperature,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-mwe,period,period.stress_period_data.mwesetting.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-mwe,period,period.stress_period_data.mwesetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-nam,options,options.list,case_sensitive,True,K1,layout,config,, +gwe-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +gwe-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +gwe-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +gwe-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +gwe-oc,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-oc,period,period.stress_period_data.saverecord.rtype,valid,"['TEMPERATURE', 'BUDGET']",K2,output,data,,enum domain +gwe-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwe-oc,period,period.stress_period_data.printrecord.rtype,valid,"['TEMPERATURE', 'BUDGET']",K2,output,data,,enum domain +gwe-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwe-sfe,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-sfe,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-sfe,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-sfe,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-sfe,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-sfe,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwe-sfe,packagedata,packagedata.packagedata.rno,index,True,K1,layout,data,,1-based index +gwe-sfe,packagedata,packagedata.packagedata.rno,pk,True,K6,logical,data,, +gwe-sfe,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-sfe,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-sfe,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-sfe,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-sfe,period,period.stress_period_data.rno,index,True,K1,layout,data,,1-based index +gwe-sfe,period,period.stress_period_data.rno,fk,packagedata.rno,K7,logical,data,,intra-component +gwe-sfe,period,period.stress_period_data.reachsetting.temperature,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,period,period.stress_period_data.reachsetting.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,period,period.stress_period_data.reachsetting.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,period,period.stress_period_data.reachsetting.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,period,period.stress_period_data.reachsetting.inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-sfe,period,period.stress_period_data.reachsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-ssm,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-ssm,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwe-ssm,sources,,write_if_empty,True,K1,layout,data,, +gwe-ssm,sources,sources.sources,required,True,K4,logical,data,unconditional, +gwe-ssm,fileinput,fileinput.fileinput.spc6_filename,component,"['utl-spc', 'utl-spca']",K8,logical,data,binding,child component type +gwe-uze,,,parent,gwe-nam,K8,logical,config,binding,may be bound only under this parent type +gwe-uze,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwe-uze,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwe-uze,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwe-uze,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwe-uze,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwe-uze,packagedata,packagedata.packagedata.uzfno,index,True,K1,layout,data,,1-based index +gwe-uze,packagedata,packagedata.packagedata.uzfno,pk,True,K6,logical,data,, +gwe-uze,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwe-uze,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-uze,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwe-uze,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwe-uze,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwe-uze,period,period.stress_period_data.uzfno,index,True,K1,layout,data,,1-based index +gwe-uze,period,period.stress_period_data.uzfno,fk,packagedata.uzfno,K7,logical,data,,intra-component +gwe-uze,period,period.stress_period_data.uzesetting.temperature,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-uze,period,period.stress_period_data.uzesetting.infiltration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-uze,period,period.stress_period_data.uzesetting.uzet,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwe-uze,period,period.stress_period_data.uzesetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-api,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-api,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-api,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-api,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-api,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-buy,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-buy,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-buy,,dims.nrhospecies,dims,nrhospecies,K3,logical,config,,local dimension +gwf-buy,options,options.denseref,default,1000.0,fill,logical,config,,"absence semantics, not a constraint" +gwf-buy,dimensions,dimensions.nrhospecies,required,True,K4,logical,config,unconditional, +gwf-buy,packagedata,packagedata.packagedata,shape,['nrhospecies'],K3,logical,data,,local +gwf-buy,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-buy,packagedata,packagedata.packagedata.irhospec,index,True,K1,layout,data,,1-based index +gwf-buy,packagedata,packagedata.packagedata.irhospec,pk,True,K6,logical,data,, +gwf-chd,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-chd,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-chd,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-chd,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-chd,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-chd,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-chd,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-chd,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-chd,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-chd,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-chd,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-chd,period,period.stress_period_data.head,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-chd,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-chd,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-chd,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-chdg,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-chdg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-chdg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-chdg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-chdg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-chdg,options,options.readarraygrid,required,True,K4,logical,config,unconditional, +gwf-chdg,options,options.readarraygrid,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-chdg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-chdg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-chdg,period,period.head,layered,True,K1,layout,data,, +gwf-chdg,period,period.head,netcdf,True,K1,layout,data,, +gwf-chdg,period,period.head,shape,['nodes'],K3,logical,data<-config,,grid +gwf-chdg,period,period.head,required,True,K4,logical,data,unconditional, +gwf-chdg,period,period.head,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-chdg,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-chdg,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-chdg,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-chdg,period,period.aux.aux,layered,True,K1,layout,data,, +gwf-chdg,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-chdg,period,period.aux.aux,shape,['nodes'],K3,logical,data<-config,,grid +gwf-csub,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-csub,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-csub,,dims.ninterbeds,dims,ninterbeds,K3,logical,config,,local dimension +gwf-csub,,dims.maxsig0,dims,maxsig0,K3,logical,config,,local dimension +gwf-csub,options,options.gammaw,default,9806.65,fill,logical,config,,"absence semantics, not a constraint" +gwf-csub,options,options.beta,default,4.6512e-10,fill,logical,config,,"absence semantics, not a constraint" +gwf-csub,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-csub,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-csub,dimensions,dimensions.ninterbeds,required,True,K4,logical,config,unconditional, +gwf-csub,griddata,griddata.cg_ske_cr,layered,True,K1,layout,data,, +gwf-csub,griddata,griddata.cg_ske_cr,netcdf,True,K1,layout,data,, +gwf-csub,griddata,griddata.cg_ske_cr,shape,['nodes'],K3,logical,data<-config,,grid +gwf-csub,griddata,griddata.cg_ske_cr,required,True,K4,logical,data,unconditional, +gwf-csub,griddata,griddata.cg_ske_cr,default,1e-05,fill,logical,data,,"absence semantics, not a constraint" +gwf-csub,griddata,griddata.cg_theta,layered,True,K1,layout,data,, +gwf-csub,griddata,griddata.cg_theta,netcdf,True,K1,layout,data,, +gwf-csub,griddata,griddata.cg_theta,shape,['nodes'],K3,logical,data<-config,,grid +gwf-csub,griddata,griddata.cg_theta,required,True,K4,logical,data,unconditional, +gwf-csub,griddata,griddata.cg_theta,default,0.2,fill,logical,data,,"absence semantics, not a constraint" +gwf-csub,griddata,griddata.sgm,layered,True,K1,layout,data,, +gwf-csub,griddata,griddata.sgm,netcdf,True,K1,layout,data,, +gwf-csub,griddata,griddata.sgm,shape,['nodes'],K3,logical,data<-config,,grid +gwf-csub,griddata,griddata.sgs,layered,True,K1,layout,data,, +gwf-csub,griddata,griddata.sgs,netcdf,True,K1,layout,data,, +gwf-csub,griddata,griddata.sgs,shape,['nodes'],K3,logical,data<-config,,grid +gwf-csub,packagedata,packagedata.packagedata,shape,['ninterbeds'],K3,logical,data,,local +gwf-csub,packagedata,packagedata.packagedata.icsubno,index,True,K1,layout,data,,1-based index +gwf-csub,packagedata,packagedata.packagedata.icsubno,pk,True,K6,logical,data,, +gwf-csub,packagedata,packagedata.packagedata.theta,default,0.2,fill,logical,data,,"absence semantics, not a constraint" +gwf-csub,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-csub,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-csub,period,period.stress_period_data,shape,['<=maxsig0'],K3,logical,data,,bound +gwf-csub,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-csub,period,period.stress_period_data.sig0,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-dis,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-dis,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-dis,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwf-dis,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +gwf-dis,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +gwf-dis,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwf-dis,,dims.nodes,dims,nlay * nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwf-dis,,dims.ncelldim,dims,3,K3,logical,config,binding:model,cross-component dimension +gwf-dis,options,options.crs,case_sensitive,True,K1,layout,config,, +gwf-dis,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwf-dis,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwf-dis,dimensions,dimensions.nlay,default,1,fill,logical,config,,"absence semantics, not a constraint" +gwf-dis,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +gwf-dis,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwf-dis,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +gwf-dis,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwf-dis,griddata,griddata.delr,netcdf,True,K1,layout,data,, +gwf-dis,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +gwf-dis,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +gwf-dis,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-dis,griddata,griddata.delc,netcdf,True,K1,layout,data,, +gwf-dis,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +gwf-dis,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +gwf-dis,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-dis,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwf-dis,griddata,griddata.top,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +gwf-dis,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwf-dis,griddata,griddata.top,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-dis,griddata,griddata.botm,layered,True,K1,layout,data,, +gwf-dis,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwf-dis,griddata,griddata.botm,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwf-dis,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwf-dis,griddata,griddata.botm,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-dis,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwf-dis,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwf-dis,griddata,griddata.idomain,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwf-disu,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-disu,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-disu,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +gwf-disu,,dims.nja,dims,nja,K3,logical,config,binding:model,cross-component dimension +gwf-disu,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwf-disu,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +gwf-disu,,dims.njas,dims,(nja - nodes) / 2,K3,logical,config,,local dimension +gwf-disu,options,options.vertical_offset_tolerance,default,0.0,fill,logical,config,,"absence semantics, not a constraint" +gwf-disu,options,options.crs,case_sensitive,True,K1,layout,config,, +gwf-disu,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +gwf-disu,dimensions,dimensions.nja,required,True,K4,logical,config,unconditional, +gwf-disu,griddata,griddata.top,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwf-disu,griddata,griddata.bot,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,griddata,griddata.bot,required,True,K4,logical,data,unconditional, +gwf-disu,griddata,griddata.area,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,griddata,griddata.area,required,True,K4,logical,data,unconditional, +gwf-disu,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.iac,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.iac,required,True,K4,logical,data,unconditional, +gwf-disu,connectiondata,connectiondata.ja,index,True,K1,layout,data,,1-based index +gwf-disu,connectiondata,connectiondata.ja,shape,['nja'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.ja,required,True,K4,logical,data,unconditional, +gwf-disu,connectiondata,connectiondata.ihc,shape,['nja'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.ihc,required,True,K4,logical,data,unconditional, +gwf-disu,connectiondata,connectiondata.cl12,shape,['nja'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.cl12,required,True,K4,logical,data,unconditional, +gwf-disu,connectiondata,connectiondata.hwva,shape,['nja'],K3,logical,data<-config,,grid +gwf-disu,connectiondata,connectiondata.hwva,required,True,K4,logical,data,unconditional, +gwf-disu,connectiondata,connectiondata.angldegx,shape,['nja'],K3,logical,data<-config,,grid +gwf-disu,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwf-disu,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwf-disu,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwf-disu,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +gwf-disu,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwf-disu,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwf-disu,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwf-disu,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwf-disv,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-disv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-disv,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwf-disv,,dims.ncpl,dims,ncpl,K3,logical,config,binding:model,cross-component dimension +gwf-disv,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwf-disv,,dims.nodes,dims,nlay * ncpl,K3,logical,config,binding:model,cross-component dimension +gwf-disv,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +gwf-disv,options,options.crs,case_sensitive,True,K1,layout,config,, +gwf-disv,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwf-disv,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwf-disv,dimensions,dimensions.ncpl,required,True,K4,logical,config,unconditional, +gwf-disv,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +gwf-disv,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwf-disv,griddata,griddata.top,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-disv,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwf-disv,griddata,griddata.botm,layered,True,K1,layout,data,, +gwf-disv,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwf-disv,griddata,griddata.botm,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwf-disv,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwf-disv,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwf-disv,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwf-disv,griddata,griddata.idomain,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwf-disv,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwf-disv,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +gwf-disv,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwf-disv,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwf-disv,cell2d,cell2d.cell2d,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-disv,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +gwf-disv,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwf-disv,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwf-disv,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwf-disv,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwf-drn,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-drn,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-drn,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-drn,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-drn,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-drn,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-drn,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-drn,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-drn,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-drn,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-drn,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-drn,period,period.stress_period_data.elev,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-drn,period,period.stress_period_data.cond,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-drn,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-drn,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-drn,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-drng,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-drng,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-drng,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-drng,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-drng,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-drng,options,options.readarraygrid,required,True,K4,logical,config,unconditional, +gwf-drng,options,options.readarraygrid,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-drng,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-drng,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-drng,period,period.elev,layered,True,K1,layout,data,, +gwf-drng,period,period.elev,netcdf,True,K1,layout,data,, +gwf-drng,period,period.elev,shape,['nodes'],K3,logical,data<-config,,grid +gwf-drng,period,period.elev,required,True,K4,logical,data,unconditional, +gwf-drng,period,period.elev,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-drng,period,period.cond,layered,True,K1,layout,data,, +gwf-drng,period,period.cond,netcdf,True,K1,layout,data,, +gwf-drng,period,period.cond,shape,['nodes'],K3,logical,data<-config,,grid +gwf-drng,period,period.cond,required,True,K4,logical,data,unconditional, +gwf-drng,period,period.cond,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-drng,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-drng,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-drng,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-drng,period,period.aux.aux,layered,True,K1,layout,data,, +gwf-drng,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-drng,period,period.aux.aux,shape,['nodes'],K3,logical,data<-config,,grid +gwf-evt,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-evt,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-evt,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-evt,,dims.nseg,dims,nseg,K3,logical,config,,local dimension +gwf-evt,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-evt,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-evt,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-evt,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-evt,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-evt,dimensions,dimensions.nseg,required,True,K4,logical,config,unconditional, +gwf-evt,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-evt,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-evt,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-evt,period,period.stress_period_data.surface,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.depth,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.pxdp,shape,['nseg-1'],K3,logical,data,,local +gwf-evt,period,period.stress_period_data.pxdp,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.petm,shape,['nseg-1'],K3,logical,data,,local +gwf-evt,period,period.stress_period_data.petm,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.petm0,when,surf_rate_specified,K4,logical,data,option:surf_rate_specified[logical/schema], +gwf-evt,period,period.stress_period_data.petm0,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-evt,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evt,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-evta,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-evta,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-evta,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-evta,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-evta,options,options.readasarrays,required,True,K4,logical,config,unconditional, +gwf-evta,options,options.readasarrays,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-evta,options,options.tas_filerecord.tas6_filename,component,utl-tas,K8,logical,config,binding,child component type +gwf-evta,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-evta,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-evta,period,period.ievt,index,True,K1,layout,data,,1-based index +gwf-evta,period,period.ievt,netcdf,True,K1,layout,data,, +gwf-evta,period,period.ievt,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-evta,period,period.surface,netcdf,True,K1,layout,data,, +gwf-evta,period,period.surface,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-evta,period,period.surface,required,True,K4,logical,data,unconditional, +gwf-evta,period,period.surface,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-evta,period,period.rate,netcdf,True,K1,layout,data,, +gwf-evta,period,period.rate,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-evta,period,period.rate,required,True,K4,logical,data,unconditional, +gwf-evta,period,period.rate,default,0.001,fill,logical,data,,"absence semantics, not a constraint" +gwf-evta,period,period.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-evta,period,period.depth,netcdf,True,K1,layout,data,, +gwf-evta,period,period.depth,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-evta,period,period.depth,required,True,K4,logical,data,unconditional, +gwf-evta,period,period.depth,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-evta,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-evta,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-evta,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-evta,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-evta,period,period.aux.aux,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-evta,period,period.aux.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-ghb,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-ghb,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-ghb,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-ghb,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-ghb,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-ghb,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-ghb,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-ghb,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-ghb,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-ghb,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-ghb,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-ghb,period,period.stress_period_data.bhead,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-ghb,period,period.stress_period_data.cond,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-ghb,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-ghb,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-ghb,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-ghbg,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-ghbg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-ghbg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-ghbg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-ghbg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-ghbg,options,options.readarraygrid,required,True,K4,logical,config,unconditional, +gwf-ghbg,options,options.readarraygrid,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-ghbg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-ghbg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-ghbg,period,period.bhead,layered,True,K1,layout,data,, +gwf-ghbg,period,period.bhead,netcdf,True,K1,layout,data,, +gwf-ghbg,period,period.bhead,shape,['nodes'],K3,logical,data<-config,,grid +gwf-ghbg,period,period.bhead,required,True,K4,logical,data,unconditional, +gwf-ghbg,period,period.bhead,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-ghbg,period,period.cond,layered,True,K1,layout,data,, +gwf-ghbg,period,period.cond,netcdf,True,K1,layout,data,, +gwf-ghbg,period,period.cond,shape,['nodes'],K3,logical,data<-config,,grid +gwf-ghbg,period,period.cond,required,True,K4,logical,data,unconditional, +gwf-ghbg,period,period.cond,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-ghbg,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-ghbg,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-ghbg,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-ghbg,period,period.aux.aux,layered,True,K1,layout,data,, +gwf-ghbg,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-ghbg,period,period.aux.aux,shape,['nodes'],K3,logical,data<-config,,grid +gwf-gnc,,,parent,"['exg-gwfgwf', 'gwf-nam']",K8,logical,config,binding,may be bound only under this parent type +gwf-gnc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-gnc,,dims.numgnc,dims,numgnc,K3,logical,config,,local dimension +gwf-gnc,,dims.numalphaj,dims,numalphaj,K3,logical,config,,local dimension +gwf-gnc,dimensions,dimensions.numgnc,required,True,K4,logical,config,unconditional, +gwf-gnc,dimensions,dimensions.numalphaj,required,True,K4,logical,config,unconditional, +gwf-gnc,gncdata,gncdata.gncdata,shape,['numgnc'],K3,logical,data,,local +gwf-gnc,gncdata,gncdata.gncdata,required,True,K4,logical,data,unconditional, +gwf-gnc,gncdata,gncdata.gncdata.cellidn,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +gwf-gnc,gncdata,gncdata.gncdata.cellidm,model,2,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +gwf-gnc,gncdata,gncdata.gncdata.cellidsj,allow_zero,True,K2,layout,data,,0 is a null placeholder in a cellid list +gwf-gnc,gncdata,gncdata.gncdata.cellidsj,shape,['numalphaj'],K3,logical,data,,local +gwf-gnc,gncdata,gncdata.gncdata.cellidsj,model,1,K7,logical,data,binding,cellid resolves in the exchange's model 1/2 +gwf-gnc,gncdata,gncdata.gncdata.alphasj,shape,['numalphaj'],K3,logical,data,,local +gwf-hfb,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-hfb,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-hfb,,dims.maxhfb,dims,maxhfb,K3,logical,config,,local dimension +gwf-hfb,dimensions,dimensions.maxhfb,required,True,K4,logical,config,unconditional, +gwf-hfb,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-hfb,period,period.stress_period_data,shape,['<=maxhfb'],K3,logical,data,,bound +gwf-hfb,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-ic,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +gwf-ic,griddata,griddata.strt,netcdf,True,K1,layout,data,, +gwf-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +gwf-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +gwf-ic,griddata,griddata.strt,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-lak,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-lak,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-lak,,dims.nlakes,dims,nlakes,K3,logical,config,,local dimension +gwf-lak,,dims.noutlets,dims,noutlets,K3,logical,config,,local dimension +gwf-lak,,dims.ntables,dims,ntables,K3,logical,config,,local dimension +gwf-lak,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-lak,,dims.nlakeconn,dims,sum(packagedata.nlakeconn),K3,logical,config,,local dimension +gwf-lak,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-lak,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-lak,dimensions,dimensions.nlakes,required,True,K4,logical,config,unconditional, +gwf-lak,dimensions,dimensions.noutlets,required,True,K4,logical,config,unconditional, +gwf-lak,dimensions,dimensions.ntables,required,True,K4,logical,config,unconditional, +gwf-lak,packagedata,packagedata.packagedata,shape,['nlakes'],K3,logical,data,,local +gwf-lak,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-lak,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwf-lak,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwf-lak,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-lak,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-lak,connectiondata,connectiondata.connectiondata,shape,['nlakeconn'],K3,logical,data,,local +gwf-lak,connectiondata,connectiondata.connectiondata,required,True,K4,logical,data,unconditional, +gwf-lak,connectiondata,connectiondata.connectiondata.ifno,index,True,K1,layout,data,,1-based index +gwf-lak,connectiondata,connectiondata.connectiondata.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,connectiondata,connectiondata.connectiondata.iconn,index,True,K1,layout,data,,1-based index +gwf-lak,tables,tables.tables,shape,['ntables'],K3,logical,data,,local +gwf-lak,tables,tables.tables,required,True,K4,logical,data,unconditional, +gwf-lak,tables,tables.tables.ifno,index,True,K1,layout,data,,1-based index +gwf-lak,tables,tables.tables.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,tables,tables.tables.tab6_filename,component,utl-laktab,K8,logical,data,binding,child component type +gwf-lak,outlets,outlets.outlets,shape,['noutlets'],K3,logical,data,,local +gwf-lak,outlets,outlets.outlets.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,outlets,outlets.outlets.outletno,pk,True,K6,logical,data,, +gwf-lak,outlets,outlets.outlets.lakein,index,True,K1,layout,data,,1-based index +gwf-lak,outlets,outlets.outlets.lakein,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,outlets,outlets.outlets.lakeout,index,True,K1,layout,data,,1-based index +gwf-lak,outlets,outlets.outlets.lakeout,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,outlets,outlets.outlets.invert,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,outlets,outlets.outlets.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,outlets,outlets.outlets.rough,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,outlets,outlets.outlets.slope,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-lak,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-lak,period,period.stress_period_data.status.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.status.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.stage.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.stage.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.stage.stage,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.rainfall.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.rainfall.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.rainfall.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.evaporation.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.evaporation.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.evaporation.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.runoff.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.runoff.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.runoff.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.inflow.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.inflow.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.inflow.inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.withdrawal.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.withdrawal.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.withdrawal.withdrawal,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.rate.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.rate.outletno,fk,outlets.outletno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.rate.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.invert.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.invert.outletno,fk,outlets.outletno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.invert.invert,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.width.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.width.outletno,fk,outlets.outletno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.width.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.slope.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.slope.outletno,fk,outlets.outletno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.slope.slope,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.rough.outletno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.rough.outletno,fk,outlets.outletno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.rough.rough,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-lak,period,period.stress_period_data.auxiliaryrecord.lakeno,index,True,K1,layout,data,,1-based index +gwf-lak,period,period.stress_period_data.auxiliaryrecord.lakeno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-lak,period,period.stress_period_data.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-maw,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-maw,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-maw,,dims.nmawwells,dims,nmawwells,K3,logical,config,,local dimension +gwf-maw,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-maw,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-maw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-maw,dimensions,dimensions.nmawwells,required,True,K4,logical,config,unconditional, +gwf-maw,packagedata,packagedata.packagedata,shape,['nmawwells'],K3,logical,data,,local +gwf-maw,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-maw,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwf-maw,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwf-maw,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-maw,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-maw,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-maw,connectiondata,connectiondata.connectiondata,required,True,K4,logical,data,unconditional, +gwf-maw,connectiondata,connectiondata.connectiondata.ifno,index,True,K1,layout,data,,1-based index +gwf-maw,connectiondata,connectiondata.connectiondata.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-maw,connectiondata,connectiondata.connectiondata.icon,index,True,K1,layout,data,,1-based index +gwf-maw,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-maw,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-maw,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwf-maw,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-maw,period,period.stress_period_data.mawsetting.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-maw,period,period.stress_period_data.mawsetting.well_head,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-maw,period,period.stress_period_data.mawsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-mvr,,,parent,"['exg-gwfgwf', 'gwf-nam']",K8,logical,config,binding,may be bound only under this parent type +gwf-mvr,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-mvr,,dims.maxmvr,dims,maxmvr,K3,logical,config,,local dimension +gwf-mvr,,dims.maxpackages,dims,maxpackages,K3,logical,config,,local dimension +gwf-mvr,dimensions,dimensions.maxmvr,required,True,K4,logical,config,unconditional, +gwf-mvr,dimensions,dimensions.maxpackages,required,True,K4,logical,config,unconditional, +gwf-mvr,packages,packages.packages,shape,['maxpackages'],K3,logical,data,,local +gwf-mvr,packages,packages.packages,required,True,K4,logical,data,unconditional, +gwf-mvr,packages,packages.packages.mname,when,modelnames,K4,layout,data,option:modelnames[layout], +gwf-mvr,packages,packages.packages.mname,fk,sim-nam.models.mname,K7,logical,data,,cross-component +gwf-mvr,packages,packages.packages.pname,pk,{'scope': 'mname'},K6,logical,data,, +gwf-mvr,packages,packages.packages.pname,fk,"{'path': 'gwf-nam.packages.pname', 'scope': 'mname'}",K7,logical,data,column:mname,"cross-component, scoped" +gwf-mvr,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-mvr,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-mvr,period,period.stress_period_data.mname1,when,modelnames,K4,layout,data,option:modelnames[layout], +gwf-mvr,period,period.stress_period_data.mname1,fk,sim-nam.models.mname,K7,logical,data,,cross-component +gwf-mvr,period,period.stress_period_data.pname1,fk,"{'path': 'packages.pname', 'scope': 'mname1'}",K7,logical,data,column:mname1,"intra-component, scoped" +gwf-mvr,period,period.stress_period_data.id1,index,True,K1,layout,data,,1-based index +gwf-mvr,period,period.stress_period_data.mname2,when,modelnames,K4,layout,data,option:modelnames[layout], +gwf-mvr,period,period.stress_period_data.mname2,fk,sim-nam.models.mname,K7,logical,data,,cross-component +gwf-mvr,period,period.stress_period_data.pname2,fk,"{'path': 'packages.pname', 'scope': 'mname2'}",K7,logical,data,column:mname2,"intra-component, scoped" +gwf-mvr,period,period.stress_period_data.id2,index,True,K1,layout,data,,1-based index +gwf-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-nam,options,options.list,case_sensitive,True,K1,layout,config,, +gwf-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +gwf-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +gwf-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +gwf-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +gwf-npf,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-npf,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-npf,options,options.alternative_cell_averaging,valid,"['logarithmic', 'amt-lmk', 'amt-hmk']",K2,logical,config,,enum domain +gwf-npf,options,options.tvk_filerecord.tvk6_filename,component,utl-tvk,K8,logical,config,binding,child component type +gwf-npf,griddata,griddata.icelltype,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.icelltype,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.icelltype,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.icelltype,default,0,fill,logical,data,,"absence semantics, not a constraint" +gwf-npf,griddata,griddata.k,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.k,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.k,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.k,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwf-npf,griddata,griddata.k22,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.k22,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.k22,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.k33,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.k33,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.k33,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.angle1,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle1,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle1,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.angle2,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle2,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle2,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.angle3,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle3,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.angle3,shape,['nodes'],K3,logical,data<-config,,grid +gwf-npf,griddata,griddata.wetdry,layered,True,K1,layout,data,, +gwf-npf,griddata,griddata.wetdry,netcdf,True,K1,layout,data,, +gwf-npf,griddata,griddata.wetdry,shape,['nodes'],K3,logical,data<-config,,grid +gwf-oc,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-oc,period,period.stress_period_data.saverecord.rtype,valid,"['HEAD', 'BUDGET']",K2,output,data,,enum domain +gwf-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwf-oc,period,period.stress_period_data.printrecord.rtype,valid,"['HEAD', 'BUDGET']",K2,output,data,,enum domain +gwf-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwf-rch,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-rch,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-rch,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-rch,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rch,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rch,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-rch,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-rch,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-rch,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-rch,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-rch,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-rch,period,period.stress_period_data.recharge,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-rch,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-rch,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-rch,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-rcha,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-rcha,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-rcha,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rcha,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rcha,options,options.readasarrays,required,True,K4,logical,config,unconditional, +gwf-rcha,options,options.readasarrays,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-rcha,options,options.tas_filerecord.tas6_filename,component,utl-tas,K8,logical,config,binding,child component type +gwf-rcha,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-rcha,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-rcha,period,period.irch,index,True,K1,layout,data,,1-based index +gwf-rcha,period,period.irch,netcdf,True,K1,layout,data,, +gwf-rcha,period,period.irch,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-rcha,period,period.recharge,netcdf,True,K1,layout,data,, +gwf-rcha,period,period.recharge,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-rcha,period,period.recharge,required,True,K4,logical,data,unconditional, +gwf-rcha,period,period.recharge,default,0.001,fill,logical,data,,"absence semantics, not a constraint" +gwf-rcha,period,period.recharge,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-rcha,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-rcha,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-rcha,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-rcha,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-rcha,period,period.aux.aux,shape,['ncpl'],K3,logical,data<-config,,grid +gwf-rcha,period,period.aux.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-riv,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-riv,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-riv,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-riv,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-riv,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-riv,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-riv,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-riv,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-riv,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-riv,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-riv,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-riv,period,period.stress_period_data.stage,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-riv,period,period.stress_period_data.cond,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-riv,period,period.stress_period_data.rbot,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-riv,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-riv,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-riv,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-rivg,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-rivg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-rivg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-rivg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rivg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-rivg,options,options.readarraygrid,required,True,K4,logical,config,unconditional, +gwf-rivg,options,options.readarraygrid,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-rivg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-rivg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-rivg,period,period.stage,layered,True,K1,layout,data,, +gwf-rivg,period,period.stage,netcdf,True,K1,layout,data,, +gwf-rivg,period,period.stage,shape,['nodes'],K3,logical,data<-config,,grid +gwf-rivg,period,period.stage,required,True,K4,logical,data,unconditional, +gwf-rivg,period,period.stage,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-rivg,period,period.cond,layered,True,K1,layout,data,, +gwf-rivg,period,period.cond,netcdf,True,K1,layout,data,, +gwf-rivg,period,period.cond,shape,['nodes'],K3,logical,data<-config,,grid +gwf-rivg,period,period.cond,required,True,K4,logical,data,unconditional, +gwf-rivg,period,period.cond,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-rivg,period,period.rbot,layered,True,K1,layout,data,, +gwf-rivg,period,period.rbot,netcdf,True,K1,layout,data,, +gwf-rivg,period,period.rbot,shape,['nodes'],K3,logical,data<-config,,grid +gwf-rivg,period,period.rbot,required,True,K4,logical,data,unconditional, +gwf-rivg,period,period.rbot,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-rivg,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-rivg,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-rivg,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-rivg,period,period.aux.aux,layered,True,K1,layout,data,, +gwf-rivg,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-rivg,period,period.aux.aux,shape,['nodes'],K3,logical,data<-config,,grid +gwf-sfr,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-sfr,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-sfr,,dims.nreaches,dims,nreaches,K3,logical,config,,local dimension +gwf-sfr,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-sfr,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-sfr,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-sfr,dimensions,dimensions.nreaches,required,True,K4,logical,config,unconditional, +gwf-sfr,packagedata,packagedata.packagedata,shape,['nreaches'],K3,logical,data,,local +gwf-sfr,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-sfr,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwf-sfr,packagedata,packagedata.packagedata.cellid,allow_zero,True,K2,layout,data,,0 is a null placeholder in a cellid list +gwf-sfr,packagedata,packagedata.packagedata.rhk,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,packagedata,packagedata.packagedata.man,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,packagedata,packagedata.packagedata.ustrf,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-sfr,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-sfr,crosssections,crosssections.crosssections.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,crosssections,crosssections.crosssections.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,crosssections,crosssections.crosssections.tab6_filename,component,utl-sfrtab,K8,logical,data,binding,child component type +gwf-sfr,connectiondata,connectiondata.connectiondata,required,True,K4,logical,data,unconditional, +gwf-sfr,connectiondata,connectiondata.connectiondata.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,connectiondata,connectiondata.connectiondata.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,connectiondata,connectiondata.connectiondata.ic,index,signed,K1,layout,data,,signed index: sign encodes a second attribute +gwf-sfr,connectiondata,connectiondata.connectiondata.ic,shape,['packagedata.ncon(ifno)'],K3,logical,data,,local +gwf-sfr,connectiondata,connectiondata.connectiondata.ic,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,diversions,diversions.diversions.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,diversions,diversions.diversions.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,diversions,diversions.diversions.idv,index,True,K1,layout,data,,1-based index +gwf-sfr,diversions,diversions.diversions.iconr,index,True,K1,layout,data,,1-based index +gwf-sfr,diversions,diversions.diversions.iconr,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,initialstages,initialstages.initialstages.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,initialstages,initialstages.initialstages.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-sfr,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-sfr,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwf-sfr,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-sfr,period,period.stress_period_data.sfrsetting.bedk,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.manning,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.stage,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.diversionrecord.idv,index,True,K1,layout,data,,1-based index +gwf-sfr,period,period.stress_period_data.sfrsetting.diversionrecord.divflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sfr,period,period.stress_period_data.sfrsetting.cross_sectionrecord.tab6_filename,component,utl-sfrtab,K8,logical,data,binding,child component type +gwf-sfr,period,period.stress_period_data.sfrsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-sto,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-sto,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-sto,options,options.tvs_filerecord.tvs6_filename,component,utl-tvs,K8,logical,config,binding,child component type +gwf-sto,griddata,griddata.iconvert,layered,True,K1,layout,data,, +gwf-sto,griddata,griddata.iconvert,netcdf,True,K1,layout,data,, +gwf-sto,griddata,griddata.iconvert,shape,['nodes'],K3,logical,data<-config,,grid +gwf-sto,griddata,griddata.iconvert,required,True,K4,logical,data,unconditional, +gwf-sto,griddata,griddata.iconvert,default,0,fill,logical,data,,"absence semantics, not a constraint" +gwf-sto,griddata,griddata.ss,layered,True,K1,layout,data,, +gwf-sto,griddata,griddata.ss,netcdf,True,K1,layout,data,, +gwf-sto,griddata,griddata.ss,shape,['nodes'],K3,logical,data<-config,,grid +gwf-sto,griddata,griddata.ss,required,True,K4,logical,data,unconditional, +gwf-sto,griddata,griddata.ss,default,1e-05,fill,logical,data,,"absence semantics, not a constraint" +gwf-sto,griddata,griddata.sy,layered,True,K1,layout,data,, +gwf-sto,griddata,griddata.sy,netcdf,True,K1,layout,data,, +gwf-sto,griddata,griddata.sy,shape,['nodes'],K3,logical,data<-config,,grid +gwf-sto,griddata,griddata.sy,required,True,K4,logical,data,unconditional, +gwf-sto,griddata,griddata.sy,default,0.15,fill,logical,data,,"absence semantics, not a constraint" +gwf-sto,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-sto,period,period.stress_period_data,valid,"['steady-state', 'transient']",K2,logical,data,,enum domain +gwf-uzf,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-uzf,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-uzf,,dims.nuzfcells,dims,nuzfcells,K3,logical,config,,local dimension +gwf-uzf,,dims.ntrailwaves,dims,ntrailwaves,K3,logical,config,,local dimension +gwf-uzf,,dims.nwavesets,dims,nwavesets,K3,logical,config,,local dimension +gwf-uzf,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-uzf,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-uzf,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-uzf,dimensions,dimensions.nuzfcells,required,True,K4,logical,config,unconditional, +gwf-uzf,dimensions,dimensions.ntrailwaves,required,True,K4,logical,config,unconditional, +gwf-uzf,dimensions,dimensions.ntrailwaves,default,7,fill,logical,config,,"absence semantics, not a constraint" +gwf-uzf,dimensions,dimensions.nwavesets,required,True,K4,logical,config,unconditional, +gwf-uzf,dimensions,dimensions.nwavesets,default,40,fill,logical,config,,"absence semantics, not a constraint" +gwf-uzf,packagedata,packagedata.packagedata,shape,['nuzfcells'],K3,logical,data,,local +gwf-uzf,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-uzf,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwf-uzf,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwf-uzf,packagedata,packagedata.packagedata.ivertcon,index,True,K1,layout,data,,1-based index +gwf-uzf,packagedata,packagedata.packagedata.ivertcon,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-uzf,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-uzf,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-uzf,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-uzf,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwf-uzf,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwf-uzf,period,period.stress_period_data.finf,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.pet,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.extdp,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.extwc,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.ha,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.hroot,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.rootact,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-uzf,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-uzf,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-vsc,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-vsc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwf-vsc,,dims.nviscspecies,dims,nviscspecies,K3,logical,config,,local dimension +gwf-vsc,options,options.viscref,default,1.0,fill,logical,config,,"absence semantics, not a constraint" +gwf-vsc,options,options.thermal_formulation,valid,"['linear', 'nonlinear']",K2,logical,config,,enum domain +gwf-vsc,options,options.thermal_a2,default,10.0,fill,logical,config,,"absence semantics, not a constraint" +gwf-vsc,options,options.thermal_a3,default,248.37,fill,logical,config,,"absence semantics, not a constraint" +gwf-vsc,options,options.thermal_a4,default,133.15,fill,logical,config,,"absence semantics, not a constraint" +gwf-vsc,dimensions,dimensions.nviscspecies,required,True,K4,logical,config,unconditional, +gwf-vsc,packagedata,packagedata.packagedata,shape,['nviscspecies'],K3,logical,data,,local +gwf-vsc,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwf-vsc,packagedata,packagedata.packagedata.iviscspec,index,True,K1,layout,data,,1-based index +gwf-vsc,packagedata,packagedata.packagedata.iviscspec,pk,True,K6,logical,data,, +gwf-wel,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-wel,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-wel,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-wel,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-wel,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-wel,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwf-wel,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-wel,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwf-wel,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-wel,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwf-wel,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwf-wel,period,period.stress_period_data.q,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-wel,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwf-wel,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwf-wel,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwf-welg,,,parent,gwf-nam,K8,logical,config,binding,may be bound only under this parent type +gwf-welg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwf-welg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwf-welg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-welg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwf-welg,options,options.readarraygrid,required,True,K4,logical,config,unconditional, +gwf-welg,options,options.readarraygrid,default,True,fill,logical,config,,"absence semantics, not a constraint" +gwf-welg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwf-welg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwf-welg,period,period.q,layered,True,K1,layout,data,, +gwf-welg,period,period.q,netcdf,True,K1,layout,data,, +gwf-welg,period,period.q,shape,['nodes'],K3,logical,data<-config,,grid +gwf-welg,period,period.q,required,True,K4,logical,data,unconditional, +gwf-welg,period,period.q,default,3e+30,fill,logical,data,,"absence semantics, not a constraint" +gwf-welg,period,period.aux,shape,['<=auxiliary'],K3,logical,data,,bound +gwf-welg,period,period.aux.auxname,fk,options.auxiliary,K7,logical,data,,intra-component +gwf-welg,period,period.aux.auxname,dynamic,True,K7,logical,data,option:auxiliary,array named by data; name must be in AUXILIARY +gwf-welg,period,period.aux.aux,layered,True,K1,layout,data,, +gwf-welg,period,period.aux.aux,netcdf,True,K1,layout,data,, +gwf-welg,period,period.aux.aux,shape,['nodes'],K3,logical,data<-config,,grid +gwt-adv,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-adv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-adv,options,options.scheme,valid,"['central', 'upstream', 'tvd', 'utvd']",K2,logical,config,,enum domain +gwt-api,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-api,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-api,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwt-api,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-api,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwt-cnc,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-cnc,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-cnc,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwt-cnc,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-cnc,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-cnc,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-cnc,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-cnc,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwt-cnc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-cnc,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwt-cnc,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-cnc,period,period.stress_period_data.conc,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-cnc,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-cnc,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-cnc,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-dis,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-dis,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-dis,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwt-dis,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +gwt-dis,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +gwt-dis,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwt-dis,,dims.nodes,dims,nlay * nrow * ncol,K3,logical,config,binding:model,cross-component dimension +gwt-dis,,dims.ncelldim,dims,3,K3,logical,config,binding:model,cross-component dimension +gwt-dis,options,options.crs,case_sensitive,True,K1,layout,config,, +gwt-dis,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwt-dis,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwt-dis,dimensions,dimensions.nlay,default,1,fill,logical,config,,"absence semantics, not a constraint" +gwt-dis,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +gwt-dis,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwt-dis,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +gwt-dis,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +gwt-dis,griddata,griddata.delr,netcdf,True,K1,layout,data,, +gwt-dis,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +gwt-dis,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +gwt-dis,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwt-dis,griddata,griddata.delc,netcdf,True,K1,layout,data,, +gwt-dis,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +gwt-dis,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +gwt-dis,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwt-dis,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwt-dis,griddata,griddata.top,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +gwt-dis,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwt-dis,griddata,griddata.top,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +gwt-dis,griddata,griddata.botm,layered,True,K1,layout,data,, +gwt-dis,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwt-dis,griddata,griddata.botm,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwt-dis,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwt-dis,griddata,griddata.botm,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwt-dis,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwt-dis,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwt-dis,griddata,griddata.idomain,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +gwt-disu,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-disu,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-disu,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +gwt-disu,,dims.nja,dims,nja,K3,logical,config,binding:model,cross-component dimension +gwt-disu,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwt-disu,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +gwt-disu,,dims.njas,dims,(nja - nodes) / 2,K3,logical,config,,local dimension +gwt-disu,options,options.vertical_offset_tolerance,default,0.0,fill,logical,config,,"absence semantics, not a constraint" +gwt-disu,options,options.crs,case_sensitive,True,K1,layout,config,, +gwt-disu,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +gwt-disu,dimensions,dimensions.nja,required,True,K4,logical,config,unconditional, +gwt-disu,griddata,griddata.top,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwt-disu,griddata,griddata.bot,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,griddata,griddata.bot,required,True,K4,logical,data,unconditional, +gwt-disu,griddata,griddata.area,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,griddata,griddata.area,required,True,K4,logical,data,unconditional, +gwt-disu,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.iac,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.iac,required,True,K4,logical,data,unconditional, +gwt-disu,connectiondata,connectiondata.ja,index,True,K1,layout,data,,1-based index +gwt-disu,connectiondata,connectiondata.ja,shape,['nja'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.ja,required,True,K4,logical,data,unconditional, +gwt-disu,connectiondata,connectiondata.ihc,shape,['nja'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.ihc,required,True,K4,logical,data,unconditional, +gwt-disu,connectiondata,connectiondata.cl12,shape,['nja'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.cl12,required,True,K4,logical,data,unconditional, +gwt-disu,connectiondata,connectiondata.hwva,shape,['nja'],K3,logical,data<-config,,grid +gwt-disu,connectiondata,connectiondata.hwva,required,True,K4,logical,data,unconditional, +gwt-disu,connectiondata,connectiondata.angldegx,shape,['nja'],K3,logical,data<-config,,grid +gwt-disu,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwt-disu,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwt-disu,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwt-disu,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +gwt-disu,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwt-disu,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwt-disu,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwt-disu,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwt-disv,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-disv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-disv,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +gwt-disv,,dims.ncpl,dims,ncpl,K3,logical,config,binding:model,cross-component dimension +gwt-disv,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +gwt-disv,,dims.nodes,dims,nlay * ncpl,K3,logical,config,binding:model,cross-component dimension +gwt-disv,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +gwt-disv,options,options.crs,case_sensitive,True,K1,layout,config,, +gwt-disv,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +gwt-disv,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +gwt-disv,dimensions,dimensions.ncpl,required,True,K4,logical,config,unconditional, +gwt-disv,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +gwt-disv,griddata,griddata.top,netcdf,True,K1,layout,data,, +gwt-disv,griddata,griddata.top,shape,['ncpl'],K3,logical,data<-config,,grid +gwt-disv,griddata,griddata.top,required,True,K4,logical,data,unconditional, +gwt-disv,griddata,griddata.botm,layered,True,K1,layout,data,, +gwt-disv,griddata,griddata.botm,netcdf,True,K1,layout,data,, +gwt-disv,griddata,griddata.botm,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwt-disv,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +gwt-disv,griddata,griddata.idomain,layered,True,K1,layout,data,, +gwt-disv,griddata,griddata.idomain,netcdf,True,K1,layout,data,, +gwt-disv,griddata,griddata.idomain,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +gwt-disv,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +gwt-disv,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +gwt-disv,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +gwt-disv,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +gwt-disv,cell2d,cell2d.cell2d,shape,['ncpl'],K3,logical,data<-config,,grid +gwt-disv,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +gwt-disv,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +gwt-disv,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +gwt-disv,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +gwt-disv,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +gwt-dsp,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-dsp,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-dsp,griddata,griddata.diffc,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.diffc,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.diffc,shape,['nodes'],K3,logical,data<-config,,grid +gwt-dsp,griddata,griddata.alh,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.alh,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.alh,shape,['nodes'],K3,logical,data<-config,,grid +gwt-dsp,griddata,griddata.alv,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.alv,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.alv,shape,['nodes'],K3,logical,data<-config,,grid +gwt-dsp,griddata,griddata.ath1,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.ath1,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.ath1,shape,['nodes'],K3,logical,data<-config,,grid +gwt-dsp,griddata,griddata.ath2,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.ath2,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.ath2,shape,['nodes'],K3,logical,data<-config,,grid +gwt-dsp,griddata,griddata.atv,layered,True,K1,layout,data,, +gwt-dsp,griddata,griddata.atv,netcdf,True,K1,layout,data,, +gwt-dsp,griddata,griddata.atv,shape,['nodes'],K3,logical,data<-config,,grid +gwt-fmi,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-fmi,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-ic,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +gwt-ic,griddata,griddata.strt,netcdf,True,K1,layout,data,, +gwt-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +gwt-ic,griddata,griddata.strt,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +gwt-ist,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-ist,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-ist,options,options.sorption,valid,"['linear', 'freundlich', 'langmuir']",K2,logical,config,,enum domain +gwt-ist,griddata,griddata.porosity,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.porosity,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.porosity,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.porosity,required,True,K4,logical,data,unconditional, +gwt-ist,griddata,griddata.volfrac,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.volfrac,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.volfrac,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.volfrac,required,True,K4,logical,data,unconditional, +gwt-ist,griddata,griddata.zetaim,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.zetaim,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.zetaim,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.zetaim,required,True,K4,logical,data,unconditional, +gwt-ist,griddata,griddata.cim,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.cim,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.cim,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.decay,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.decay,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.decay,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.decay_sorbed,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.decay_sorbed,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.decay_sorbed,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.bulk_density,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.bulk_density,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.bulk_density,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.distcoef,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.distcoef,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.distcoef,shape,['nodes'],K3,logical,data<-config,,grid +gwt-ist,griddata,griddata.sp2,layered,True,K1,layout,data,, +gwt-ist,griddata,griddata.sp2,netcdf,True,K1,layout,data,, +gwt-ist,griddata,griddata.sp2,shape,['nodes'],K3,logical,data<-config,,grid +gwt-lkt,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-lkt,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-lkt,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-lkt,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-lkt,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-lkt,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwt-lkt,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwt-lkt,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwt-lkt,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-lkt,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-lkt,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-lkt,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-lkt,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwt-lkt,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwt-lkt,period,period.stress_period_data.laksetting.concentration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,period,period.stress_period_data.laksetting.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,period,period.stress_period_data.laksetting.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,period,period.stress_period_data.laksetting.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,period,period.stress_period_data.laksetting.ext-inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-lkt,period,period.stress_period_data.laksetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-mst,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-mst,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-mst,options,options.sorption,valid,"['linear', 'freundlich', 'langmuir']",K2,logical,config,,enum domain +gwt-mst,griddata,griddata.porosity,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.porosity,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.porosity,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mst,griddata,griddata.porosity,required,True,K4,logical,data,unconditional, +gwt-mst,griddata,griddata.decay,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.decay,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.decay,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mst,griddata,griddata.decay_sorbed,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.decay_sorbed,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.decay_sorbed,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mst,griddata,griddata.bulk_density,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.bulk_density,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.bulk_density,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mst,griddata,griddata.distcoef,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.distcoef,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.distcoef,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mst,griddata,griddata.sp2,layered,True,K1,layout,data,, +gwt-mst,griddata,griddata.sp2,netcdf,True,K1,layout,data,, +gwt-mst,griddata,griddata.sp2,shape,['nodes'],K3,logical,data<-config,,grid +gwt-mvt,,,parent,"['exg-gwtgwt', 'gwt-nam']",K8,logical,config,binding,may be bound only under this parent type +gwt-mvt,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-mwt,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-mwt,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-mwt,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-mwt,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-mwt,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-mwt,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwt-mwt,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwt-mwt,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwt-mwt,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-mwt,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-mwt,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-mwt,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-mwt,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-mwt,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwt-mwt,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwt-mwt,period,period.stress_period_data.mwtsetting.concentration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-mwt,period,period.stress_period_data.mwtsetting.rate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-mwt,period,period.stress_period_data.mwtsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-nam,options,options.list,case_sensitive,True,K1,layout,config,, +gwt-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +gwt-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +gwt-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +gwt-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +gwt-oc,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-oc,period,period.stress_period_data.saverecord.rtype,valid,"['CONCENTRATION', 'BUDGET']",K2,output,data,,enum domain +gwt-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwt-oc,period,period.stress_period_data.printrecord.rtype,valid,"['CONCENTRATION', 'BUDGET']",K2,output,data,,enum domain +gwt-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +gwt-sft,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-sft,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-sft,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-sft,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-sft,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-sft,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwt-sft,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwt-sft,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwt-sft,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-sft,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-sft,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-sft,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-sft,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwt-sft,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwt-sft,period,period.stress_period_data.reachsetting.concentration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,period,period.stress_period_data.reachsetting.rainfall,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,period,period.stress_period_data.reachsetting.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,period,period.stress_period_data.reachsetting.runoff,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,period,period.stress_period_data.reachsetting.inflow,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-sft,period,period.stress_period_data.reachsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-src,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-src,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-src,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +gwt-src,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-src,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-src,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-src,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-src,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +gwt-src,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-src,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +gwt-src,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-src,period,period.stress_period_data.smassrate,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-src,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-src,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-src,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-ssm,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-ssm,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +gwt-ssm,sources,,write_if_empty,True,K1,layout,data,, +gwt-ssm,sources,sources.sources,required,True,K4,logical,data,unconditional, +gwt-ssm,fileinput,fileinput.fileinput.spc6_filename,component,"['utl-spc', 'utl-spca']",K8,logical,data,binding,child component type +gwt-uzt,,,parent,gwt-nam,K8,logical,config,binding,may be bound only under this parent type +gwt-uzt,,,multi,True,K8,logical,config,binding,cardinality per parent: many +gwt-uzt,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +gwt-uzt,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +gwt-uzt,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +gwt-uzt,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +gwt-uzt,packagedata,packagedata.packagedata.ifno,index,True,K1,layout,data,,1-based index +gwt-uzt,packagedata,packagedata.packagedata.ifno,pk,True,K6,logical,data,, +gwt-uzt,packagedata,packagedata.packagedata.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +gwt-uzt,packagedata,packagedata.packagedata.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-uzt,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +gwt-uzt,period,header,fill_forward,True,K9,logical,data,,period block carried forward +gwt-uzt,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +gwt-uzt,period,period.stress_period_data.ifno,index,True,K1,layout,data,,1-based index +gwt-uzt,period,period.stress_period_data.ifno,fk,packagedata.ifno,K7,logical,data,,intra-component +gwt-uzt,period,period.stress_period_data.uztsetting.concentration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-uzt,period,period.stress_period_data.uztsetting.infiltration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-uzt,period,period.stress_period_data.uztsetting.uzet,time_series,True,K7,logical,data,binding:TS6,value may name a time series +gwt-uzt,period,period.stress_period_data.uztsetting.auxiliaryrecord.auxval,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-cdb,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-cdb,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-cdb,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-cdb,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-cdb,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-cdb,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-cdb,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-cdb,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-cdb,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-cdb,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-cdb,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +olf-cdb,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-cdb,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-cdb,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-cdb,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +olf-chd,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-chd,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-chd,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-chd,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-chd,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-chd,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +olf-chd,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-chd,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-chd,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-chd,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-chd,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-chd,period,period.stress_period_data.head,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-chd,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-chd,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-chd,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +olf-cxs,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-cxs,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-cxs,,dims.nsections,dims,nsections,K3,logical,config,,local dimension +olf-cxs,,dims.npoints,dims,npoints,K3,logical,config,,local dimension +olf-cxs,dimensions,dimensions.nsections,required,True,K4,logical,config,unconditional, +olf-cxs,dimensions,dimensions.npoints,required,True,K4,logical,config,unconditional, +olf-cxs,packagedata,packagedata.packagedata,shape,['nsections'],K3,logical,data,,local +olf-cxs,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +olf-cxs,packagedata,packagedata.packagedata.idcxs,index,True,K1,layout,data,,1-based index +olf-cxs,packagedata,packagedata.packagedata.idcxs,pk,True,K6,logical,data,, +olf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,shape,['npoints'],K3,logical,data,,local +olf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,required,True,K4,logical,data,unconditional, +olf-dfw,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-dfw,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-dfw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-dfw,griddata,griddata.manningsn,shape,['nodes'],K3,logical,data<-config,,grid +olf-dfw,griddata,griddata.idcxs,index,True,K1,layout,data,,1-based index +olf-dfw,griddata,griddata.idcxs,shape,['nodes'],K3,logical,data<-config,,grid +olf-dfw,griddata,griddata.idcxs,fk,chf-cxs.packagedata.idcxs,K7,logical,data,,cross-component +olf-dis2d,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-dis2d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-dis2d,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +olf-dis2d,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +olf-dis2d,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +olf-dis2d,,dims.nodes,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +olf-dis2d,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +olf-dis2d,options,options.crs,case_sensitive,True,K1,layout,config,, +olf-dis2d,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +olf-dis2d,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +olf-dis2d,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +olf-dis2d,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +olf-dis2d,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +olf-dis2d,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +olf-dis2d,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +olf-dis2d,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +olf-dis2d,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +olf-dis2d,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +olf-dis2d,griddata,griddata.bottom,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +olf-dis2d,griddata,griddata.bottom,required,True,K4,logical,data,unconditional, +olf-dis2d,griddata,griddata.bottom,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +olf-dis2d,griddata,griddata.idomain,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +olf-disv1d,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-disv1d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-disv1d,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +olf-disv1d,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +olf-disv1d,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +olf-disv1d,options,options.crs,case_sensitive,True,K1,layout,config,, +olf-disv1d,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +olf-disv1d,griddata,griddata.length,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv1d,griddata,griddata.width,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv1d,griddata,griddata.bottom,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv1d,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv1d,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +olf-disv1d,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +olf-disv1d,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +olf-disv1d,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +olf-disv1d,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv1d,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +olf-disv1d,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +olf-disv1d,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +olf-disv1d,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +olf-disv1d,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +olf-disv2d,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-disv2d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-disv2d,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +olf-disv2d,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +olf-disv2d,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +olf-disv2d,options,options.crs,case_sensitive,True,K1,layout,config,, +olf-disv2d,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +olf-disv2d,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +olf-disv2d,griddata,griddata.bottom,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv2d,griddata,griddata.bottom,required,True,K4,logical,data,unconditional, +olf-disv2d,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv2d,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +olf-disv2d,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +olf-disv2d,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +olf-disv2d,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +olf-disv2d,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +olf-disv2d,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +olf-disv2d,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +olf-disv2d,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +olf-disv2d,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +olf-disv2d,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +olf-evp,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-evp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-evp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-evp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-evp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-evp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +olf-evp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-evp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-evp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-evp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-evp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-evp,period,period.stress_period_data.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-evp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-evp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-evp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +olf-flw,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-flw,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-flw,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-flw,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-flw,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-flw,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +olf-flw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-flw,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-flw,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-flw,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-flw,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-flw,period,period.stress_period_data.q,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-flw,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-flw,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-flw,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +olf-ic,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +olf-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +olf-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +olf-ic,griddata,griddata.strt,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +olf-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +olf-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-nam,options,options.list,case_sensitive,True,K1,layout,config,, +olf-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +olf-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +olf-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +olf-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +olf-oc,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-oc,period,period.stress_period_data.saverecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +olf-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +olf-oc,period,period.stress_period_data.printrecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +olf-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +olf-pcp,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-pcp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-pcp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-pcp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-pcp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-pcp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +olf-pcp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-pcp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-pcp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-pcp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-pcp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-pcp,period,period.stress_period_data.precipitation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-pcp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-pcp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-pcp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +olf-sto,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-sto,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +olf-sto,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-sto,period,period.stress_period_data,valid,"['steady-state', 'transient']",K2,logical,data,,enum domain +olf-zdg,,,parent,olf-nam,K8,logical,config,binding,may be bound only under this parent type +olf-zdg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +olf-zdg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +olf-zdg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +olf-zdg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +olf-zdg,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +olf-zdg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +olf-zdg,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +olf-zdg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +olf-zdg,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +olf-zdg,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +olf-zdg,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +olf-zdg,period,period.stress_period_data.idcxs,fk,chf-cxs.packagedata.idcxs,K7,logical,data,,cross-component +olf-zdg,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-zdg,period,period.stress_period_data.slope,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-zdg,period,period.stress_period_data.rough,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-zdg,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +olf-zdg,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +olf-zdg,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +prt-dis,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-dis,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-dis,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +prt-dis,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +prt-dis,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +prt-dis,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +prt-dis,,dims.nodes,dims,nlay * nrow * ncol,K3,logical,config,binding:model,cross-component dimension +prt-dis,,dims.ncelldim,dims,3,K3,logical,config,binding:model,cross-component dimension +prt-dis,options,options.crs,case_sensitive,True,K1,layout,config,, +prt-dis,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +prt-dis,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +prt-dis,dimensions,dimensions.nlay,default,1,fill,logical,config,,"absence semantics, not a constraint" +prt-dis,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +prt-dis,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +prt-dis,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +prt-dis,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +prt-dis,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +prt-dis,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +prt-dis,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +prt-dis,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +prt-dis,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +prt-dis,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +prt-dis,griddata,griddata.top,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +prt-dis,griddata,griddata.top,required,True,K4,logical,data,unconditional, +prt-dis,griddata,griddata.top,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +prt-dis,griddata,griddata.botm,layered,True,K1,layout,data,, +prt-dis,griddata,griddata.botm,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +prt-dis,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +prt-dis,griddata,griddata.botm,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +prt-dis,griddata,griddata.idomain,layered,True,K1,layout,data,, +prt-dis,griddata,griddata.idomain,shape,"['ncol', 'nrow', 'nlay']",K3,logical,data<-config,,grid +prt-disv,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-disv,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-disv,,dims.nlay,dims,nlay,K3,logical,config,binding:model,cross-component dimension +prt-disv,,dims.ncpl,dims,ncpl,K3,logical,config,binding:model,cross-component dimension +prt-disv,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +prt-disv,,dims.nodes,dims,nlay * ncpl,K3,logical,config,binding:model,cross-component dimension +prt-disv,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +prt-disv,options,options.crs,case_sensitive,True,K1,layout,config,, +prt-disv,options,options.ncf_filerecord.ncf6_filename,component,utl-ncf,K8,logical,config,binding,child component type +prt-disv,dimensions,dimensions.nlay,required,True,K4,logical,config,unconditional, +prt-disv,dimensions,dimensions.ncpl,required,True,K4,logical,config,unconditional, +prt-disv,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +prt-disv,griddata,griddata.top,shape,['ncpl'],K3,logical,data<-config,,grid +prt-disv,griddata,griddata.top,required,True,K4,logical,data,unconditional, +prt-disv,griddata,griddata.botm,layered,True,K1,layout,data,, +prt-disv,griddata,griddata.botm,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +prt-disv,griddata,griddata.botm,required,True,K4,logical,data,unconditional, +prt-disv,griddata,griddata.idomain,layered,True,K1,layout,data,, +prt-disv,griddata,griddata.idomain,shape,"['ncpl', 'nlay']",K3,logical,data<-config,,grid +prt-disv,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +prt-disv,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +prt-disv,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +prt-disv,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +prt-disv,cell2d,cell2d.cell2d,shape,['ncpl'],K3,logical,data<-config,,grid +prt-disv,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +prt-disv,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +prt-disv,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +prt-disv,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +prt-disv,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +prt-fmi,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-fmi,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-mip,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-mip,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-mip,griddata,griddata.porosity,layered,True,K1,layout,data,, +prt-mip,griddata,griddata.porosity,shape,['nodes'],K3,logical,data<-config,,grid +prt-mip,griddata,griddata.porosity,required,True,K4,logical,data,unconditional, +prt-mip,griddata,griddata.retfactor,layered,True,K1,layout,data,, +prt-mip,griddata,griddata.retfactor,shape,['nodes'],K3,logical,data<-config,,grid +prt-mip,griddata,griddata.izone,layered,True,K1,layout,data,, +prt-mip,griddata,griddata.izone,shape,['nodes'],K3,logical,data<-config,,grid +prt-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +prt-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +prt-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +prt-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +prt-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +prt-oc,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +prt-oc,,dims.ntracktimes,dims,ntracktimes,K3,logical,config,,local dimension +prt-oc,options,options.track_timesfilerecord.timesfile,case_sensitive,True,K1,layout,config,, +prt-oc,tracktimes,tracktimes.tracktimes,shape,['ntracktimes'],K3,logical,data,,local +prt-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +prt-oc,period,period.stress_period_data.saverecord.rtype,valid,['BUDGET'],K2,output,data,,enum domain +prt-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +prt-oc,period,period.stress_period_data.printrecord.rtype,valid,['BUDGET'],K2,output,data,,enum domain +prt-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +prt-prp,,,parent,prt-nam,K8,logical,config,binding,may be bound only under this parent type +prt-prp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +prt-prp,,dims.nreleasepts,dims,nreleasepts,K3,logical,config,,local dimension +prt-prp,,dims.nreleasetimes,dims,nreleasetimes,K3,logical,config,,local dimension +prt-prp,options,options.exit_solve_tolerance,default,1e-05,fill,logical,config,,"absence semantics, not a constraint" +prt-prp,options,options.release_timesfilerecord.timesfile,case_sensitive,True,K1,layout,config,, +prt-prp,options,options.dry_tracking_method,valid,"['drop', 'stop', 'stay']",K2,logical,config,,enum domain +prt-prp,options,options.dev_forceternary,required,True,K4,logical,config,unconditional, +prt-prp,options,options.coordinate_check_method,valid,"['none', 'eager']",K2,output,config,,enum domain +prt-prp,options,options.coordinate_check_method,default,eager,fill,logical,config,,"absence semantics, not a constraint" +prt-prp,dimensions,dimensions.nreleasepts,required,True,K4,logical,config,unconditional, +prt-prp,dimensions,dimensions.nreleasetimes,required,True,K4,logical,config,unconditional, +prt-prp,packagedata,packagedata.packagedata,shape,['nreleasepts'],K3,logical,data,,local +prt-prp,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +prt-prp,packagedata,packagedata.packagedata.irptno,index,True,K1,layout,data,,1-based index +prt-prp,packagedata,packagedata.packagedata.irptno,pk,True,K6,logical,data,, +prt-prp,packagedata,packagedata.packagedata.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +prt-prp,releasetimes,releasetimes.releasetimes,shape,['nreleasetimes'],K3,logical,data,,local +prt-prp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +prt-prp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +prt-prp,period,period.stress_period_data.steps,shape,['<=nstp'],K3,logical,data,,bound +prt-prp,period,period.stress_period_data.fraction,shape,['<=nstp'],K3,logical,data,,bound +sim-nam,options,options.hpc_filerecord.hpc6_filename,component,utl-hpc,K8,logical,config,binding,child component type +sim-nam,timing,timing.tdis6,required,True,K4,logical,data,unconditional, +sim-nam,timing,timing.tdis6,component,sim-tdis,K8,logical,data,binding,child component type +sim-nam,models,models.models.mfname,component,model,K8,logical,data,binding,child component type +sim-nam,models,models.models.mfname,component_ftype,mtype,K8,logical,data,binding,row's type column selects child component type +sim-nam,models,models.models.mname,pk,True,K6,logical,data,, +sim-nam,exchanges,exchanges.exchanges.exgfile,component,exchange,K8,logical,data,binding,child component type +sim-nam,exchanges,exchanges.exchanges.exgfile,component_ftype,exgtype,K8,logical,data,binding,row's type column selects child component type +sim-nam,exchanges,exchanges.exchanges.exgmnamea,fk,models.mname,K7,logical,data,,intra-component +sim-nam,exchanges,exchanges.exchanges.exgmnameb,fk,models.mname,K7,logical,data,,intra-component +sim-nam,solutiongroup,header,fill_forward,False,K9,logical,data,, +sim-nam,solutiongroup,solutiongroup.solutiongroup,required,True,K4,logical,data,unconditional, +sim-nam,solutiongroup,solutiongroup.solutiongroup.slntype,valid,"['ims6', 'ems6']",K2,logical,data,,enum domain +sim-nam,solutiongroup,solutiongroup.solutiongroup.slnfname,component,solution,K8,logical,data,binding,child component type +sim-nam,solutiongroup,solutiongroup.solutiongroup.slnfname,component_ftype,slntype,K8,logical,data,binding,row's type column selects child component type +sim-nam,solutiongroup,solutiongroup.solutiongroup.slnmnames,fk,models.mname,K7,logical,data,,intra-component +sim-tdis,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +sim-tdis,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +sim-tdis,,dims.nper,dims,nper,K3,logical,config,,local dimension +sim-tdis,options,options.ats_filerecord.ats6_filename,component,utl-ats,K8,logical,config,binding,child component type +sim-tdis,dimensions,dimensions.nper,required,True,K4,logical,config,unconditional, +sim-tdis,dimensions,dimensions.nper,default,1,fill,logical,config,,"absence semantics, not a constraint" +sim-tdis,perioddata,perioddata.perioddata,shape,['nper'],K3,logical,data,,local +sim-tdis,perioddata,perioddata.perioddata,required,True,K4,logical,data,unconditional, +sim-tdis,perioddata,perioddata.perioddata,default,"[[1.0, 1, 1.0]]",fill,logical,data,,"absence semantics, not a constraint" +sln-ems,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +sln-ems,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +sln-ims,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +sln-ims,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +sln-pts,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +sln-pts,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +sln-pts,nonlinear,nonlinear.outer_maximum,required,True,K4,logical,data,unconditional, +swf-cdb,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-cdb,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-cdb,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-cdb,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-cdb,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-cdb,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-cdb,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-cdb,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-cdb,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-cdb,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-cdb,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +swf-cdb,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-cdb,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-cdb,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-cdb,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +swf-chd,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-chd,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-chd,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-chd,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-chd,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-chd,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +swf-chd,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-chd,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-chd,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-chd,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-chd,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-chd,period,period.stress_period_data.head,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-chd,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-chd,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-chd,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +swf-cxs,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-cxs,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-cxs,,dims.nsections,dims,nsections,K3,logical,config,,local dimension +swf-cxs,,dims.npoints,dims,npoints,K3,logical,config,,local dimension +swf-cxs,dimensions,dimensions.nsections,required,True,K4,logical,config,unconditional, +swf-cxs,dimensions,dimensions.npoints,required,True,K4,logical,config,unconditional, +swf-cxs,packagedata,packagedata.packagedata,shape,['nsections'],K3,logical,data,,local +swf-cxs,packagedata,packagedata.packagedata,required,True,K4,logical,data,unconditional, +swf-cxs,packagedata,packagedata.packagedata.idcxs,index,True,K1,layout,data,,1-based index +swf-cxs,packagedata,packagedata.packagedata.idcxs,pk,True,K6,logical,data,, +swf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,shape,['npoints'],K3,logical,data,,local +swf-cxs,crosssectiondata,crosssectiondata.crosssectiondata,required,True,K4,logical,data,unconditional, +swf-dfw,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-dfw,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-dfw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-dfw,griddata,griddata.manningsn,shape,['nodes'],K3,logical,data<-config,,grid +swf-dfw,griddata,griddata.idcxs,index,True,K1,layout,data,,1-based index +swf-dfw,griddata,griddata.idcxs,shape,['nodes'],K3,logical,data<-config,,grid +swf-dis2d,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-dis2d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-dis2d,,dims.nrow,dims,nrow,K3,logical,config,binding:model,cross-component dimension +swf-dis2d,,dims.ncol,dims,ncol,K3,logical,config,binding:model,cross-component dimension +swf-dis2d,,dims.ncpl,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +swf-dis2d,,dims.nodes,dims,nrow * ncol,K3,logical,config,binding:model,cross-component dimension +swf-dis2d,,dims.ncelldim,dims,2,K3,logical,config,binding:model,cross-component dimension +swf-dis2d,options,options.crs,case_sensitive,True,K1,layout,config,, +swf-dis2d,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +swf-dis2d,dimensions,dimensions.nrow,default,2,fill,logical,config,,"absence semantics, not a constraint" +swf-dis2d,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +swf-dis2d,dimensions,dimensions.ncol,default,2,fill,logical,config,,"absence semantics, not a constraint" +swf-dis2d,griddata,griddata.delr,shape,['ncol'],K3,logical,data<-config,,grid +swf-dis2d,griddata,griddata.delr,required,True,K4,logical,data,unconditional, +swf-dis2d,griddata,griddata.delr,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +swf-dis2d,griddata,griddata.delc,shape,['nrow'],K3,logical,data<-config,,grid +swf-dis2d,griddata,griddata.delc,required,True,K4,logical,data,unconditional, +swf-dis2d,griddata,griddata.delc,default,1.0,fill,logical,data,,"absence semantics, not a constraint" +swf-dis2d,griddata,griddata.bottom,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +swf-dis2d,griddata,griddata.bottom,required,True,K4,logical,data,unconditional, +swf-dis2d,griddata,griddata.bottom,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +swf-dis2d,griddata,griddata.idomain,shape,"['ncol', 'nrow']",K3,logical,data<-config,,grid +swf-disv1d,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-disv1d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-disv1d,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +swf-disv1d,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +swf-disv1d,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +swf-disv1d,options,options.crs,case_sensitive,True,K1,layout,config,, +swf-disv1d,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +swf-disv1d,griddata,griddata.width,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv1d,griddata,griddata.bottom,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv1d,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv1d,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +swf-disv1d,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +swf-disv1d,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +swf-disv1d,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +swf-disv1d,cell1d,cell1d.cell1d,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv1d,cell1d,cell1d.cell1d,required,True,K4,logical,data,unconditional, +swf-disv1d,cell1d,cell1d.cell1d.icell1d,index,True,K1,layout,data,,1-based index +swf-disv1d,cell1d,cell1d.cell1d.icell1d,pk,True,K6,logical,data,, +swf-disv1d,cell1d,cell1d.cell1d.icvert,index,True,K1,layout,data,,1-based index +swf-disv1d,cell1d,cell1d.cell1d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +swf-disv2d,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-disv2d,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-disv2d,,dims.nodes,dims,nodes,K3,logical,config,binding:model,cross-component dimension +swf-disv2d,,dims.nvert,dims,nvert,K3,logical,config,binding:model,cross-component dimension +swf-disv2d,,dims.ncelldim,dims,1,K3,logical,config,binding:model,cross-component dimension +swf-disv2d,options,options.crs,case_sensitive,True,K1,layout,config,, +swf-disv2d,dimensions,dimensions.nodes,required,True,K4,logical,config,unconditional, +swf-disv2d,dimensions,dimensions.nvert,required,True,K4,logical,config,unconditional, +swf-disv2d,griddata,griddata.bottom,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv2d,griddata,griddata.bottom,required,True,K4,logical,data,unconditional, +swf-disv2d,griddata,griddata.idomain,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv2d,vertices,vertices.vertices,shape,['nvert'],K3,logical,data<-config,,grid +swf-disv2d,vertices,vertices.vertices,required,True,K4,logical,data,unconditional, +swf-disv2d,vertices,vertices.vertices.iv,index,True,K1,layout,data,,1-based index +swf-disv2d,vertices,vertices.vertices.iv,pk,True,K6,logical,data,, +swf-disv2d,cell2d,cell2d.cell2d,shape,['nodes'],K3,logical,data<-config,,grid +swf-disv2d,cell2d,cell2d.cell2d,required,True,K4,logical,data,unconditional, +swf-disv2d,cell2d,cell2d.cell2d.icell2d,index,True,K1,layout,data,,1-based index +swf-disv2d,cell2d,cell2d.cell2d.icell2d,pk,True,K6,logical,data,, +swf-disv2d,cell2d,cell2d.cell2d.icvert,index,True,K1,layout,data,,1-based index +swf-disv2d,cell2d,cell2d.cell2d.icvert,shape,['ncvert'],K3,logical,data<-config,,grid +swf-evp,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-evp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-evp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-evp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-evp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-evp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +swf-evp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-evp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-evp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-evp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-evp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-evp,period,period.stress_period_data.evaporation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-evp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-evp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-evp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +swf-flw,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-flw,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-flw,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-flw,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-flw,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-flw,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +swf-flw,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-flw,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-flw,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-flw,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-flw,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-flw,period,period.stress_period_data.q,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-flw,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-flw,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-flw,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +swf-ic,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-ic,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-ic,griddata,griddata.strt,layered,True,K1,layout,data,, +swf-ic,griddata,griddata.strt,shape,['nodes'],K3,logical,data<-config,,grid +swf-ic,griddata,griddata.strt,required,True,K4,logical,data,unconditional, +swf-ic,griddata,griddata.strt,default,0.0,fill,logical,data,,"absence semantics, not a constraint" +swf-nam,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +swf-nam,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-nam,options,options.list,case_sensitive,True,K1,layout,config,, +swf-nam,packages,packages.packages,required,True,K4,logical,data,unconditional, +swf-nam,packages,packages.packages.fname,component,package,K8,logical,data,binding,child component type +swf-nam,packages,packages.packages.fname,component_ftype,ftype,K8,logical,data,binding,row's type column selects child component type +swf-nam,packages,packages.packages.pname,pk,True,K6,logical,data,, +swf-oc,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-oc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-oc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-oc,period,period.stress_period_data.saverecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +swf-oc,period,period.stress_period_data.saverecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +swf-oc,period,period.stress_period_data.printrecord.rtype,valid,"['STAGE', 'BUDGET']",K2,output,data,,enum domain +swf-oc,period,period.stress_period_data.printrecord.ocsetting.steps,shape,['<=nstp'],K3,logical,data,,bound +swf-pcp,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-pcp,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-pcp,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-pcp,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-pcp,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-pcp,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +swf-pcp,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-pcp,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-pcp,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-pcp,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-pcp,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-pcp,period,period.stress_period_data.precipitation,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-pcp,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-pcp,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-pcp,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +swf-sto,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-sto,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +swf-sto,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-sto,period,period.stress_period_data,valid,"['steady-state', 'transient']",K2,logical,data,,enum domain +swf-zdg,,,parent,swf-nam,K8,logical,config,binding,may be bound only under this parent type +swf-zdg,,,multi,True,K8,logical,config,binding,cardinality per parent: many +swf-zdg,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +swf-zdg,,dims.auxiliary,dims,len(auxiliary),K3,logical,config,,local dimension +swf-zdg,,dims.naux,dims,len(auxiliary),K3,logical,config,,local dimension +swf-zdg,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +swf-zdg,options,options.obs_filerecord.obs6_filename,component,utl-obs,K8,logical,config,binding,child component type +swf-zdg,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +swf-zdg,period,header,fill_forward,True,K9,logical,data,,period block carried forward +swf-zdg,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +swf-zdg,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +swf-zdg,period,period.stress_period_data.idcxs,index,True,K1,layout,data,,1-based index +swf-zdg,period,period.stress_period_data.width,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-zdg,period,period.stress_period_data.slope,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-zdg,period,period.stress_period_data.rough,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-zdg,period,period.stress_period_data.aux,shape,['auxiliary'],K3,logical,data,option:auxiliary,schema +swf-zdg,period,period.stress_period_data.aux,time_series,True,K7,logical,data,binding:TS6,value may name a time series +swf-zdg,period,period.stress_period_data.boundname,when,boundnames,K4,logical,data,option:boundnames[logical/schema], +utl-ats,,,parent,sim-tdis,K8,logical,config,binding,may be bound only under this parent type +utl-ats,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-ats,,dims.maxats,dims,maxats,K3,logical,config,,local dimension +utl-ats,dimensions,dimensions.maxats,required,True,K4,logical,config,unconditional, +utl-ats,dimensions,dimensions.maxats,default,1,fill,logical,config,,"absence semantics, not a constraint" +utl-ats,perioddata,perioddata.perioddata,shape,['<=maxats'],K3,logical,data,,bound +utl-ats,perioddata,perioddata.perioddata,required,True,K4,logical,data,unconditional, +utl-ats,perioddata,perioddata.perioddata.iperats,index,True,K1,layout,data,,1-based index +utl-ats,perioddata,perioddata.perioddata.iperats,pk,True,K6,logical,data,, +utl-hpc,,,parent,sim-nam,K8,logical,config,binding,may be bound only under this parent type +utl-hpc,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-hpc,partitions,partitions.partitions.mname,fk,sim-nam.models.mname,K7,logical,data,,cross-component +utl-laktab,,,parent,gwf-lak,K8,logical,config,binding,may be bound only under this parent type +utl-laktab,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-laktab,,dims.nrow,dims,nrow,K3,logical,config,,local dimension +utl-laktab,,dims.ncol,dims,ncol,K3,logical,config,,local dimension +utl-laktab,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +utl-laktab,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +utl-laktab,table,table.table,shape,['nrow'],K3,logical,data<-config,,grid +utl-laktab,table,table.table,required,True,K4,logical,data,unconditional, +utl-ncf,,,parent,"['gwe-dis', 'gwe-disv', 'gwf-dis', 'gwf-disv', 'gwt-dis', 'gwt-disv', 'prt-dis', 'prt-disv']",K8,logical,config,binding,may be bound only under this parent type +utl-ncf,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-ncf,,dims.ncpl,dims,ncpl,K3,logical,config,,local dimension +utl-ncf,griddata,griddata.latitude,shape,['ncpl'],K3,logical,data<-config,,grid +utl-ncf,griddata,griddata.longitude,shape,['ncpl'],K3,logical,data<-config,,grid +utl-obs,,,parent,"['model', 'package']",K8,logical,config,binding,may be bound only under this parent type +utl-obs,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-obs,continuous,header,fill_forward,False,K9,logical,data,, +utl-obs,continuous,continuous.continuous,required,True,K4,logical,data,unconditional, +utl-obs,continuous,continuous.continuous.id.index,index,True,K1,layout,data,,1-based index +utl-obs,continuous,continuous.continuous.id2.index,index,True,K1,layout,data,,1-based index +utl-sfrtab,,,parent,gwf-sfr,K8,logical,config,binding,may be bound only under this parent type +utl-sfrtab,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-sfrtab,,dims.nrow,dims,nrow,K3,logical,config,,local dimension +utl-sfrtab,,dims.ncol,dims,ncol,K3,logical,config,,local dimension +utl-sfrtab,dimensions,dimensions.nrow,required,True,K4,logical,config,unconditional, +utl-sfrtab,dimensions,dimensions.ncol,required,True,K4,logical,config,unconditional, +utl-sfrtab,table,table.table,shape,['nrow'],K3,logical,data<-config,,grid +utl-sfrtab,table,table.table,required,True,K4,logical,data,unconditional, +utl-spc,,,parent,"['gwe-ssm', 'gwt-ssm']",K8,logical,config,binding,may be bound only under this parent type +utl-spc,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-spc,,dims.maxbound,dims,maxbound,K3,logical,config,,local dimension +utl-spc,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +utl-spc,dimensions,dimensions.maxbound,required,True,K4,logical,config,unconditional, +utl-spc,period,header,fill_forward,True,K9,logical,data,,period block carried forward +utl-spc,period,period.stress_period_data,shape,['<=maxbound'],K3,logical,data,,bound +utl-spc,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +utl-spc,period,period.stress_period_data.bndno,index,True,K1,layout,data,,1-based index +utl-spc,period,period.stress_period_data.bndno,pk,True,K6,logical,data,, +utl-spc,period,period.stress_period_data.spcsetting.concentration,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-spc,period,period.stress_period_data.spcsetting.temperature,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-spca,,,parent,"['gwe-ssm', 'gwt-ssm']",K8,logical,config,binding,may be bound only under this parent type +utl-spca,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-spca,options,options.readasarrays,required,True,K4,logical,config,unconditional, +utl-spca,options,options.readasarrays,default,True,fill,logical,config,,"absence semantics, not a constraint" +utl-spca,options,options.tas_filerecord.tas6_filename,component,utl-tas,K8,logical,config,binding,child component type +utl-spca,period,header,fill_forward,True,K9,logical,data,,period block carried forward +utl-spca,period,period.concentration,shape,['ncpl'],K3,logical,data<-config,,grid +utl-spca,period,period.temperature,shape,['ncpl'],K3,logical,data<-config,,grid +utl-tas,,,parent,"['gwf-evta', 'gwf-rcha', 'utl-spca']",K8,logical,config,binding,may be bound only under this parent type +utl-tas,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-tas,,dims.time_series_name,dims,len(time_series_name),K3,logical,config,,local dimension +utl-tas,attributes,attributes.interpolation_methodrecord.interpolation_method,valid,"['stepwise', 'linear', 'linearend']",K2,logical,data,,enum domain +utl-tas,attributes,attributes.sfacrecord.sfacval,shape,['time_series_name'],K3,logical,data,,local +utl-tas,time,header,fill_forward,False,K9,logical,data,, +utl-tas,time,time.tas_array,shape,['ncpl'],K3,logical,data<-config,,grid +utl-tas,time,time.tas_array,required,True,K4,logical,data,unconditional, +utl-ts,,,parent,package,K8,logical,config,binding,may be bound only under this parent type +utl-ts,,,multi,True,K8,logical,config,binding,cardinality per parent: many +utl-ts,,dims.time_series_names,dims,len(time_series_names),K3,logical,config,,local dimension +utl-ts,attributes,attributes.interpolation_methodrecord.interpolation_method,shape,['time_series_names'],K3,logical,data,,local +utl-ts,attributes,attributes.method,required,True,K4,logical,data,unconditional, +utl-ts,attributes,attributes.interpolation_method_single,valid,"['stepwise', 'linear', 'linearend']",K2,logical,data,,enum domain +utl-ts,attributes,attributes.interpolation_method_single,required,True,K4,logical,data,unconditional, +utl-ts,attributes,attributes.sfacrecord.sfacval,shape,['time_series_names'],K3,logical,data,,local +utl-ts,attributes,attributes.sfac,required,True,K4,logical,data,unconditional, +utl-ts,timeseries,timeseries.timeseries,required,True,K4,logical,data,unconditional, +utl-ts,timeseries,timeseries.timeseries.ts_array,shape,['time_series_names'],K3,logical,data,,local +utl-tvk,,,parent,gwf-npf,K8,logical,config,binding,may be bound only under this parent type +utl-tvk,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-tvk,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +utl-tvk,period,header,fill_forward,True,K9,logical,data,,period block carried forward +utl-tvk,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +utl-tvk,period,period.stress_period_data.tvksetting.k,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-tvk,period,period.stress_period_data.tvksetting.k22,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-tvk,period,period.stress_period_data.tvksetting.k33,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-tvs,,,parent,gwf-sto,K8,logical,config,binding,may be bound only under this parent type +utl-tvs,,,multi,False,K8,logical,config,binding,cardinality per parent: at most one +utl-tvs,options,options.ts_filerecord.ts6_filename,component,utl-ts,K8,logical,config,binding,child component type +utl-tvs,period,header,fill_forward,True,K9,logical,data,,period block carried forward +utl-tvs,period,period.stress_period_data,required,True,K4,logical,data,unconditional, +utl-tvs,period,period.stress_period_data.tvssetting.ss,time_series,True,K7,logical,data,binding:TS6,value may name a time series +utl-tvs,period,period.stress_period_data.tvssetting.sy,time_series,True,K7,logical,data,binding:TS6,value may name a time series diff --git a/constraint-audit/dfn_pass.py b/constraint-audit/dfn_pass.py new file mode 100644 index 00000000..b34a20f7 --- /dev/null +++ b/constraint-audit/dfn_pass.py @@ -0,0 +1,167 @@ +"""Pass 1 of constraint-audit-plan.md: constraint-bearing keys in the dev3 DFN snapshots. + +Usage: python dfn_pass.py +Writes dfn-constraints.csv (one row per constraint-bearing key) and prints tallies. +""" + +import collections +import csv +import glob +import os +import sys +import tomllib + +snap, labels_csv, out = sys.argv[1:4] +LABEL = {r["name"]: (r["category"], r["tags"]) for r in csv.DictReader(open(labels_csv))} + +GRID_DIMS = {"nodes", "ncpl", "nlay", "nrow", "ncol", "nja", "nvert", "ncvert", "ncelldim", + "nodesuser", "ncell2d", "ncell1d"} + +rows = [] + + +def add(comp, block, path, key, value, kind, level, cfgdata, trigger="", note=""): + rows.append(dict(component=comp, block=block, path=".".join(path), key=key, + value=str(value), kind=kind, level=level, cfgdata=cfgdata, + trigger=trigger, note=note)) + + +def cfgdata_of(block): + return "config" if block in ("options", "dimensions") else "data" + + +def shape_kind(dims, comp_dims): + """Classify a shape: grid-sized, bounded row count, schema-derived, or other.""" + kinds = set() + for d in dims: + d = str(d) + if d.startswith("<="): + kinds.add("bound") # row count bounded by a dimension (config vs data) + elif d in ("auxiliary", "naux"): + kinds.add("schema") # columns follow an option list + elif d in GRID_DIMS or comp_dims.get(d, {}).get("scope") == "model": + kinds.add("grid") + else: + kinds.add("local") + return "+".join(sorted(kinds)) + + +def walk(comp, block, path, f, comp_dims, in_record=False): + cd = cfgdata_of(block) + is_option = block == "options" and len(path) == 2 + opt = LABEL.get(path[-1]) if is_option else None + t = f.get("type") + + if "valid" in f: + lvl = "output" if (opt and opt[0] == "output") or comp.endswith("-oc") else "logical" + add(comp, block, path, "valid", f["valid"], "K2", lvl, cd, note="enum domain") + if t in ("integer", "double", "string", "keyword") or "dtype" in f: + pass # type itself: counted in tally only (every field has one) + if "allow_zero" in f: + add(comp, block, path, "allow_zero", f["allow_zero"], "K2", "layout", cd, + note="0 is a null placeholder in a cellid list") + if "index" in f: + note = "1-based index" if f["index"] is True else "signed index: sign encodes a second attribute" + add(comp, block, path, "index", f["index"], "K1", "layout", cd, note=note) + for k in ("tagged", "case_sensitive", "layered", "netcdf", "mode_keyword"): + if k in f and f[k]: + add(comp, block, path, k, f[k], "K1", "layout", cd) + if "shape" in f: + sk = shape_kind(f["shape"], comp_dims) + trig = "option:auxiliary" if "schema" in sk else "" + add(comp, block, path, "shape", f["shape"], "K3", "logical", cd + ("" if "grid" not in sk else "<-config"), + trigger=trig, note=sk) + if not f.get("optional") and not in_record and t not in ("record", "union") \ + and not f.get("removed") and not f.get("deprecated"): + add(comp, block, path, "required", True, "K4", "logical", cd, trigger="unconditional") + if "when" in f: + w = f["when"] + lab = LABEL.get(w, ("?", "")) + add(comp, block, path, "when", w, "K4", "logical" if lab[0] == "logical" else lab[0], cd, + trigger=f"option:{w}[{lab[0]}{'/' + lab[1] if lab[1] else ''}]") + if "default" in f: + add(comp, block, path, "default", f["default"], "fill", "logical", cd, + note="absence semantics, not a constraint") + if "pk" in f: + add(comp, block, path, "pk", f["pk"], "K6", "logical", cd) + if "fk" in f: + fk = f["fk"] + tgt = fk if isinstance(fk, str) else fk.get("path") + scoped = isinstance(fk, dict) and "scope" in fk + cross = "-" in str(tgt).split(".")[0] + add(comp, block, path, "fk", fk, "K7", "logical", cd, + trigger="column:" + fk["scope"] if scoped else "", + note=("cross-component" if cross else "intra-component") + (", scoped" if scoped else "")) + if "model" in f: + add(comp, block, path, "model", f["model"], "K7", "logical", cd, trigger="binding", + note="cellid resolves in the exchange's model 1/2") + if f.get("time_series"): + add(comp, block, path, "time_series", True, "K7", "logical", cd, trigger="binding:TS6", + note="value may name a time series") + if f.get("dynamic"): + add(comp, block, path, "dynamic", True, "K7", "logical", cd, trigger="option:auxiliary", + note="array named by data; name must be in AUXILIARY") + if "component" in f: + add(comp, block, path, "component", f["component"], "K8", "logical", cd, trigger="binding", + note="child component type") + if "component_ftype" in f: + add(comp, block, path, "component_ftype", f["component_ftype"], "K8", "logical", cd, + trigger="binding", note="row's type column selects child component type") + + for n, sub in (f.get("fields") or {}).items(): + walk(comp, block, path + [n], sub, comp_dims, in_record or t in ("record", "union")) + if isinstance(f.get("item"), dict): + walk(comp, block, path, f["item"], comp_dims, True) + for n, sub in (f.get("arms") or {}).items(): + walk(comp, block, path + [n], sub, comp_dims, True) + + +ncomp = 0 +nfields = 0 +for p in sorted(glob.glob(os.path.join(snap, "*.toml"))): + d = tomllib.load(open(p, "rb")) + comp = d["name"] + ncomp += 1 + comp_dims = d.get("dims", {}) or {} + if "parent" in d: + add(comp, "", [], "parent", d["parent"], "K8", "logical", "config", trigger="binding", + note="may be bound only under this parent type") + add(comp, "", [], "multi", d.get("multi", False), "K8", "logical", "config", + trigger="binding", note="cardinality per parent: many" if d.get("multi") else + "cardinality per parent: at most one") + for dn, dd in comp_dims.items(): + sc = dd.get("scope", "") + add(comp, "", ["dims", dn], "dims", dd.get("value"), "K3", "logical", "config", + trigger="binding:model" if sc == "model" else "", + note="cross-component dimension" if sc == "model" else "local dimension") + for b, bd in (d.get("blocks") or {}).items(): + h = bd.get("header") + if isinstance(h, dict): + add(comp, b, ["header"], "fill_forward", h.get("fill_forward", False), "K9", "logical", + "data", note="period block carried forward" if h.get("fill_forward") else "") + if bd.get("write_if_empty"): + add(comp, b, [], "write_if_empty", True, "K1", "layout", "data") + for n, f in (bd.get("fields") or {}).items(): + nfields += 1 + walk(comp, b, [b, n], f, comp_dims) + +os.makedirs(out, exist_ok=True) +with open(os.path.join(out, "dfn-constraints.csv"), "w", newline="") as fh: + w = csv.DictWriter(fh, fieldnames=list(rows[0])) + w.writeheader() + w.writerows(rows) + +print(f"{ncomp} components, {nfields} top-level fields, {len(rows)} constraint rows\n") +byk = collections.Counter((r["kind"], r["key"]) for r in rows) +for (k, key), n in sorted(byk.items()): + print(f" {k:5} {key:16} {n:5}") +print() +kind = collections.Counter(r["kind"] for r in rows) +print("by kind:", dict(sorted(kind.items()))) +print("by level:", dict(collections.Counter(r["level"] for r in rows))) +print("shape classes:", dict(collections.Counter(r["note"] for r in rows if r["key"] == "shape"))) +print("when triggers:", dict(collections.Counter(r["trigger"] for r in rows if r["key"] == "when"))) +print("fk notes:", dict(collections.Counter(r["note"] for r in rows if r["key"] == "fk"))) +print("multi:", dict(collections.Counter(r["note"] for r in rows if r["key"] == "multi"))) +print("valid on:", dict(collections.Counter((r["block"], r["level"]) for r in rows if r["key"] == "valid"))) +print("required by block:", dict(collections.Counter(r["block"] for r in rows if r["key"] == "required").most_common(12))) diff --git a/constraint-audit/mf6-checks.csv b/constraint-audit/mf6-checks.csv new file mode 100644 index 00000000..65a2d98c --- /dev/null +++ b/constraint-audit/mf6-checks.csv @@ -0,0 +1,1280 @@ +file,line,routine,family,call,cluster,kind,message +Model/Connection/GridConnection.f90,324,buildConnections,Connection,store_warning,binding-context,K8,(3a) Exchange will not use vertex and/or cell coordinate data because not all connected models have those data specified +Model/Connection/GweGweConnection.f90,355,validateConnection,Connection,store_error,binding-context,K8,"(1x,a,a,a) Cannot connect GWE models in exchange because one model is configured with ADV and the other one is not" +Model/Connection/GweGweConnection.f90,365,validateConnection,Connection,store_error,binding-context,K8,"(1x,a,a,a) Cannot connect GWE models in exchange because one model is configured with CND and the other one is not" +Model/Connection/GweGweConnection.f90,566,validateGweExchange,Connection,store_error,binding-context,K8,(2a) Energy Storage and Transfer package should be enabled/disabled simultaneously in models connected with the interface model for exchange +Model/Connection/GweGweConnection.f90,575,validateGweExchange,Connection,store_error,binding-context,K8,"(2a) Use of XT3D to calculate conduction should be the same in both models, either both use XT3D or neither for exchange" +Model/Connection/GweGweConnection.f90,597,validateGweExchange,Connection,store_error,binding-context,K8,(6a) Energy storage and transfer (EST) packages in model and should be equivalent to construct an interface model for exchange +Model/Connection/GwfGwfConnection.f90,403,validateGwfExchange,Connection,store_error,binding-context,K8,(2a) Ghost node correction not supported combined with XT3D for exchange +Model/Connection/GwfGwfConnection.f90,408,validateGwfExchange,Connection,store_error,binding-context,K8,(2a) Ghost node correction not supported in parallel run for exchange +Model/Connection/GwfGwfConnection.f90,426,validateGwfExchange,Connection,store_error,binding-context,K8,(2a) Buoyancy package should be enabled/disabled simultaneously in models connected with the interface model for exchange +Model/Connection/GwfGwfConnection.f90,436,validateGwfExchange,Connection,store_error,binding-context,K8,(2a) Connecting models with BUY package not allowed with XT3D enabled on exchange +Model/Connection/GwfGwfConnection.f90,459,validateGwfExchange,Connection,store_error,binding-context,K8,(6a) Buoyancy packages in model and should be equivalent to construct an interface model for exchange +Model/Connection/GwtGwtConnection.f90,333,validateConnection,Connection,store_error,binding-context,K8,"(a,a,a) Cannot connect GWT models in exchange because one model is configured with ADV and the other one is not" +Model/Connection/GwtGwtConnection.f90,343,validateConnection,Connection,store_error,binding-context,K8,"(a,a,a) Cannot connect GWT models in exchange because one model is configured with DSP and the other one is not" +Model/Connection/SpatialModelConnection.f90,635,validateConnection,Connection,store_error,binding-context,K8,"(a,a,a,a,a) XT3D configured on the exchange but the discretization in model has no ANGLDEGX specified" +Model/Connection/SpatialModelConnection.f90,641,validateConnection,Connection,store_error,binding-context,K8,"(a,a,a,a,a) XT3D configured on the exchange but the discretization in model has no ANGLDEGX specified" +Model/Discretization/Dis.f90,264,source_dimensions,Discretization,store_error,range-count,K3,NLAY was not specified or was specified incorrectly. +Model/Discretization/Dis.f90,269,source_dimensions,Discretization,store_error,range-count,K3,NROW was not specified or was specified incorrectly. +Model/Discretization/Dis.f90,274,source_dimensions,Discretization,store_error,range-count,K3,NCOL was not specified or was specified incorrectly. +Model/Discretization/Dis.f90,418,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Ensure IDOMAIN array has some values greater than zero. +Model/Discretization/Dis.f90,439,grid_finalize,Discretization,store_error,unclassified,?, +Model/Discretization/Dis.f90,728,nodeu_to_array,Discretization,store_error,not-input,-,"(a,i0,a,i0,a) Program error: nodeu_to_array size of array ( ) is not equal to the discretization dimension ( )" +Model/Discretization/Dis.f90,759,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i0,a) Node number ( ) less than 1 or greater than the number of nodes." +Model/Discretization/Dis.f90,791,get_nodenumber_idx3,Discretization,store_error,unclassified,?,"('Error in structured-grid cell indices: layer = ',i0,', row = ',i0,', column = ',i0)" +Model/Discretization/Dis.f90,800,get_nodenumber_idx3,Discretization,store_error,range-count,K3,Layer less than one or greater than nlay +Model/Discretization/Dis.f90,802,get_nodenumber_idx3,Discretization,store_error,range-count,K3,Row less than one or greater than nrow +Model/Discretization/Dis.f90,804,get_nodenumber_idx3,Discretization,store_error,range-count,K3,Column less than one or greater than ncol +Model/Discretization/Dis.f90,810,get_nodenumber_idx3,Discretization,store_error,range-count,K3,"(a,i0,a) Node number ( )less than 1 or greater than nodes." +Model/Discretization/Dis.f90,949,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Dis.f90,1040,nodeu_from_cellid,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Cell number cannot be determined for cellid ( ) and results in a user node number ( ) that is outside of the grid." +Model/Discretization/Dis.f90,1601,nlarray_to_nodelist,Discretization,store_error,type-parse,K1,"(a,1x,i0) Invalid layer number:" +Model/Discretization/Dis.f90,1620,nlarray_to_nodelist,Discretization,store_error,unclassified,?,"(a,1x,i0) MAXBOUND dimension is too small. INCREASE MAXBOUND TO:" +Model/Discretization/Dis.f90,1637,nlarray_to_nodelist,Discretization,store_error,type-parse,K1,"(a,1x,i0) Invalid node number:" +Model/Discretization/Dis2d.f90,256,source_dimensions,Discretization,store_error,range-count,K3,NROW was not specified or was specified incorrectly. +Model/Discretization/Dis2d.f90,261,source_dimensions,Discretization,store_error,range-count,K3,NCOL was not specified or was specified incorrectly. +Model/Discretization/Dis2d.f90,389,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Ensure IDOMAIN array has some values greater than zero. +Model/Discretization/Dis2d.f90,652,nodeu_to_array,Discretization,store_error,not-input,-,"(a,i0,a,i0,a) Program error: nodeu_to_array size of array ( ) is not equal to the discretization dimension ( )" +Model/Discretization/Dis2d.f90,682,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i0,a) Node number ( ) less than 1 or greater than the number of nodes." +Model/Discretization/Dis2d.f90,715,get_nodenumber_idx2,Discretization,store_error,unclassified,?,"('Error in structured-grid cell indices: row = ',i0,', column = ',i0)" +Model/Discretization/Dis2d.f90,724,get_nodenumber_idx2,Discretization,store_error,range-count,K3,Row less than one or greater than nrow +Model/Discretization/Dis2d.f90,726,get_nodenumber_idx2,Discretization,store_error,range-count,K3,Column less than one or greater than ncol +Model/Discretization/Dis2d.f90,732,get_nodenumber_idx2,Discretization,store_error,range-count,K3,"(a,i0,a) Node number ( )less than 1 or greater than nodes." +Model/Discretization/Dis2d.f90,862,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Dis2d.f90,947,nodeu_from_cellid,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Cell number cannot be determined for cellid ( ) and results in a user node number ( ) that is outside of the grid." +Model/Discretization/Dis2d.f90,1376,nlarray_to_nodelist,Discretization,store_error,type-parse,K1,"(a,1x,i0) Invalid layer number:" +Model/Discretization/Dis2d.f90,1395,nlarray_to_nodelist,Discretization,store_error,unclassified,?,"(a,1x,i0) MAXBOUND dimension is too small. INCREASE MAXBOUND TO:" +Model/Discretization/Dis2d.f90,1412,nlarray_to_nodelist,Discretization,store_error,type-parse,K1,"(a,1x,i0) Invalid node number:" +Model/Discretization/Disu.f90,234,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Ensure IDOMAIN array has some values greater than zero. +Model/Discretization/Disu.f90,413,disu_ck,Discretization,store_error,geometry,K10,"('Cell ', i0, ' must have its first connection be itself. Found: ', i0)" +Model/Discretization/Disu.f90,422,disu_ck,Discretization,store_error,geometry,K10,"('Cell ', i0, ' has invalid connection in JA. Found: ', i0)" +Model/Discretization/Disu.f90,438,disu_ck,Discretization,store_error,cellid-grid,K7,"('Invalid idomain value ', i0, ' specified for node ', i0)" +Model/Discretization/Disu.f90,449,disu_ck,Discretization,store_error,unclassified,?, +Model/Discretization/Disu.f90,453,disu_ck,Discretization,store_error,unclassified,?, +Model/Discretization/Disu.f90,463,disu_ck,Discretization,store_error,sign-value,K2,"(a, 1pg15.6) Vertical offset tolerance must be greater than zero. Found" +Model/Discretization/Disu.f90,479,disu_ck,Discretization,store_error,geometry,K10,"('Top elevation (', 1pg15.6, ') for cell ', i0, ' is above bottom elevation (', 1pg15.6, ') for cell ', i0, '. Based on node numbering rules cell ', i0, ' must be below cell ', i0, '.')" +Model/Discretization/Disu.f90,613,disu_ck,Discretization,store_warning,unclassified,?, +Model/Discretization/Disu.f90,619,disu_ck,Discretization,store_warning,unclassified,?, +Model/Discretization/Disu.f90,624,disu_ck,Discretization,store_warning,unclassified,?, +Model/Discretization/Disu.f90,629,disu_ck,Discretization,store_warning,unclassified,?, +Model/Discretization/Disu.f90,799,nodeu_to_array,Discretization,store_error,not-input,-,"(a,i0,a,i0,a) Program error: nodeu_to_array size of array ( ) is not equal to the discretization dimension ( )" +Model/Discretization/Disu.f90,895,source_dimensions,Discretization,store_error,range-count,K3,NODES was not specified or was specified incorrectly. +Model/Discretization/Disu.f90,899,source_dimensions,Discretization,store_error,range-count,K3,NJA was not specified or was specified incorrectly. +Model/Discretization/Disu.f90,1102,source_vertices,Discretization,store_error,block-structure,K1,Required Vertex arrays not found. +Model/Discretization/Disu.f90,1135,define_cellverts,Discretization,store_error,geometry,K10,ICELL2D input sequence violation. +Model/Discretization/Disu.f90,1178,source_cell2d,Discretization,store_error,block-structure,K1,Required cell vertex arrays not found. +Model/Discretization/Disu.f90,1192,source_cell2d,Discretization,store_error,block-structure,K1,Required cell center arrays not found. +Model/Discretization/Disu.f90,1372,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i0,a,i0,a) Node number ( ) is less than 1 or greater than nodes ( )." +Model/Discretization/Disu.f90,1460,connection_vector,Discretization,store_error,geometry,K10,(a) Cannot calculate unit vector components for DISU grid if VERTEX data are not specified +Model/Discretization/Disu.f90,1636,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Disu.f90,1696,nodeu_from_cellid,Discretization,store_error,cellid-grid,K7,"(a,i0,a) Cell number cannot be determined for cellid ( ) and results in a user node number ( ) that is outside of the grid." +Model/Discretization/Disv.f90,283,source_dimensions,Discretization,store_error,range-count,K3,NLAY was not specified or was specified incorrectly. +Model/Discretization/Disv.f90,288,source_dimensions,Discretization,store_error,range-count,K3,NCPL was not specified or was specified incorrectly. +Model/Discretization/Disv.f90,293,source_dimensions,Discretization,store_error,range-count,K3,NVERT was not specified or was specified incorrectly. +Model/Discretization/Disv.f90,419,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Ensure IDOMAIN array has some values greater than zero. +Model/Discretization/Disv.f90,438,grid_finalize,Discretization,store_error,unclassified,?, +Model/Discretization/Disv.f90,539,source_vertices,Discretization,store_error,block-structure,K1,Required Vertex arrays not found. +Model/Discretization/Disv.f90,572,define_cellverts,Discretization,store_error,geometry,K10,ICELL2D input sequence violation. +Model/Discretization/Disv.f90,615,source_cell2d,Discretization,store_error,block-structure,K1,"Required cell vertex array(s) [ICELL2D, NCVERT, ICVERT] not found." +Model/Discretization/Disv.f90,630,source_cell2d,Discretization,store_error,block-structure,K1,Required cell center arrays not found. +Model/Discretization/Disv.f90,673,connect,Discretization,store_error,sign-value,K2,"(a,i0,a) Calculated CELL2D area less than zero for cell ." +Model/Discretization/Disv.f90,679,connect,Discretization,store_error,sign-value,K2,"(a,i0,a) Calculated CELL2D area is zero for cell ." +Model/Discretization/Disv.f90,690,connect,Discretization,store_error,sign-value,K2,"(i0,a) cell(s) have an area less than zero. Calculated cell areas must be greater than zero. Negative areas often mean vertices are not listed in clockwise order." +Model/Discretization/Disv.f90,698,connect,Discretization,store_error,sign-value,K2,"(i0,a) cell(s) have an area equal to zero. Calculated cell areas must be greater than zero. Calculated cell areas equal to zero indicate that the cell is not defined by a valid polygon." +Model/Discretization/Disv.f90,915,nodeu_to_array,Discretization,store_error,not-input,-,"(a,i0,a,i0,a) Program error: nodeu_to_array size of array ( ) is not equal to the discretization dimension ( )." +Model/Discretization/Disv.f90,947,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i0,a,i0,a) Node number ( ) is less than 1 or greater than nodes ( )." +Model/Discretization/Disv.f90,977,get_nodenumber_idx2,Discretization,store_error,unclassified,?,"('Error in disv grid cell indices: layer = ',i0,', node = ',i0)" +Model/Discretization/Disv.f90,1006,get_nodenumber_idx2,Discretization,store_error,range-count,K3,"(a,1x,a,i0,a,i0,a) Node number ( ) is less than 1 or greater than nodes ( )." +Model/Discretization/Disv.f90,1320,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Disv.f90,1405,nodeu_from_cellid,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Cell number cannot be determined for cellid ( ) and results in a user node number ( ) that is outside of the grid." +Model/Discretization/Disv.f90,1824,nlarray_to_nodelist,Discretization,store_error,type-parse,K1,"(a,i0,a) Invalid layer number ( )." +Model/Discretization/Disv.f90,1842,nlarray_to_nodelist,Discretization,store_error,unclassified,?,"(a,i0,a) MAXBOUND dimension is too small. Increase MAXBOUND to ." +Model/Discretization/Disv1d.f90,367,source_dimensions,Discretization,store_error,range-count,K3,NODES was not specified or was specified incorrectly. +Model/Discretization/Disv1d.f90,372,source_dimensions,Discretization,store_warning,block-structure,K1,NVERT was not specified or was specified as zero. The VERTICES and CELL1D blocks will not be read for the DISV1D6 Package in model . +Model/Discretization/Disv1d.f90,448,source_griddata,Discretization,store_error,block-structure,K1,(a) Error in GRIDDATA block: WIDTH not found. +Model/Discretization/Disv1d.f90,453,source_griddata,Discretization,store_error,block-structure,K1,(a) Error in GRIDDATA block: BOTTOM not found. +Model/Discretization/Disv1d.f90,520,source_vertices,Discretization,store_error,block-structure,K1,Required Vertex arrays not found. +Model/Discretization/Disv1d.f90,563,source_cell1d,Discretization,store_error,block-structure,K1,Required cell vertex arrays not found. +Model/Discretization/Disv1d.f90,575,source_cell1d,Discretization,store_error,block-structure,K1,Required fdc array not found. +Model/Discretization/Disv1d.f90,616,define_cellverts,Discretization,store_error,unclassified,?,ICELL1D input sequence violation. +Model/Discretization/Disv1d.f90,737,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Make sure IDOMAIN has some values greater than zero. +Model/Discretization/Disv1d.f90,1008,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i10) Nodenumber less than 1 or greater than nodes:" +Model/Discretization/Disv1d.f90,1085,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Disv2d.f90,296,source_dimensions,Discretization,store_error,range-count,K3,NODES was not specified or was specified incorrectly. +Model/Discretization/Disv2d.f90,301,source_dimensions,Discretization,store_error,range-count,K3,NVERT was not specified or was specified incorrectly. +Model/Discretization/Disv2d.f90,409,grid_finalize,Discretization,store_error,not-found-ref,K7,Model does not have any active nodes. Ensure IDOMAIN array has some values greater than zero. +Model/Discretization/Disv2d.f90,493,source_vertices,Discretization,store_error,block-structure,K1,Required Vertex arrays not found. +Model/Discretization/Disv2d.f90,526,define_cellverts,Discretization,store_error,geometry,K10,ICELL2D input sequence violation. +Model/Discretization/Disv2d.f90,569,source_cell2d,Discretization,store_error,block-structure,K1,"Required cell vertex array(s) [ICELL2D, NCVERT, ICVERT] not found." +Model/Discretization/Disv2d.f90,584,source_cell2d,Discretization,store_error,block-structure,K1,Required cell center arrays not found. +Model/Discretization/Disv2d.f90,624,connect,Discretization,store_error,sign-value,K2,"(a,i0,a) Calculated CELL2D area less than zero for cell ." +Model/Discretization/Disv2d.f90,630,connect,Discretization,store_error,sign-value,K2,"(a,i0,a) Calculated CELL2D area is zero for cell ." +Model/Discretization/Disv2d.f90,641,connect,Discretization,store_error,sign-value,K2,"(i0,a) cell(s) have an area less than zero. Calculated cell areas must be greater than zero. Negative areas often mean vertices are not listed in clockwise order." +Model/Discretization/Disv2d.f90,649,connect,Discretization,store_error,sign-value,K2,"(i0,a) cell(s) have an area equal to zero. Calculated cell areas must be greater than zero. Calculated cell areas equal to zero indicate that the cell is not defined by a valid polygon." +Model/Discretization/Disv2d.f90,856,nodeu_to_array,Discretization,store_error,not-input,-,"(a,i0,a,i0,a) Program error: nodeu_to_array size of array ( ) is not equal to the discretization dimension ( )." +Model/Discretization/Disv2d.f90,887,get_nodenumber_idx1,Discretization,store_error,range-count,K3,"(a,i0,a,i0,a) Node number ( ) is less than 1 or greater than nodes ( )." +Model/Discretization/Disv2d.f90,1176,nodeu_from_string,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Node number in list ( ) is outside of the grid. Cell number cannot be determined in line ' '." +Model/Discretization/Disv2d.f90,1254,nodeu_from_cellid,Discretization,store_error,cellid-grid,K7,"(a,1x,a,i0,a) Cell number cannot be determined for cellid ( ) and results in a user node number ( ) that is outside of the grid." +Model/Geometry/CircularGeometry.f90,140,set_attribute,Geometry,store_error,unclassified,?,"(a,a) Unknown circular geometry attribute:" +Model/Geometry/RectangularGeometry.f90,130,set_attribute,Geometry,store_error,unclassified,?,"(a,a) Unknown rectangular geometry attribute:" +Model/GroundWaterEnergy/gwe-cnd.f90,616,source_griddata,GroundWaterEnergy,store_error,unclassified,?,"(1x,a) If dispersivities are specified then ALH and ATH1 are required." +Model/GroundWaterEnergy/gwe-ctp.f90,160,ctp_rp,GroundWaterEnergy,store_error,unclassified,?,Cell is already a constant : +Model/GroundWaterEnergy/gwe-ctp.f90,224,ctp_ck,GroundWaterEnergy,store_error,unclassified,?, +Model/GroundWaterEnergy/gwe-ctp.f90,477,ctp_bound_value,GroundWaterEnergy,store_error,not-input,-,(3a) Programming error. bound value requested column ''outside range of ncolbnd (1). +Model/GroundWaterEnergy/gwe-esl.f90,166,esl_ck,GroundWaterEnergy,store_warning,unclassified,?, +Model/GroundWaterEnergy/gwe-est.f90,774,source_options,GroundWaterEnergy,store_error,sign-value,K2,(a) Specified value for the heat capacity of water must be greater than 0.0. +Model/GroundWaterEnergy/gwe-est.f90,782,source_options,GroundWaterEnergy,store_error,sign-value,K2,(a) Specified value for the density of water must be greater than 0.0. +Model/GroundWaterEnergy/gwe-est.f90,884,source_data,GroundWaterEnergy,store_error,block-structure,K1,(a) Porosity not specified in griddata block. +Model/GroundWaterEnergy/gwe-est.f90,888,source_data,GroundWaterEnergy,store_error,block-structure,K1,(a) HEAT_CAPACITY_SOLID not specified in griddata block. +Model/GroundWaterEnergy/gwe-est.f90,892,source_data,GroundWaterEnergy,store_error,block-structure,K1,(a) DENSITY_SOLID not specified in griddata block. +Model/GroundWaterEnergy/gwe-est.f90,917,source_data,GroundWaterEnergy,store_error,block-structure,K1,(a) Zero order decay in either the aqueous or solid phase is active but the corresponding zero-order rate coefficient is not specified. Either DECAY_WATER or DECAY_SOLID must be specified in the griddata block. +Model/GroundWaterEnergy/gwe-est.f90,925,source_data,GroundWaterEnergy,store_warning,sign-value,K2,(a) Zero order decay in the aqueous phase has not been activated but DECAY_WATER has been specified. Zero order decay in the aqueous phase will have no affect on simulation results. +Model/GroundWaterEnergy/gwe-est.f90,932,source_data,GroundWaterEnergy,store_warning,sign-value,K2,(a) Zero order decay in the solid phase has not been activated but DECAY_SOLID has been specified. Zero order decay in the solid phase will have no affect on simulation results. +Model/GroundWaterEnergy/gwe-lke.f90,222,find_lke_package,GroundWaterEnergy,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterEnergy/gwe-lke.f90,1186,lke_read_cvs,GroundWaterEnergy,store_error,sign-value,K2,"(4x,a) ****ERROR. Specified thickness used for thermal conduction MUST BE > 0 else divide by zero error occurs" +Model/GroundWaterEnergy/gwe-mwe.f90,215,find_mwe_package,GroundWaterEnergy,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterEnergy/gwe-mwe.f90,992,mwe_read_cvs,GroundWaterEnergy,store_error,sign-value,K2,"(4x,a) ****ERROR. Specified thickness used for thermal conduction MUST BE > 0 else divide by zero error occurs" +Model/GroundWaterEnergy/gwe-sfe.f90,227,find_sfe_package,GroundWaterEnergy,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterEnergy/gwe-sfe.f90,1154,sfe_read_cvs,GroundWaterEnergy,store_error,sign-value,K2,"(4x,a) ****ERROR. Specified thickness used for thermal conduction MUST BE > 0 else divide by zero error occurs" +Model/GroundWaterEnergy/gwe-uze.f90,211,find_uze_package,GroundWaterEnergy,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterEnergy/gwe-uze.f90,1395,area_error,GroundWaterEnergy,store_error,geometry,K10,"(3a) In a GWE model, the area of every UZF object must be equal to that of the host cell. This condition is violated in cell . Check use of AUXMULTNAME option in UZF package." +Model/GroundWaterEnergy/gwe.f90,170,gwe_df,GroundWaterEnergy,store_error,binding-context,K8,"Flow model has boundary packages, but there is no SSM package. The SSM package must be activated." +Model/GroundWaterEnergy/gwe.f90,778,package_create,GroundWaterEnergy,store_error,type-parse,K1,Invalid package type: +Model/GroundWaterEnergy/gwe.f90,789,package_create,GroundWaterEnergy,store_error,duplicate,K6,"(a,a) Cannot create package. Package name already exists:" +Model/GroundWaterFlow/TvBase.f90,255,rp,GroundWaterFlow,store_error,cellid-grid,K7,"(a,i0,a) CELLID at PERIOD row is not in the active model domain." +Model/GroundWaterFlow/gwf-api.f90,166,source_dimensions,GroundWaterFlow,store_error,sign-value,K2,(a) MAXBOUND must be an integer greater than zero. +Model/GroundWaterFlow/gwf-buy.f90,186,buy_ar,GroundWaterFlow,store_error,option-coherence,K5,Error in model . The XT3D option cannot be used with the BUY Package. +Model/GroundWaterFlow/gwf-buy.f90,264,buy_rp,GroundWaterFlow,store_error,binding-context,K8,"('Buoyancy Package does not have a concentration set for species ',i0,'. One or more model names may be specified incorrectly in the PACKAGEDATA block or a gwf-gwt exchange may need to be activated.')" +Model/GroundWaterFlow/gwf-buy.f90,954,source_dimensions,GroundWaterFlow,store_error,sign-value,K2,NRHOSPECIES must be greater than zero. +Model/GroundWaterFlow/gwf-buy.f90,1002,source_packagedata,GroundWaterFlow,store_error,duplicate,K6,"('Invalid value for IRHOSPEC (',i0,') detected in BUY Package. IRHOSPEC must be > 0 and <= NRHOSPECIES, and duplicate values are not allowed.')" +Model/GroundWaterFlow/gwf-buy.f90,1006,source_packagedata,GroundWaterFlow,store_error,duplicate,K6,"('Invalid value for IRHOSPEC (',i0,') detected in BUY Package. IRHOSPEC must be > 0 and <= NRHOSPECIES, and duplicate values are not allowed.')" +Model/GroundWaterFlow/gwf-chd.f90,154,chd_rp,GroundWaterFlow,store_error,unclassified,?,(3a) Cell is already a constant head ( ). +Model/GroundWaterFlow/gwf-chd.f90,219,chd_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-chd.f90,454,chd_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. CHD bound value requested column ''outside range of ncolbnd (1). +Model/GroundWaterFlow/gwf-csub.f90,455,csub_ar,GroundWaterFlow,store_error,sign-value,K2,"(a,3(1x,a)) Specific storage values in the storage (STO) package must be zero in all active cells when using the package." +Model/GroundWaterFlow/gwf-csub.f90,544,csub_ar,GroundWaterFlow,store_error,geometry,K10,"(a,a,g0,a) . Total interbed thickness= ." +Model/GroundWaterFlow/gwf-csub.f90,915,csub_source_dimensions,GroundWaterFlow,store_error,unclassified,?,(a) NINTERBEDS is a required dimension. +Model/GroundWaterFlow/gwf-csub.f90,1242,csub_source_griddata,GroundWaterFlow,store_error,option-coherence,K5,CG_SKE GRIDDATA must be specified. +Model/GroundWaterFlow/gwf-csub.f90,1246,csub_source_griddata,GroundWaterFlow,store_error,option-coherence,K5,CG_THETA GRIDDATA must be specified. +Model/GroundWaterFlow/gwf-csub.f90,1320,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,2(1x,a),1x,i0,a) Interbed number ( ) must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-csub.f90,1348,csub_source_packagedata,GroundWaterFlow,store_error,cellid-grid,K7,(a) CSUB configured for inactive cell: . +Model/GroundWaterFlow/gwf-csub.f90,1373,csub_source_packagedata,GroundWaterFlow,store_error,type-parse,K1,"(a,1x,a,1x,i0,1x,a) Invalid CDELAY for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1388,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,g0,2(a,1x),g0,1x,a,1x,i0,a) THICK ( ) MUST BE greater than or equal to 0 and less than or equal to than for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1396,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,a) FRAC MUST BE greater than 0 and less than or equal to 1 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1411,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,g0,a,1x,a,1x,i0,a) RNB ( ) must be greater than or equal to 1 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1423,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(2(a,1x),i0,a) (SKV,CI) must be greater than or equal to 0 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1432,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(2(a,1x),i0,a) (SKE,RCI) must be greater than or equal to 0 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1448,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,a) THETA must be greater than 0 and less than or equal to 1 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1460,csub_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,a) KV must be greater than 0 for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1474,csub_source_packagedata,GroundWaterFlow,store_error,option-coherence,K5,"(a,1x,i0,a) BOUNDNAME must be specified for packagedata entry ." +Model/GroundWaterFlow/gwf-csub.f90,1610,csub_source_packagedata,GroundWaterFlow,store_error,option-coherence,K5,"(a,i0,a,1x,a) NDELAYCELLS ( ) must be an odd number when using the effective stress formulation." +Model/GroundWaterFlow/gwf-csub.f90,2138,csub_fp,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,a,1x,i0,1x,a,1x,a) Delay interbed cell heads were less than the top of the interbed cell in interbed cells in non-convertible GWF cells for at least one time step during the simulation." +Model/GroundWaterFlow/gwf-csub.f90,2166,csub_da,GroundWaterFlow,store_warning,sign-value,K2,"(a,1x,i0,1x,3a) CSUB negative effective stress was regularized in time step(s); see the model listing file for the number of cells regularized in each time step. This typically occurs in uppermost cells where simulated water levels rise above land surface." +Model/GroundWaterFlow/gwf-csub.f90,2511,csub_ad,GroundWaterFlow,store_error,temporal,K9,"(a,i0,a,1x,a,1x,a,1x,i0,1x,a) Only the first and last ( ) stress period can be steady if interbeds are simulated. Stress period has been defined to be steady state." +Model/GroundWaterFlow/gwf-csub.f90,2828,csub_initialize_tables,GroundWaterFlow,store_warning,not-found-ref,K7,"(a,1x,3a) Package convergence data is requested but delay interbeds are not included in package ( )." +Model/GroundWaterFlow/gwf-csub.f90,3928,csub_cg_chk_stress,GroundWaterFlow,store_error,sign-value,K2,"(a,g0,a,1x,a,1x,a,4(g0,a)) Small to negative effective stress ( ) in cell . ( = - ( - )." +Model/GroundWaterFlow/gwf-csub.f90,3945,csub_cg_chk_stress,GroundWaterFlow,store_error,binding-context,K8,"(a,1x,i0,3(1x,a)) Solution: small to negative effective stress values in cells can be eliminated by increasing storage values and/or adding/modifying stress boundaries to prevent water-levels from exceeding the top of the model." +Model/GroundWaterFlow/gwf-csub.f90,3983,csub_nodelay_update,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,g0,a) Adjusted thickness for no-delay interbed is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,3989,csub_nodelay_update,GroundWaterFlow,store_error,not-input,-,"(a,1x,i0,1x,a,g0,a) Adjusted theta for no-delay interbed is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,4298,csub_set_initial_state,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,a) Negative recompression index calculated for cell ." +Model/GroundWaterFlow/gwf-csub.f90,4345,csub_set_initial_state,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,2(1x,a),a) Negative compression indices calculated for interbed in cell ." +Model/GroundWaterFlow/gwf-csub.f90,4968,csub_cg_update,GroundWaterFlow,store_error,geometry,K10,"(a,1x,a,1x,a,g0,a) Adjusted thickness for cell is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,4974,csub_cg_update,GroundWaterFlow,store_error,not-input,-,"(a,1x,a,1x,a,g0,a) Adjusted theta for cell is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,6535,csub_delay_update,GroundWaterFlow,store_error,geometry,K10,"(2(a,i0),a,g0,a) Adjusted thickness for delay interbed ( ) cell ( ) is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,6541,csub_delay_update,GroundWaterFlow,store_error,not-input,-,"(2(a,i0),a,g0,a) Adjusted theta for delay interbed ( ) cell ( is less than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-csub.f90,6924,csub_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type ' '. +Model/GroundWaterFlow/gwf-csub.f90,7044,csub_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: ' '. +Model/GroundWaterFlow/gwf-csub.f90,7139,csub_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,i0,1x,a,i0,a) interbed cell must be greater than 0 and less than or equal to (specified value is )." +Model/GroundWaterFlow/gwf-csub.f90,7167,csub_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,i0,1x,a,i0,a) interbed cell must be greater than 0 and less than or equal to (specified value is )." +Model/GroundWaterFlow/gwf-csub.f90,7184,csub_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,i0,1x,a,i0,a) interbed cell must be greater than 0 and less than or equal to (specified value is )." +Model/GroundWaterFlow/gwf-csub.f90,7316,csub_process_obsID,GroundWaterFlow,store_error,not-found-ref,K7,(5a) BOUNDNAME (' ') not allowed for CSUB observation type ' '. +Model/GroundWaterFlow/gwf-csub.f90,7341,csub_process_obsID,GroundWaterFlow,store_error,not-found-ref,K7,(5a) BOUNDNAME (' ') not allowed for CSUB observation type ' ' idcellno. +Model/GroundWaterFlow/gwf-drn.f90,214,drn_options,GroundWaterFlow,store_error,unclassified,?,"(a,2(1x,a)) AUXDEPTHNAME was specified as but no AUX variables specified." +Model/GroundWaterFlow/gwf-drn.f90,231,drn_options,GroundWaterFlow,store_error,unclassified,?,"(a,2(1x,a)) AUXDEPTHNAME was specified as but no AUX variable found with this name." +Model/GroundWaterFlow/gwf-drn.f90,316,drn_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-drn.f90,322,drn_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-drn.f90,327,drn_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-drn.f90,662,drn_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. DRN bound value requested column ''outside range of ncolbnd (2). +Model/GroundWaterFlow/gwf-evt.f90,202,evt_source_options,GroundWaterFlow,store_error,option-coherence,K5,READASARRAYS option is not compatible with the SURF_RATE_SPECIFIED option. +Model/GroundWaterFlow/gwf-evt.f90,283,evt_source_dimensions,GroundWaterFlow,store_error,sign-value,K2,"('Error: In EVT, NSEG must be > 0 but is specified as ',i0)" +Model/GroundWaterFlow/gwf-evt.f90,291,evt_source_dimensions,GroundWaterFlow,store_error,sign-value,K2,"In the EVT package, NSEG cannot be greater than 1 when READASARRAYS is used." +Model/GroundWaterFlow/gwf-evt.f90,383,check_pxdp,GroundWaterFlow,store_error,range-count,K3,"('PXDP must be between 0 and 1. Found ', G0, ' for cell ', A)" +Model/GroundWaterFlow/gwf-evt.f90,395,check_pxdp,GroundWaterFlow,store_error,unclassified,?,"('PXDP is not monotonically increasing for cell ', A)" +Model/GroundWaterFlow/gwf-evt.f90,764,evt_bound_value,GroundWaterFlow,store_error,not-input,-,"(a,i0,a) Programming error. EVT bound value requested column ''outside range of ncolbnd ( )." +Model/GroundWaterFlow/gwf-ghb.f90,215,ghb_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-ghb.f90,221,ghb_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-ghb.f90,226,ghb_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-ghb.f90,414,ghb_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. GHB bound value requested column ''outside range of ncolbnd (2). +Model/GroundWaterFlow/gwf-hfb.f90,576,source_dimensions,GroundWaterFlow,store_error,sign-value,K2,(a) MAXHFB must be specified with value greater than zero. +Model/GroundWaterFlow/gwf-hfb.f90,662,source_data,GroundWaterFlow,store_warning,cellid-grid,K7,(a) HFB connection between inactive cell(s) will be excluded: to . +Model/GroundWaterFlow/gwf-hfb.f90,684,source_data,GroundWaterFlow,store_error,unclassified,?,Errors encountered in HFB input file. +Model/GroundWaterFlow/gwf-hfb.f90,746,check_data,GroundWaterFlow,store_error,unclassified,?,"(1x, 'HFB no. ',i0, ' is between two unconnected cells: ', a, ' and ', a)" +Model/GroundWaterFlow/gwf-hfb.f90,761,check_data,GroundWaterFlow,store_error,geometry,K10,"(1x, 'HFB no. ',i0, ' and HFB no. ',i0, ' are both between cells ', a, ' and ', a, '. Only one HFB can be assigned to a cell connection.')" +Model/GroundWaterFlow/gwf-ic.f90,159,source_griddata,GroundWaterFlow,store_error,block-structure,K1,(a) Error in GRIDDATA block: STRT not found. +Model/GroundWaterFlow/gwf-lak.f90,782,lak_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0) nlakeconn MUST BE >= 0 for lake" +Model/GroundWaterFlow/gwf-lak.f90,792,lak_source_packagedata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a) lake has no connections; the IMPLICIT option requires each lake to have at least one GWF connection." +Model/GroundWaterFlow/gwf-lak.f90,916,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0) lakeno MUST BE > 0 and <=" +Model/GroundWaterFlow/gwf-lak.f90,922,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0) iconn FOR LAKE MUST BE > 0 and <=" +Model/GroundWaterFlow/gwf-lak.f90,936,lak_source_connectiondata,GroundWaterFlow,store_error,cellid-grid,K7,"(a,1x,i0,1x,a,1x,i0) INVALID cellid FOR LAKE connection" +Model/GroundWaterFlow/gwf-lak.f90,956,lak_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,a,a) UNKNOWN ctype FOR LAKE connection ( )" +Model/GroundWaterFlow/gwf-lak.f90,985,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a) bedleak FOR LAKE MUST BE >= 0" +Model/GroundWaterFlow/gwf-lak.f90,1003,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,a,1x,a) connection length (connlen) FOR LAKE , CONNECTION NO. , MUST BE > 0 FOR SPECIFIED connection type (ctype)" +Model/GroundWaterFlow/gwf-lak.f90,1017,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a) cell width (connwidth) FOR LAKE HORIZONTAL CONNECTION MUST BE >= 0" +Model/GroundWaterFlow/gwf-lak.f90,1042,lak_source_connectiondata,GroundWaterFlow,store_error,range-count,K3,"(a,1x,i0,1x,a,1x,i0,1x,a) nlakeconn FOR LAKE EMBEDDED CONNECTION EXCEEDS 1." +Model/GroundWaterFlow/gwf-lak.f90,1064,lak_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,1x,a) EMBEDDED LAKE CANNOT COINCIDE WITH VERTICAL CONNECTION IN LAKE ." +Model/GroundWaterFlow/gwf-lak.f90,1092,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a) telev FOR LAKE HORIZONTAL CONNECTION MUST BE >= belev" +Model/GroundWaterFlow/gwf-lak.f90,1097,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.7,1x,a) belev FOR LAKE HORIZONTAL CONNECTION MUST BE >= cell bottom ( )" +Model/GroundWaterFlow/gwf-lak.f90,1102,lak_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.7,1x,a) telev FOR LAKE HORIZONTAL CONNECTION MUST BE <= cell top ( )" +Model/GroundWaterFlow/gwf-lak.f90,1117,lak_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0) NO DATA SPECIFIED FOR LAKE CONNECTION" +Model/GroundWaterFlow/gwf-lak.f90,1122,lak_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,1x,a) DATA FOR LAKE CONNECTION SPECIFIED TIMES" +Model/GroundWaterFlow/gwf-lak.f90,1171,lak_source_tables,GroundWaterFlow,store_error,block-structure,K1,(a) NTABLES > 0 BUT A TABLES BLOCK WAS NOT FOUND IN THE LAK INPUT. +Model/GroundWaterFlow/gwf-lak.f90,1203,lak_source_tables,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0) TABLE DATA ARE SPECIFIED TIMES BUT NTABLES IS SET TO" +Model/GroundWaterFlow/gwf-lak.f90,1321,lak_source_table,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0) LAKE TABLE NROW MUST BE > 0 FOR LAKE" +Model/GroundWaterFlow/gwf-lak.f90,1333,lak_source_table,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0) LAKE TABLE NCOL MUST BE >= FOR LAKE" +Model/GroundWaterFlow/gwf-lak.f90,1387,lak_source_table,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g15.6,1x,a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.6,1x,a) TABLE STAGE ENTRY ( ) FOR LAKE MUST BE GREATER THAN THE PREVIOUS STAGE ENTRY ( )" +Model/GroundWaterFlow/gwf-lak.f90,1397,lak_source_table,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g15.6,1x,a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.6,1x,a) TABLE VOLUME ENTRY ( ) FOR LAKE MUST BE GREATER THAN THE PREVIOUS VOLUME ENTRY ( )" +Model/GroundWaterFlow/gwf-lak.f90,1407,lak_source_table,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g15.6,1x,a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.6,1x,a) TABLE SURFACE AREA ENTRY ( ) FOR LAKE MUST BE GREATER THAN OR EQUAL TO THE PREVIOUS SURFACE AREA ENTRY ( )" +Model/GroundWaterFlow/gwf-lak.f90,1418,lak_source_table,GroundWaterFlow,store_error,binding-context,K8,"(a,1x,i0,1x,a,1x,g15.6,1x,a,1x,i0,1x,a,1x,i0,1x,a,1x,g15.6,1x,a) TABLE EXCHANGE AREA ENTRY ( ) FOR LAKE MUST BE GREATER THAN OR EQUAL TO THE PREVIOUS EXCHANGE AREA ENTRY ( )" +Model/GroundWaterFlow/gwf-lak.f90,1466,lak_source_outlets,GroundWaterFlow,store_error,block-structure,K1,(a) NOUTLETS > 0 BUT AN OUTLETS BLOCK WAS NOT FOUND IN THE LAK INPUT. +Model/GroundWaterFlow/gwf-lak.f90,1520,lak_source_outlets,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0) lakein FOR OUTLET MUST BE > 0 and <=" +Model/GroundWaterFlow/gwf-lak.f90,1528,lak_source_outlets,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0) lakeout FOR OUTLET MUST BE >= 0 and <=" +Model/GroundWaterFlow/gwf-lak.f90,1544,lak_source_outlets,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,a,a) UNKNOWN couttype FOR OUTLET ( )" +Model/GroundWaterFlow/gwf-lak.f90,1611,lak_source_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) NLAKES WAS NOT SPECIFIED OR WAS SPECIFIED INCORRECTLY. +Model/GroundWaterFlow/gwf-lak.f90,1621,lak_source_dimensions,GroundWaterFlow,store_error,range-count,K3,"(a,i0,a,i0,a) DEV_FORCE_LEGACY_LAKE ( ) MUST BE BETWEEN 1 AND NLAKES ( )." +Model/GroundWaterFlow/gwf-lak.f90,2913,lak_check_valid,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,a) LAKENO must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-lak.f90,2921,lak_check_valid,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,a) IOUTLET must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-lak.f90,2946,lak_set_nonneg_value,GroundWaterFlow,store_error,sign-value,K2,"(a,i0,a,g0,a) Lake was assigned a value of . Value must be positive." +Model/GroundWaterFlow/gwf-lak.f90,2967,lak_set_attribute_error,GroundWaterFlow,store_error,unclassified,?,"(a,1x,a,1x,i0,1x,a) for LAKE" +Model/GroundWaterFlow/gwf-lak.f90,3116,lak_source_options,GroundWaterFlow,store_error,geometry,K10,DEV_FORCE_LEGACY and DEV_FORCE_LEGACY_LAKE cannot both be specified. +Model/GroundWaterFlow/gwf-lak.f90,3126,lak_source_options,GroundWaterFlow,store_error,geometry,K10,DEV_FORCE_LEGACY and DEV_FORCE_LEGACY_LAKE cannot both be specified. +Model/GroundWaterFlow/gwf-lak.f90,3242,lak_ar,GroundWaterFlow,store_warning,unclassified,?,"(a) PACKAGE_CONVERGENCE output file "" "" is not written when the IMPLICIT option is active; the lake stage is part of the solver (IMS) convergence check." +Model/GroundWaterFlow/gwf-lak.f90,3268,lak_ar,GroundWaterFlow,store_error,not-found-ref,K7,Time series links for OUTLETS not found in input. +Model/GroundWaterFlow/gwf-lak.f90,3348,lak_rp,GroundWaterFlow,store_error,unknown-keyword,K1,"(a,a) Unknown STATUS keyword: ." +Model/GroundWaterFlow/gwf-lak.f90,5007,lak_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/GroundWaterFlow/gwf-lak.f90,5090,lak_rp_obs,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-lak.f90,5106,lak_rp_obs,GroundWaterFlow,store_error,not-input,-,Programming error in lak_rp_obs +Model/GroundWaterFlow/gwf-lak.f90,5118,lak_rp_obs,GroundWaterFlow,store_error,duplicate,K6,"(a,3(1x,a)) for observation must be assigned to a lake with a unique boundname." +Model/GroundWaterFlow/gwf-lak.f90,5133,lak_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) outlet must be > 0 and <= (specified value is )" +Model/GroundWaterFlow/gwf-lak.f90,5146,lak_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) lake connection number must be > 0 and <= (specified value is )" +Model/GroundWaterFlow/gwf-lak.f90,5157,lak_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) lake must be > 0 and <= (specified value is )" +Model/GroundWaterFlow/gwf-lak.f90,5208,lak_process_obsID,GroundWaterFlow,store_error,option-coherence,K5,"(a,1x,a,a,1x,a,1x,a) For observation type , ID given as an integer and not as boundname, but ID2 (iconn) is missing. Either change ID to valid boundname or supply valid entry for ID2." +Model/GroundWaterFlow/gwf-lak.f90,6455,lak_activate_density,GroundWaterFlow,store_error,option-coherence,K5,(a) The IMPLICIT option cannot be used with the BUY package until the implicit formulation includes density terms. Remove the IMPLICIT option from LAK package to simulate density. +Model/GroundWaterFlow/gwf-maw.f90,805,maw_source_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) NMAWWELLS was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-maw.f90,924,maw_source_packagedata,GroundWaterFlow,store_error,not-found-ref,K7,Time series links for STRT not found in input. +Model/GroundWaterFlow/gwf-maw.f90,952,maw_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a) Radius for well must be greater than zero." +Model/GroundWaterFlow/gwf-maw.f90,972,maw_source_packagedata,GroundWaterFlow,store_warning,unclassified,?,"(a,a,a,a,a,a) CONDEQN in ' ' should be corrected from ' ' to 'THIEM'." +Model/GroundWaterFlow/gwf-maw.f90,983,maw_source_packagedata,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a) CONDEQN for well must be 'CUMULATIVE', 'THIEM', 'MEAN', or 'SKIN'." +Model/GroundWaterFlow/gwf-maw.f90,991,maw_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a) NGWFNODES for well must be greater than zero." +Model/GroundWaterFlow/gwf-maw.f90,1104,maw_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,a) IMAW must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-maw.f90,1112,maw_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,a) JCONN for well must be greater than 1 and less than or equal to ." +Model/GroundWaterFlow/gwf-maw.f90,1169,maw_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a,g0,a) Screen bottom for maw well connection ( ) is less than the well bottom ( )." +Model/GroundWaterFlow/gwf-maw.f90,1193,maw_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a,g0,a) Screen radius for maw well connection ( ) is less than or equal to the well radius ( )." +Model/GroundWaterFlow/gwf-maw.f90,1209,maw_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,a) No data specified for maw well connection ." +Model/GroundWaterFlow/gwf-maw.f90,1214,maw_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,1x,a) Data for maw well connection specified times." +Model/GroundWaterFlow/gwf-maw.f90,1234,maw_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,3(1x,a)) Only one connection can be specified for maw well connection to gwf cell unless the mean condeqn is specified." +Model/GroundWaterFlow/gwf-maw.f90,1250,maw_source_connectiondata,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,a,1x,a,1x,i0,1x,a) The screen tops in multi-aquifer well package were reset to the top of the connected cell times." +Model/GroundWaterFlow/gwf-maw.f90,1257,maw_source_connectiondata,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,a,1x,a,1x,i0,1x,a) The screen bottoms in multi-aquifer well package were reset to the bottom of the connected cell times." +Model/GroundWaterFlow/gwf-maw.f90,1264,maw_source_connectiondata,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,a,1x,a,1x,i0,1x,a) The well bottoms in multi-aquifer well package were reset to the bottom of the connected cell times." +Model/GroundWaterFlow/gwf-maw.f90,1311,maw_source_angledata,GroundWaterFlow,store_error,block-structure,K1,An ANGLEDATA block was specified but the NON_VERTICAL_WELLS option was not specified in the OPTIONS block. +Model/GroundWaterFlow/gwf-maw.f90,1339,maw_source_angledata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,a) IFNO must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-maw.f90,1349,maw_source_angledata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,a) ICON for well must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-maw.f90,1360,maw_source_angledata,GroundWaterFlow,store_error,duplicate,K6,"(a,1x,i0,1x,a,1x,i0,1x,a) ANGLEDATA for maw well connection is specified more than once." +Model/GroundWaterFlow/gwf-maw.f90,1374,maw_source_angledata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a) ANGLE for maw well connection ( ) must be greater than or equal to 0.0 and less than or equal to 90.0 degrees." +Model/GroundWaterFlow/gwf-maw.f90,1396,maw_source_angledata,GroundWaterFlow,store_error,block-structure,K1,"(a,1x,i0,1x,a,1x,i0,1x,a) The screen top must be greater than the screen bottom for maw well connection listed in the ANGLEDATA block." +Model/GroundWaterFlow/gwf-maw.f90,1425,maw_source_angledata,GroundWaterFlow,store_error,block-structure,K1,"(a,1x,i0,1x,a,1x,i0,1x,a) A connection length must be specified for the (near) horizontal maw well connection listed in the ANGLEDATA block." +Model/GroundWaterFlow/gwf-maw.f90,1434,maw_source_angledata,GroundWaterFlow,store_error,block-structure,K1,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a) The calculated in-cell screen length for maw well connection ( ) is not greater than zero. Specify a connection length in the ANGLEDATA block." +Model/GroundWaterFlow/gwf-maw.f90,1473,maw_source_angledata,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a,g0,a) The horizontal distance spanned by maw well connection ( ) is greater than the of the connected cell ( ), so the screen extends beyond the cell it is connected to. Reduce ANGLE or CONN_LENGTH, or specify a separate connection to each cell the connection passes thr" +Model/GroundWaterFlow/gwf-maw.f90,1488,maw_source_angledata,GroundWaterFlow,store_warning,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a) The (near) horizontal maw well connection uses a radial conductance equation (THIEM, SKIN, or CUMULATIVE). The calculated conductance is an approximation for horizontal connections; the MEAN conductance equation is recommended for horizontal connections." +Model/GroundWaterFlow/gwf-maw.f90,1505,maw_source_angledata,GroundWaterFlow,store_warning,not-found-ref,K7,"(a,1x,i0,1x,a,1x,i0,1x,a,g0,a) The vertical screen extent (SCRN_TOP - SCRN_BOT) for the (near) horizontal maw well connection is not equal to the well diameter (2 * RADIUS = ). The vertical screen extent is used to determine the saturation of a connection and should equal the well diameter for a hor" +Model/GroundWaterFlow/gwf-maw.f90,1526,maw_source_angledata,GroundWaterFlow,store_warning,block-structure,K1,(a) The NON_VERTICAL_WELLS option was specified but an ANGLEDATA block was not found. All multi-aquifer well connections will be treated as vertical. +Model/GroundWaterFlow/gwf-maw.f90,1890,maw_set_attribute_error,GroundWaterFlow,store_error,unclassified,?,"(a,1x,a,1x,i0,1x,a) for MAW well" +Model/GroundWaterFlow/gwf-maw.f90,2184,maw_rp,GroundWaterFlow,store_error,sign-value,K2,"(2(a,1x),i0,a) IMAW must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-maw.f90,2206,maw_rp,GroundWaterFlow,store_error,unknown-keyword,K1,(2a) Unknown maw status keyword: ' '. +Model/GroundWaterFlow/gwf-maw.f90,2229,maw_rp,GroundWaterFlow,store_error,type-parse,K1,Could not read HEAD_LIMIT value. +Model/GroundWaterFlow/gwf-maw.f90,2247,maw_rp,GroundWaterFlow,store_warning,unclassified,?, +Model/GroundWaterFlow/gwf-maw.f90,2339,maw_rp,GroundWaterFlow,store_warning,not-input,-,"(a,a,a,1x,a,1x,a) HEAD_LIMIT in ' ' was below the well bottom for one or more multi-aquifer well(s). This may result in convergence failures for some models." +Model/GroundWaterFlow/gwf-maw.f90,3161,maw_cc,GroundWaterFlow,store_warning,geometry,K10,"('MAW well ', a, ' requests an extraction rate (', g0.5, ') larger than the maximum rate (', g0.5, ') the aquifer can supply with the well head at its bottom. This may be preventing convergence. Consider reducing the requested rate, applying or widening RATE_SCALING, or reviewing the connection cond" +Model/GroundWaterFlow/gwf-maw.f90,3793,maw_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/GroundWaterFlow/gwf-maw.f90,3868,maw_rp_obs,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-maw.f90,3883,maw_rp_obs,GroundWaterFlow,store_error,not-input,-,Programming error in maw_rp_obs +Model/GroundWaterFlow/gwf-maw.f90,3895,maw_rp_obs,GroundWaterFlow,store_error,duplicate,K6,"(a,3(1x,a)) for observation must be assigned to a multi-aquifer well with a unique boundname." +Model/GroundWaterFlow/gwf-maw.f90,3912,maw_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(3(a,1x),i0,1x,a,i0,a) multi-aquifer well connection number must be greater than 0 and less than (specified value is )." +Model/GroundWaterFlow/gwf-maw.f90,3924,maw_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(3(a,1x),i0,1x,a,i0,a) multi-aquifer well must be greater than 0 and less than or equal to (specified value is )." +Model/GroundWaterFlow/gwf-maw.f90,3976,maw_process_obsID,GroundWaterFlow,store_error,option-coherence,K5,"(a,1x,a,a,1x,a,1x,a) For observation type , ID given as an integer and not as boundname, but ID2 (icon) is missing. Either change ID to valid boundname or supply valid entry for ID2." +Model/GroundWaterFlow/gwf-maw.f90,4164,maw_calculate_satcond,GroundWaterFlow,store_error,range-count,K3,"(a,g0,a,1x,i0,1x,a,1x,i0,a,4(1x,a)) Invalid calculated transmissivity contrast ( ) for maw well connection . This happens when the skin transmissivity equals or exceeds the aquifer transmissivity. Consider decreasing HK_SKIN for the connection or using the CUMULATIVE or MEAN conductance equations." +Model/GroundWaterFlow/gwf-maw.f90,4199,maw_calculate_satcond,GroundWaterFlow,store_error,range-count,K3,"(a,g0,a,1x,i0,1x,a,1x,i0,a,4(1x,a)) Invalid calculated negative conductance ( ) for maw well connection . this happens when the skin transmissivity equals or exceeds the aquifer transmissivity. consider decreasing hk_skin for the connection or using the mean conductance equation." +Model/GroundWaterFlow/gwf-mvr.f90,323,mvr_rp,GroundWaterFlow,store_error,enum,K2,"('Error. Looking for BEGIN PERIOD iper. Found ', a, ' instead.')" +Model/GroundWaterFlow/gwf-mvr.f90,373,mvr_rp,GroundWaterFlow,store_error,block-structure,K1,"(a,a,a) Provider not listed in packages block." +Model/GroundWaterFlow/gwf-mvr.f90,379,mvr_rp,GroundWaterFlow,store_error,block-structure,K1,"(a,a,a) Receiver not listed in packages block." +Model/GroundWaterFlow/gwf-mvr.f90,727,read_options,GroundWaterFlow,store_error,unclassified,?,OPTIONAL BUDGET KEYWORD MUST BE FOLLOWED BY FILEOUT +Model/GroundWaterFlow/gwf-mvr.f90,741,read_options,GroundWaterFlow,store_error,unclassified,?,OPTIONAL BUDGETCSV KEYWORD MUST BE FOLLOWED BY FILEOUT +Model/GroundWaterFlow/gwf-mvr.f90,760,read_options,GroundWaterFlow,store_error,binding-context,K8,"(a,a) MODELNAMES cannot be specified unless the mover package is for an exchange." +Model/GroundWaterFlow/gwf-mvr.f90,765,read_options,GroundWaterFlow,store_error,unknown-keyword,K1,"(a,a) Unknown MVR option:" +Model/GroundWaterFlow/gwf-mvr.f90,789,check_options,GroundWaterFlow,store_error,binding-context,K8,"(a,a) MODELNAMES cannot be specified unless the mover package is for an exchange." +Model/GroundWaterFlow/gwf-mvr.f90,798,check_options,GroundWaterFlow,store_error,binding-context,K8,"(a,a) MODELNAMES option must be specified because mover package is for an exchange." +Model/GroundWaterFlow/gwf-mvr.f90,839,read_dimensions,GroundWaterFlow,store_error,unclassified,?,"(a,a) Unknown MVR dimension:" +Model/GroundWaterFlow/gwf-mvr.f90,845,read_dimensions,GroundWaterFlow,store_error,block-structure,K1,Required DIMENSIONS block not found. +Model/GroundWaterFlow/gwf-mvr.f90,861,read_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) MAXMVR was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-mvr.f90,867,read_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) MAXPACKAGES was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-mvr.f90,903,read_packages,GroundWaterFlow,store_error,unclassified,?,ERROR. MAXPACKAGES NOT SET LARGE ENOUGH. +Model/GroundWaterFlow/gwf-mvr.f90,917,read_packages,GroundWaterFlow,store_error,block-structure,K1,"(a,a,a) Package listed more than once in PACKAGES block." +Model/GroundWaterFlow/gwf-mvr.f90,926,read_packages,GroundWaterFlow,store_error,block-structure,K1,ERROR. REQUIRED PACKAGES BLOCK NOT FOUND. +Model/GroundWaterFlow/gwf-mvr.f90,935,read_packages,GroundWaterFlow,store_error,range-count,K3,"(a, i0, a, i0, a) ERROR. NUMBER OF PACKAGES ( ) DOES NOT EQUAL MAXPACKAGES ( )." +Model/GroundWaterFlow/gwf-mvr.f90,963,check_packages,GroundWaterFlow,store_error,block-structure,K1,"(a, a, a) ERROR. MODEL AND PACKAGE "" "" DOES NOT HAVE MOVER SPECIFIED IN OPTIONS BLOCK." +Model/GroundWaterFlow/gwf-npf.f90,280,npf_df,GroundWaterFlow,store_error,option-coherence,K5,Error in model . The XT3D option cannot be used with the GNC Package. +Model/GroundWaterFlow/gwf-npf.f90,391,npf_ar,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a) ANGLDEGX values in the DISU Package are inconsistent for cell faces (see the warnings written after the DISU Package input in the model listing file). ANGLDEGX must be correct because it is required input for the NPF Package when XT3D, K22, or SAVE_SPECIFIC_DISCHARGE is specified." +Model/GroundWaterFlow/gwf-npf.f90,1667,source_options,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized input value for ALTERNATIVE_CELL_AVERAGING option. +Model/GroundWaterFlow/gwf-npf.f90,1715,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. NEWTON OPTION CANNOT BE USED WITH PERCHED OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1720,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. NEWTON OPTION CANNOT BE USED WITH VARIABLECV OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1725,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. NEWTON OPTION CANNOT BE USED WITH REWET OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1733,check_options,GroundWaterFlow,store_warning,option-coherence,K5,(a) HIGHEST_CELL_SATURATION option cannot be used when NEWTON option in not specified. Resetting HIGHEST_CELL_SATURATION option to off. +Model/GroundWaterFlow/gwf-npf.f90,1742,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. ALTERNATIVE_CELL_AVERAGING OPTION CANNOT BE USED WITH XT3D OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1747,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. THICKSTRT OPTION CANNOT BE USED WITH XT3D OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1752,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. PERCHED OPTION CANNOT BE USED WITH XT3D OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1757,check_options,GroundWaterFlow,store_error,option-coherence,K5,(a) ERROR IN NPF OPTIONS. VARIABLECV OPTION CANNOT BE USED WITH XT3D OPTION. +Model/GroundWaterFlow/gwf-npf.f90,1856,source_griddata,GroundWaterFlow,store_error,block-structure,K1,(a) Error in GRIDDATA block: ICELLTYPE not found. +Model/GroundWaterFlow/gwf-npf.f90,1862,source_griddata,GroundWaterFlow,store_error,block-structure,K1,(a) Error in GRIDDATA block: K not found. +Model/GroundWaterFlow/gwf-npf.f90,1868,source_griddata,GroundWaterFlow,store_error,option-coherence,K5,(a) K33OVERK option specified but K33 not specified. +Model/GroundWaterFlow/gwf-npf.f90,1874,source_griddata,GroundWaterFlow,store_error,option-coherence,K5,(a) K22OVERK option specified but K22 not specified. +Model/GroundWaterFlow/gwf-npf.f90,1942,prepcheck,GroundWaterFlow,store_error,sign-value,K2,"(1x, 'Hydraulic property ',a,' is <= 0 for cell ',a, ' ', 1pg15.6)" +Model/GroundWaterFlow/gwf-npf.f90,1948,prepcheck,GroundWaterFlow,store_error,unclassified,?,"(1x, '... ', i0,' additional errors not shown for ',a)" +Model/GroundWaterFlow/gwf-npf.f90,1964,prepcheck,GroundWaterFlow,store_error,sign-value,K2,"(1x, 'Hydraulic property ',a,' is <= 0 for cell ',a, ' ', 1pg15.6)" +Model/GroundWaterFlow/gwf-npf.f90,1970,prepcheck,GroundWaterFlow,store_error,unclassified,?,"(1x, '... ', i0,' additional errors not shown for ',a)" +Model/GroundWaterFlow/gwf-npf.f90,1981,prepcheck,GroundWaterFlow,store_error,geometry,K10,"(a) Error. ANGLDEGX not provided in discretization file, but K22 was specified." +Model/GroundWaterFlow/gwf-npf.f90,1994,prepcheck,GroundWaterFlow,store_error,sign-value,K2,"(1x, 'Hydraulic property ',a,' is <= 0 for cell ',a, ' ', 1pg15.6)" +Model/GroundWaterFlow/gwf-npf.f90,2000,prepcheck,GroundWaterFlow,store_error,unclassified,?,"(1x, '... ', i0,' additional errors not shown for ',a)" +Model/GroundWaterFlow/gwf-npf.f90,2009,prepcheck,GroundWaterFlow,store_error,block-structure,K1,"(a, a, a) Error in GRIDDATA block: not found." +Model/GroundWaterFlow/gwf-npf.f90,2032,prepcheck,GroundWaterFlow,store_error,option-coherence,K5,(a) ANGLE2 SPECIFIED BUT NOT ANGLE1. ANGLE2 REQUIRES ANGLE1. +Model/GroundWaterFlow/gwf-npf.f90,2037,prepcheck,GroundWaterFlow,store_error,geometry,K10,(a) ANGLE2 SPECIFIED BUT NOT ANGLE3. SPECIFY BOTH OR NEITHER ONE. +Model/GroundWaterFlow/gwf-npf.f90,2047,prepcheck,GroundWaterFlow,store_error,option-coherence,K5,(a) ANGLE3 SPECIFIED BUT NOT ANGLE1. ANGLE3 REQUIRES ANGLE1. +Model/GroundWaterFlow/gwf-npf.f90,2052,prepcheck,GroundWaterFlow,store_error,geometry,K10,(a) ANGLE3 SPECIFIED BUT NOT ANGLE2. SPECIFY BOTH OR NEITHER ONE. +Model/GroundWaterFlow/gwf-npf.f90,2195,preprocess_input,GroundWaterFlow,store_error,sign-value,K2,"(1X,/1X,'Negative cell thickness at (layer,row,col)', I4,',',I5,',',I5)" +Model/GroundWaterFlow/gwf-npf.f90,2197,preprocess_input,GroundWaterFlow,store_error,geometry,K10,"(1X,'Initial head, bottom elevation:',1P,2G13.5)" +Model/GroundWaterFlow/gwf-npf.f90,2206,preprocess_input,GroundWaterFlow,store_error,sign-value,K2,"(1X,/1X,'Negative cell thickness at (layer,row,col)', I4,',',I5,',',I5)" +Model/GroundWaterFlow/gwf-npf.f90,2208,preprocess_input,GroundWaterFlow,store_error,geometry,K10,"(1X,'Top elevation, bottom elevation:',1P,2G13.5)" +Model/GroundWaterFlow/gwf-npf.f90,2448,sgwf_npf_wetdry,GroundWaterFlow,store_error,sign-value,K2,"(1X,/1X,'Negative cell thickness at (layer,row,col)', I4,',',I5,',',I5)" +Model/GroundWaterFlow/gwf-npf.f90,2450,sgwf_npf_wetdry,GroundWaterFlow,store_error,geometry,K10,"(1X,'Top elevation, bottom elevation:',1P,2G13.5)" +Model/GroundWaterFlow/gwf-npf.f90,2466,sgwf_npf_wetdry,GroundWaterFlow,store_error,not-input,-,CONSTANT-HEAD CELL WENT DRY -- SIMULATION ABORTED +Model/GroundWaterFlow/gwf-npf.f90,2468,sgwf_npf_wetdry,GroundWaterFlow,store_error,geometry,K10,"(1X,'Top elevation, bottom elevation, thickness:',1P,3G13.5)" +Model/GroundWaterFlow/gwf-npf.f90,2471,sgwf_npf_wetdry,GroundWaterFlow,store_error,cellid-grid,K7,"(1X,'CELLID=',a,' ITERATION=',I0,' TIME STEP=',I0,' STRESS PERIOD=',I0)" +Model/GroundWaterFlow/gwf-npf.f90,2730,calc_spdis,GroundWaterFlow,store_error,geometry,K10,Error. ANGLDEGX not provided in discretization file. ANGLDEGX required for calculation of specific discharge. +Model/GroundWaterFlow/gwf-obs.f90,128,gwf_obs_bd,GroundWaterFlow,store_error,unknown-keyword,K1,Error: Unrecognized observation type: +Model/GroundWaterFlow/gwf-obs.f90,202,gwf_process_head_drawdown_obs_id,GroundWaterFlow,store_error,not-input,-,Error reading data from ID string +Model/GroundWaterFlow/gwf-obs.f90,235,gwf_process_intercell_obs_id,GroundWaterFlow,store_error,not-input,-,Error reading data from ID string: +Model/GroundWaterFlow/gwf-obs.f90,246,gwf_process_intercell_obs_id,GroundWaterFlow,store_error,not-input,-,Error reading data from ID string: +Model/GroundWaterFlow/gwf-obs.f90,253,gwf_process_intercell_obs_id,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-rch.f90,414,rch_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. RCH bound value requested column ''outside range of ncolbnd (1). +Model/GroundWaterFlow/gwf-riv.f90,233,riv_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-riv.f90,237,riv_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-riv.f90,241,riv_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-riv.f90,247,riv_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-riv.f90,252,riv_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-riv.f90,450,riv_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. RIV bound value requested column ''outside range of ncolbnd (3). +Model/GroundWaterFlow/gwf-sfr.f90,805,sfr_source_options,GroundWaterFlow,store_error,sign-value,K2,"(a,g0,a) ATS_COURANT SPECIFIED TO BE ' ' BUT MUST BE GREATER THAN ZERO" +Model/GroundWaterFlow/gwf-sfr.f90,909,sfr_source_options,GroundWaterFlow,store_error,sign-value,K2,"(a,g0,a) STORAGE_WEIGHT SPECIFIED TO BE ' ' BUT CANNOT BE LESS THAN 0.5 OR GREATER THAN 1.0" +Model/GroundWaterFlow/gwf-sfr.f90,966,sfr_source_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) NREACHES was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-sfr.f90,1062,sfr_source_packagedata,GroundWaterFlow,store_error,not-found-ref,K7,"Time series links for RHK, MAN and USTRF not found in input." +Model/GroundWaterFlow/gwf-sfr.f90,1112,sfr_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,i0,a) NCON for reach must be greater than or equal to 0 ( )." +Model/GroundWaterFlow/gwf-sfr.f90,1151,sfr_source_packagedata,GroundWaterFlow,store_warning,sign-value,K2,"(a,1x,a,1x,a,1x,i0,1x,a) SFR Package has reach(es) with zero connections." +Model/GroundWaterFlow/gwf-sfr.f90,1346,sfr_source_connectiondata,GroundWaterFlow,store_error,geometry,K10,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,a) Reach connectiondata specifies connections but NCON for this reach in packagedata is ." +Model/GroundWaterFlow/gwf-sfr.f90,1359,sfr_source_connectiondata,GroundWaterFlow,store_error,sign-value,K2,Missing or zero connection reach in connectiondata. +Model/GroundWaterFlow/gwf-sfr.f90,1365,sfr_source_connectiondata,GroundWaterFlow,store_error,range-count,K3,Reach number exceeds NREACHES in connectiondata. +Model/GroundWaterFlow/gwf-sfr.f90,1389,sfr_source_connectiondata,GroundWaterFlow,store_error,block-structure,K1,"(a,3(1x,a)) Could not fill package IA and JA connection data. Check connectivity data in connectiondata block." +Model/GroundWaterFlow/gwf-sfr.f90,1442,sfr_source_connectiondata,GroundWaterFlow,store_warning,binding-context,K8,(a) PACKAGE ( ) cannot calculate a Directed Asyclic Graph for reach connectivity because of circular dependency. Using the reach number for solution ordering. +Model/GroundWaterFlow/gwf-sfr.f90,1503,sfr_source_diversions,GroundWaterFlow,store_error,block-structure,K1,REQUIRED DIVERSIONS BLOCK NOT FOUND. +Model/GroundWaterFlow/gwf-sfr.f90,1513,sfr_source_diversions,GroundWaterFlow,store_error,block-structure,K1,"(a,1x,a) A diversions block should not be specified if diversions are not specified." +Model/GroundWaterFlow/gwf-sfr.f90,1541,sfr_source_diversions,GroundWaterFlow,store_error,unclassified,?,"(a,i0,a) Reach number should be between 1 and ." +Model/GroundWaterFlow/gwf-sfr.f90,1547,sfr_source_diversions,GroundWaterFlow,store_error,option-coherence,K5,"(a,i0) Diversions cannot be specified for reach" +Model/GroundWaterFlow/gwf-sfr.f90,1555,sfr_source_diversions,GroundWaterFlow,store_error,unclassified,?,"(a,i0,a,i0,a) Reach diversion number should be between 1 and ." +Model/GroundWaterFlow/gwf-sfr.f90,1565,sfr_source_diversions,GroundWaterFlow,store_error,unclassified,?,"(a,i0,a) Diversion target reach number should be between 1 and ." +Model/GroundWaterFlow/gwf-sfr.f90,1574,sfr_source_diversions,GroundWaterFlow,store_error,type-parse,K1,Invalid cprior type . +Model/GroundWaterFlow/gwf-sfr.f90,1588,sfr_source_diversions,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0) No data specified for reach diversion" +Model/GroundWaterFlow/gwf-sfr.f90,1593,sfr_source_diversions,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,1x,a) Data for reach diversion specified times" +Model/GroundWaterFlow/gwf-sfr.f90,1658,sfr_source_initialstages,GroundWaterFlow,store_error,geometry,K10,"(a,g0,a,1x,i0,1x,a,g0,a) Initial stage ( ) for reach is less than the reach top ( )." +Model/GroundWaterFlow/gwf-sfr.f90,1730,sfr_ar,GroundWaterFlow,store_error,option-coherence,K5,(a) ATS_COURANT OPTION REQUIRES STORAGE OPTION TO BE ACTIVE +Model/GroundWaterFlow/gwf-sfr.f90,1742,sfr_ar,GroundWaterFlow,store_warning,block-structure,K1,"(a) ATS_COURANT IS SPECIFIED IN THE SFR OPTIONS BLOCK BUT THE ATS PACKAGE IS NOT ACTIVE IN TDIS, SO THE TIME STEP IS NOT ADAPTED TO THE COURANT NUMBER. IF THE COURANT NUMBER EXCEEDS 1 THE EXPLICIT KINEMATIC-WAVE ROUTING MAY OSCILLATE WHILE STILL CLOSING THE BUDGET. ACTIVATE THE ATS PACKAGE OR USE A " +Model/GroundWaterFlow/gwf-sfr.f90,1855,sfr_rp,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,a) Reach number (RNO) must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-sfr.f90,1875,sfr_rp,GroundWaterFlow,store_error,unknown-keyword,K1,(2a) Unknown sfr status keyword: +Model/GroundWaterFlow/gwf-sfr.f90,1889,sfr_rp,GroundWaterFlow,store_error,option-coherence,K5,diversions cannot be specified for reach +Model/GroundWaterFlow/gwf-sfr.f90,1898,sfr_rp,GroundWaterFlow,store_error,unclassified,?,diversion number should be between 1 and . +Model/GroundWaterFlow/gwf-sfr.f90,1908,sfr_rp,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,a) cprior is type FRACTION for diversion no. , but divflow not within the range 0.0 to 1.0" +Model/GroundWaterFlow/gwf-sfr.f90,3274,sfr_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/GroundWaterFlow/gwf-sfr.f90,3334,sfr_rp_obs,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-sfr.f90,3338,sfr_rp_obs,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-sfr.f90,3345,sfr_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) reach must be greater than 0 and less than or equal to (specified value is )" +Model/GroundWaterFlow/gwf-sfr.f90,3351,sfr_rp_obs,GroundWaterFlow,store_error,not-input,-,Programming error in sfr_rp_obs +Model/GroundWaterFlow/gwf-sfr.f90,3369,sfr_rp_obs,GroundWaterFlow,store_error,duplicate,K6,"(a,3(1x,a)) for observation must be assigned to a reach with a unique boundname." +Model/GroundWaterFlow/gwf-sfr.f90,3382,sfr_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) reach must be greater than 0 and less than or equal to (specified value is )" +Model/GroundWaterFlow/gwf-sfr.f90,4170,sfr_check_conversion,GroundWaterFlow,store_error,geometry,K10,"('SFR (',a,') UNIT_CONVERSION SPECIFIED VALUE (',g0,') AND', 1x,'LENGTH_CONVERSION OR TIME_CONVERSION SPECIFIED.')" +Model/GroundWaterFlow/gwf-sfr.f90,4269,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach length must be greater than 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4274,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach width must be greater than 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4279,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach slope must be greater than 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4291,sfr_check_reaches,GroundWaterFlow,store_error,unclassified,?,= ). +Model/GroundWaterFlow/gwf-sfr.f90,4295,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach hk must be greater than or equal to 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4302,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach Manning's roughness coefficient must be greater than 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4308,sfr_check_reaches,GroundWaterFlow,store_error,sign-value,K2,Reach upstream fraction must be greater than or equal to 0.0. +Model/GroundWaterFlow/gwf-sfr.f90,4434,sfr_check_connections,GroundWaterFlow,store_error,not-found-ref,K7,Reach is connected to reach but reach is not connected to reach . +Model/GroundWaterFlow/gwf-sfr.f90,4474,sfr_check_connections,GroundWaterFlow,store_error,not-found-ref,K7,Reach is connected to reach but streamflow from reach to reach is not permitted. +Model/GroundWaterFlow/gwf-sfr.f90,4508,sfr_check_connections,GroundWaterFlow,store_error,not-found-ref,K7,Reach downstream connected reach is reach but reach is not the upstream connected reach for reach . +Model/GroundWaterFlow/gwf-sfr.f90,4675,sfr_check_diversions,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-sfr.f90,4692,sfr_check_diversions,GroundWaterFlow,store_error,not-found-ref,K7,Reach is not a upstream reach for reach as a result diversion from reach to reach is not possible. Check reach connectivity. +Model/GroundWaterFlow/gwf-sfr.f90,4875,sfr_check_ustrf,GroundWaterFlow,store_error,not-found-ref,K7,"(a,2(1x,i0,1x,a),1x,a,g0,a,2(1x,a)) Reach is connected to reach by a diversion but the upstream fraction is not equal to zero ( ). Check package diversion and package data." +Model/GroundWaterFlow/gwf-sfr.f90,4882,sfr_check_ustrf,GroundWaterFlow,store_warning,unclassified,?,"(a,3(1x,a)) A warning instead of an error is issued because the reach is only connected to the diversion reach in the downstream direction." +Model/GroundWaterFlow/gwf-sfr.f90,4961,sfr_check_ustrf,GroundWaterFlow,store_error,not-found-ref,K7,"(a,1x,i0,1x,a,g0,a,3(1x,a)) Upstream fractions for reach is not equal to one ( ). Check package reach connectivity and package data." +Model/GroundWaterFlow/gwf-sto.f90,215,sto_rp,GroundWaterFlow,store_error,unknown-keyword,K1,"(a,a) Unknown STORAGE data tag:" +Model/GroundWaterFlow/gwf-sto.f90,281,sto_fc,GroundWaterFlow,store_error,sign-value,K2,"('DETECTED TIME STEP LENGTH OF ZERO. GWF STORAGE PACKAGE CANNOT BE ', 'USED UNLESS DELT IS NON-ZERO.')" +Model/GroundWaterFlow/gwf-sto.f90,1126,source_data,GroundWaterFlow,store_error,block-structure,K1,"(a, a, a) Error in GRIDDATA block: not found." +Model/GroundWaterFlow/gwf-sto.f90,1135,source_data,GroundWaterFlow,store_error,block-structure,K1,"(a, a, a) Error in GRIDDATA block: not found." +Model/GroundWaterFlow/gwf-sto.f90,1145,source_data,GroundWaterFlow,store_error,block-structure,K1,"(a, a, a) Error in GRIDDATA block: not found." +Model/GroundWaterFlow/gwf-sto.f90,1160,source_data,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,g0,1x,a) Error in SS DATA: SS value in cell is less than zero ( )." +Model/GroundWaterFlow/gwf-sto.f90,1168,source_data,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,g0,1x,a) Error in SY DATA: SY value in cell is less than zero ( )." +Model/GroundWaterFlow/gwf-tvk.f90,126,tvk_apply_row_changes,GroundWaterFlow,store_error,cellid-grid,K7,"(a,2(1x,a)) CELLID is not in the active model domain." +Model/GroundWaterFlow/gwf-tvk.f90,144,tvk_apply_row_changes,GroundWaterFlow,store_error,cellid-grid,K7,"(a,2(1x,a)) CELLID is not in the active model domain." +Model/GroundWaterFlow/gwf-tvk.f90,162,tvk_apply_row_changes,GroundWaterFlow,store_error,cellid-grid,K7,"(a,2(1x,a)) CELLID is not in the active model domain." +Model/GroundWaterFlow/gwf-tvk.f90,236,tvk_validate_change,GroundWaterFlow,store_error,sign-value,K2,"(1x, a, ' changed hydraulic property ', a, ' is <= 0 for cell ', a, ' ', 1pg15.6) K" +Model/GroundWaterFlow/gwf-tvk.f90,246,tvk_validate_change,GroundWaterFlow,store_error,sign-value,K2,"(1x, a, ' changed hydraulic property ', a, ' is <= 0 for cell ', a, ' ', 1pg15.6) K22" +Model/GroundWaterFlow/gwf-tvk.f90,256,tvk_validate_change,GroundWaterFlow,store_error,sign-value,K2,"(1x, a, ' changed hydraulic property ', a, ' is <= 0 for cell ', a, ' ', 1pg15.6) K33" +Model/GroundWaterFlow/gwf-tvs.f90,138,tvs_apply_row_changes,GroundWaterFlow,store_error,cellid-grid,K7,"(a,2(1x,a)) CELLID is not in the active model domain." +Model/GroundWaterFlow/gwf-tvs.f90,156,tvs_apply_row_changes,GroundWaterFlow,store_error,cellid-grid,K7,"(a,2(1x,a)) CELLID is not in the active model domain." +Model/GroundWaterFlow/gwf-tvs.f90,225,tvs_validate_change,GroundWaterFlow,store_error,sign-value,K2,"(1x, a, ' changed storage property ', a, ' is < 0 for cell ', a,' ', 1pg15.6) SS" +Model/GroundWaterFlow/gwf-tvs.f90,232,tvs_validate_change,GroundWaterFlow,store_error,unclassified,?,"(1x, a, ' cannot change ', a ,' for cell ', a, ' because SY is unused in this model (all ICONVERT flags are 0).') SY" +Model/GroundWaterFlow/gwf-tvs.f90,237,tvs_validate_change,GroundWaterFlow,store_error,sign-value,K2,"(1x, a, ' changed storage property ', a, ' is < 0 for cell ', a,' ', 1pg15.6) SY" +Model/GroundWaterFlow/gwf-uzf.f90,606,uzf_source_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) NUZFCELLS (MAXBOUND) was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-uzf.f90,611,uzf_source_dimensions,GroundWaterFlow,store_error,sign-value,K2,(a) NTRAILWAVES must be greater than 1. A value of 7 is recommended. +Model/GroundWaterFlow/gwf-uzf.f90,616,uzf_source_dimensions,GroundWaterFlow,store_error,range-count,K3,(a) NWAVESETS was not specified or was specified incorrectly. +Model/GroundWaterFlow/gwf-uzf.f90,745,uzf_rp,GroundWaterFlow,store_error,sign-value,K2,"(a,a,i0,1x,a,i0,a) : UZFNO must be greater than 0 and less than or equal to ." +Model/GroundWaterFlow/gwf-uzf.f90,1601,uzf_solve,GroundWaterFlow,store_error,unclassified,?,UZF variable NWAVESETS needs to be increased. +Model/GroundWaterFlow/gwf-uzf.f90,1776,uzf_source_packagedata,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0,a) LANDFLAG for uzf cell must be 0 or 1 (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1785,uzf_source_packagedata,GroundWaterFlow,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0,a) IVERTCON for uzf cell must be 0 or less than NUZFCELLS (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1793,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g0,a) SURFDEP for uzf cell must be greater than 0 (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1799,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a) SURFDEP for uzf cell cannot be greater than the cell thickness." +Model/GroundWaterFlow/gwf-uzf.f90,1807,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g0,a) VKS for uzf cell must be greater than 0 (specified value ia )." +Model/GroundWaterFlow/gwf-uzf.f90,1815,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g0,a) THTR for uzf cell must be greater than 0 (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1823,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,g0,a) THTS for uzf cell must be greater than THTR (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1832,uzf_source_packagedata,GroundWaterFlow,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,a,1x,g0,a) THTI for uzf cell must be greater than or equal to THTR AND less than THTS (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,1840,uzf_source_packagedata,GroundWaterFlow,store_error,range-count,K3,"(a,1x,i0,1x,a,1x,g0,a) EPSILON for uzf cell must be between 3.5 and 14.0 (specified value is )." +Model/GroundWaterFlow/gwf-uzf.f90,2033,check_cell_area,GroundWaterFlow,store_error,range-count,K3,"(2(a,1x,g0,1x,a,1x,i0,1x),a) UZF cell area ( ) for cell does not equal uzf cell area ( ) for cell ." +Model/GroundWaterFlow/gwf-uzf.f90,2063,check_cell_area,GroundWaterFlow,store_error,range-count,K3,"(a,1x,g0,1x,a,1x,g0,1x,a,1x,a,1x,a,a,a) Total uzf cell area ( ) exceeds the gwf cell area ( ) of cell which includes uzf cell(s): ." +Model/GroundWaterFlow/gwf-uzf.f90,2229,uzf_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/GroundWaterFlow/gwf-uzf.f90,2300,uzf_rp_obs,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-uzf.f90,2317,uzf_rp_obs,GroundWaterFlow,store_error,duplicate,K6,"(a,3(1x,a)) for observation must be assigned to a UZF cell with a unique boundname." +Model/GroundWaterFlow/gwf-uzf.f90,2341,uzf_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,3(1x,a),1x,g0,1x,a,1x,g0,a) for observation specified depth ( ) must be greater than 0.0 and less than ." +Model/GroundWaterFlow/gwf-uzf.f90,2351,uzf_rp_obs,GroundWaterFlow,store_error,sign-value,K2,"(a,2(1x,a),1x,i0,1x,a,1x,i0,a) uzfno must be greater than 0 and less than or equal to (specified value is )." +Model/GroundWaterFlow/gwf-vsc.f90,344,vsc_rp,GroundWaterFlow,store_error,binding-context,K8,"('Viscosity Package does not have a concentration set for species ',i0,'. One or more model names may be specified incorrectly in the PACKAGEDATA block or a GWF-GWT exchange may need to be activated.')" +Model/GroundWaterFlow/gwf-vsc.f90,969,source_dimensions,GroundWaterFlow,store_error,sign-value,K2,NVISCSPECIES must be greater than zero. +Model/GroundWaterFlow/gwf-vsc.f90,1016,source_packagedata,GroundWaterFlow,store_error,duplicate,K6,"('Invalid value for IRHOSPEC (',i0,') detected in VSC Package. IRHOSPEC must be > 0 and <= NVISCSPECIES, and duplicate values are not allowed.')" +Model/GroundWaterFlow/gwf-vsc.f90,1020,source_packagedata,GroundWaterFlow,store_error,duplicate,K6,"('Invalid value for IRHOSPEC (',i0,') detected in VSC Package. IRHOSPEC must be > 0 and <= NVISCSPECIES, and duplicate values are not allowed.')" +Model/GroundWaterFlow/gwf-vsc.f90,1034,source_packagedata,GroundWaterFlow,store_error,unclassified,?,(a) More than one species in VSC input identified as . Only one species may be designated to represent temperature. +Model/GroundWaterFlow/gwf-vsc.f90,1255,source_options,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized input value for THERMAL_FORM option. +Model/GroundWaterFlow/gwf-vsc.f90,1266,source_options,GroundWaterFlow,store_error,unclassified,?,"(a) LINEAR option selected for varying viscosity with temperature, but A1, a surrogate for dVISC/dT, set equal to 0.0" +Model/GroundWaterFlow/gwf-vsc.f90,1275,source_options,GroundWaterFlow,store_warning,sign-value,K2,"(a) NONLINEAR option selected for varying viscosity with temperature, but A2 set equal to zero which may lead to unintended values for viscosity" +Model/GroundWaterFlow/gwf-vsc.f90,1281,source_options,GroundWaterFlow,store_warning,sign-value,K2,"(a) NONLINEAR option selected for varying viscosity with temperature,, but A3 set equal to zero which may lead to unintended values for viscosity" +Model/GroundWaterFlow/gwf-vsc.f90,1287,source_options,GroundWaterFlow,store_warning,sign-value,K2,"(a) NONLINEAR option selected for varying viscosity with temperature, BUT A4 SET EQUAL TO zero which may lead to unintended values for viscosity" +Model/GroundWaterFlow/gwf-wel.f90,231,wel_options,GroundWaterFlow,store_warning,geometry,K10,(a) FLOW_REDUCTION_LENGTH option specified but a AUTO_FLOW_REDUCTION value is not specified. The FLOW_REDUCTION_LENGTH option will be ignored. +Model/GroundWaterFlow/gwf-wel.f90,244,wel_options,GroundWaterFlow,store_error,sign-value,K2,(a) An AUTO_FLOW_REDUCTION value less than or equal to zero cannot be specified if the FLOW_REDUCTION_LENGTH option is specified. +Model/GroundWaterFlow/gwf-wel.f90,266,wel_options,GroundWaterFlow,store_warning,option-coherence,K5,(a) AUTO_FLOW_REDUCE_AUXNAME is specified but AUTO_FLOW_REDUCE is not specified. The AUTO_FLOW_REDUCE_AUXNAME option will be ignored. +Model/GroundWaterFlow/gwf-wel.f90,273,wel_options,GroundWaterFlow,store_warning,geometry,K10,(a) AUTO_FLOW_REDUCE_AUXNAME is specified but FLOW_REDUCTION_LENGTH is not specified. The AUTO_FLOW_REDUCE_AUXNAME value will be interpreted as a fraction of the cell thickness. +Model/GroundWaterFlow/gwf-wel.f90,280,wel_options,GroundWaterFlow,store_error,unclassified,?,"(a,2(1x,a)) AUTO_FLOW_REDUCE_AUXNAME was specified as but no AUX variables specified." +Model/GroundWaterFlow/gwf-wel.f90,294,wel_options,GroundWaterFlow,store_error,unclassified,?,"(a,2(1x,a)) AUTO_FLOW_REDUCE_AUXNAME was specified as but no AUX variable found with this name." +Model/GroundWaterFlow/gwf-wel.f90,411,wel_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-wel.f90,418,wel_ck,GroundWaterFlow,store_error,unclassified,?, +Model/GroundWaterFlow/gwf-wel.f90,741,wel_bd_obs,GroundWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/GroundWaterFlow/gwf-wel.f90,794,wel_bound_value,GroundWaterFlow,store_error,not-input,-,Programming error. WEL bound value requested column ''outside range of ncolbnd (1). +Model/GroundWaterFlow/gwf.f90,1314,package_create,GroundWaterFlow,store_error,type-parse,K1,Invalid package type: +Model/GroundWaterFlow/gwf.f90,1324,package_create,GroundWaterFlow,store_error,duplicate,K6,"(a,a) Cannot create package. Package name already exists:" +Model/GroundWaterFlow/gwf.f90,1345,ftype_check,GroundWaterFlow,store_error,unclassified,?,(a) Initial Conditions (IC6) package not specified. +Model/GroundWaterFlow/gwf.f90,1350,ftype_check,GroundWaterFlow,store_error,unclassified,?,"(a) Discretization (DIS6, DISV6, or DISU6) Package not specified." +Model/GroundWaterFlow/gwf.f90,1355,ftype_check,GroundWaterFlow,store_error,unclassified,?,(a) Node Property Flow (NPF6) Package not specified. +Model/GroundWaterFlow/gwf.f90,1360,ftype_check,GroundWaterFlow,store_error,unclassified,?,(a) One or more required package(s) not specified. +Model/GroundWaterFlow/gwf.f90,1631,steady_period_check,GroundWaterFlow,store_warning,binding-context,K8,"(a,a,a,i0,a) GWF Model ( ) is steady state for period and adaptive time stepping is active. Adaptive time stepping may not work properly for steady-state conditions." +Model/GroundWaterFlow/submodules/gwf-lak-implicit.f90,364,,GroundWaterFlow,store_warning,not-input,-,"(a) IMPLICIT lake option in package ' ' did not converge even after switching a stalled lake to the substitution solver. The default solver needs fewer outer iterations for such a lake; consider omitting IMPLICIT, adding an outlet, increasing the leakance, or running transient." +Model/GroundWaterTransport/gwt-cnc.f90,160,cnc_rp,GroundWaterTransport,store_error,unclassified,?,Cell is already a constant : +Model/GroundWaterTransport/gwt-cnc.f90,227,cnc_ck,GroundWaterTransport,store_error,unclassified,?, +Model/GroundWaterTransport/gwt-cnc.f90,484,cnc_bound_value,GroundWaterTransport,store_error,not-input,-,(3a) Programming error. bound value requested column ''outside range of ncolbnd (1). +Model/GroundWaterTransport/gwt-dsp.f90,606,source_griddata,GroundWaterTransport,store_error,unclassified,?,(a) if dispersivities are specified then ALH and ATH1 are required. +Model/GroundWaterTransport/gwt-ist.f90,212,ist_ar,GroundWaterTransport,store_error,geometry,K10,DECAY must be activated consistently between the MST and IST Packages. Activate or deactivate DECAY for both Packages. +Model/GroundWaterTransport/gwt-ist.f90,217,ist_ar,GroundWaterTransport,store_error,block-structure,K1,"SORPTION must be activated consistently between the MST and IST Packages. Activate or deactivate SORPTION for both Packages. If activated, the same type of sorption (LINEAR, FREUNDLICH, or LANGMUIR) must be specified in the options block of both the MST and IST Packages." +Model/GroundWaterTransport/gwt-ist.f90,1022,source_options,GroundWaterTransport,store_error,enum,K2,"Unknown sorption type was specified. Sorption must be specified as LINEAR, FREUNDLICH, or LANGMUIR." +Model/GroundWaterTransport/gwt-ist.f90,1217,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Sorption is active but BULK_DENSITY not specified. BULK_DENSITY must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1223,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Sorption is active but distribution coefficient not specified. DISTCOEF must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1230,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Freundlich or langmuir sorption is active but SP2 not specified. SP2 must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1238,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but BULK_DENSITY was specified. BULK_DENSITY will have no affect on simulation results. +Model/GroundWaterTransport/gwt-ist.f90,1245,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but distribution coefficient was specified. DISTCOEF will have no affect on simulation results. +Model/GroundWaterTransport/gwt-ist.f90,1252,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but SP2 was specified. SP2 will have no affect on simulation results. +Model/GroundWaterTransport/gwt-ist.f90,1263,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) First or zero order decay is active but the first rate coefficient is not specified. DECAY must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1272,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) DECAY_SORBED not provided in GRIDDATA block but decay and sorption are active. Specify DECAY_SORBED in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1279,source_data,GroundWaterTransport,store_warning,sign-value,K2,(a) First- or zero-order decay is not active but decay was specified. DECAY will have no affect on simulation results. +Model/GroundWaterTransport/gwt-ist.f90,1286,source_data,GroundWaterTransport,store_warning,sign-value,K2,(a) First- or zero-order decay is not active but DECAY_SORBED was specified. DECAY_SORBED will have no affect on simulation results. +Model/GroundWaterTransport/gwt-ist.f90,1302,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Dual domain is active but dual domain mass transfer rate (ZETAIM) was not specified. ZETAIM must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1308,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Dual domain is active but immobile domain POROSITY was not specified. POROSITY must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-ist.f90,1316,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Dual domain is active but immobile domain VOLFRAC was not specified. VOLFRAC must be specified in GRIDDATA block. This is a new requirement for MODFLOW versions later than version 6.4.1. +Model/GroundWaterTransport/gwt-lkt.f90,206,find_lkt_package,GroundWaterTransport,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterTransport/gwt-mst.f90,1136,source_options,GroundWaterTransport,store_error,enum,K2,"Unknown sorption type was specified. Sorption must be specified as LINEAR, FREUNDLICH, or LANGMUIR." +Model/GroundWaterTransport/gwt-mst.f90,1276,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) POROSITY not specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1285,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Sorption is active but BULK_DENSITY not specified. BULK_DENSITY must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1291,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Sorption is active but distribution coefficient not specified. DISTCOEF must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1298,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) Freundlich or langmuir sorption is active but SP2 not specified. SP2 must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1306,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but BULK_DENSITY was specified. BULK_DENSITY will have no affect on simulation results. +Model/GroundWaterTransport/gwt-mst.f90,1313,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but distribution coefficient was specified. DISTCOEF will have no affect on simulation results. +Model/GroundWaterTransport/gwt-mst.f90,1320,source_data,GroundWaterTransport,store_warning,option-coherence,K5,(a) Sorption is not active but SP2 was specified. SP2 will have no affect on simulation results. +Model/GroundWaterTransport/gwt-mst.f90,1331,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) First or zero order decay is active but the first rate coefficient is not specified. DECAY must be specified in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1342,source_data,GroundWaterTransport,store_error,block-structure,K1,(a) DECAY_SORBED not provided in GRIDDATA block but decay and sorption are active. Specify DECAY_SORBED in GRIDDATA block. +Model/GroundWaterTransport/gwt-mst.f90,1350,source_data,GroundWaterTransport,store_warning,sign-value,K2,(a) First- or zero-order decay is not active but decay was specified. DECAY will have no affect on simulation results. +Model/GroundWaterTransport/gwt-mst.f90,1357,source_data,GroundWaterTransport,store_warning,sign-value,K2,(a) First- or zero-order decay is not active but DECAY_SORBED was specified. DECAY_SORBED will have no affect on simulation results. +Model/GroundWaterTransport/gwt-mwt.f90,199,find_mwt_package,GroundWaterTransport,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterTransport/gwt-sft.f90,208,find_sft_package,GroundWaterTransport,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterTransport/gwt-uzt.f90,192,find_uzt_package,GroundWaterTransport,store_error,not-found-ref,K7,(a) Could not find flow package with name . +Model/GroundWaterTransport/gwt.f90,169,gwt_df,GroundWaterTransport,store_error,binding-context,K8,"Flow model has boundary packages, but there is no SSM package. The SSM package must be activated." +Model/GroundWaterTransport/gwt.f90,805,package_create,GroundWaterTransport,store_error,type-parse,K1,Invalid package type: +Model/GroundWaterTransport/gwt.f90,816,package_create,GroundWaterTransport,store_error,duplicate,K6,"(a,a) Cannot create package. Package name already exists:" +Model/ParticleTracking/prt-fmi.f90,164,prtfmi_df,ParticleTracking,store_error,range-count,K3,"(a,i0,a,i0,a) DISV grid contains a cell with lateral faces. Cells may have at most 30 lateral faces." +Model/ParticleTracking/prt-mip.f90,136,mip_ar,ParticleTracking,store_error,block-structure,K1,(a) Error in GRIDDATA block: POROSITY not found +Model/ParticleTracking/prt-oc.f90,345,prt_oc_source_dimensions,ParticleTracking,store_error,range-count,K3,(a) NTRACKTIMES WAS NOT SPECIFIED OR WAS SPECIFIED INCORRECTLY. +Model/ParticleTracking/prt-oc.f90,367,prt_oc_source_tracktimes,ParticleTracking,store_error,geometry,K10,"(a, i0) Expected TRACKTIMES with length" +Model/ParticleTracking/prt-oc.f90,383,prt_oc_source_tracktimes,ParticleTracking,store_error,unclassified,?,TRACKTIMES must strictly increase +Model/ParticleTracking/prt-prp.f90,489,prp_ad,ParticleTracking,store_warning,sign-value,K2,"(a,g0,a) Skipping negative release time (t= )." +Model/ParticleTracking/prt-prp.f90,496,prp_ad,ParticleTracking,store_warning,temporal,K9,"(a,g0,a) Skipping release time falling after the end of the simulation (t= ). Enable EXTEND_TRACKING to release particles after the simulation end time." +Model/ParticleTracking/prt-prp.f90,585,validate_release_point,ParticleTracking,store_error,unclassified,?,"(a,g0,a,g0,a,i0) Error: release point (x= , y= ) is not in cell" +Model/ParticleTracking/prt-prp.f90,592,validate_release_point,ParticleTracking,store_error,geometry,K10,"(a,g0,a,g0,a,i0) Error: release point (z= ) is above grid top" +Model/ParticleTracking/prt-prp.f90,598,validate_release_point,ParticleTracking,store_error,geometry,K10,"(a,g0,a,g0,a,i0) Error: release point (z= ) is below grid bottom" +Model/ParticleTracking/prt-prp.f90,928,prp_options,ParticleTracking,store_error,enum,K2,"(a) Unsupported dry tracking method. DRY_TRACKING_METHOD must be ""DROP"", ""STOP"", or ""STAY""" +Model/ParticleTracking/prt-prp.f90,937,prp_options,ParticleTracking,store_error,sign-value,K2,EXIT_SOLVE_TOLERANCE MUST BE POSITIVE +Model/ParticleTracking/prt-prp.f90,941,prp_options,ParticleTracking,store_warning,unclassified,?, +Model/ParticleTracking/prt-prp.f90,947,prp_options,ParticleTracking,store_error,sign-value,K2,RELEASE_TIME_TOLERANCE MUST BE POSITIVE +Model/ParticleTracking/prt-prp.f90,952,prp_options,ParticleTracking,store_error,sign-value,K2,RELEASE_TIME_FREQUENCY MUST BE POSITIVE +Model/ParticleTracking/prt-prp.f90,957,prp_options,ParticleTracking,store_error,enum,K2,DEV_EXIT_SOLVE_METHOD MUST BE 1 (BRENT) OR 2 (CHANDRUPATLA) +Model/ParticleTracking/prt-prp.f90,965,prp_options,ParticleTracking,store_error,enum,K2,"(a) Unsupported coordinate check method. COORDINATE_CHECK_METHOD must be ""NONE"" or ""EAGER""" +Model/ParticleTracking/prt-prp.f90,974,prp_options,ParticleTracking,store_error,sign-value,K2,CYCLE_DETECTION_WINDOW MUST BE NON-NEGATIVE +Model/ParticleTracking/prt-prp.f90,1143,prp_packagedata,ParticleTracking,store_error,type-parse,K1,"(a,i0,a,i0,a) Expected release points. Points must be numbered from 1 to ." +Model/ParticleTracking/prt-prp.f90,1175,prp_packagedata,ParticleTracking,store_error,not-found-ref,K7,(a) Particle release point configured for nonexistent cell: . This cell has IDOMAIN <= 0 and ''therefore does not exist in the model grid. +Model/ParticleTracking/prt-prp.f90,1182,prp_packagedata,ParticleTracking,store_error,unclassified,?,"Local z coordinate must fall in the interval [0, 1]" +Model/ParticleTracking/prt-prp.f90,1215,prp_packagedata,ParticleTracking,store_error,unclassified,?,"(a,a,1x,i0,a) No data specified for particle release point ." +Model/ParticleTracking/prt-prp.f90,1220,prp_packagedata,ParticleTracking,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0,1x,a) Data for particle release point specified times." +Model/ParticleTracking/prt-prp.f90,1261,prp_releasetimes,ParticleTracking,store_error,block-structure,K1,RELEASTIMES block expected when NRELEASETIMES dimension is non-zero. +Model/ParticleTracking/prt-prp.f90,1279,prp_releasetimes,ParticleTracking,store_error,block-structure,K1,RELEASTIMES block entries must strictly increase. +Model/ParticleTracking/prt-prp.f90,1311,prp_load_releasetimefrequency,ParticleTracking,store_error,unclassified,?,Release times must strictly increase +Model/ParticleTracking/prt.f90,1001,package_create,ParticleTracking,store_error,type-parse,K1,Invalid package type: +Model/ParticleTracking/prt.f90,1012,package_create,ParticleTracking,store_error,duplicate,K6,"(a,a) Cannot create package. Package name already exists:" +Model/ParticleTracking/prt.f90,1030,ftype_check,ParticleTracking,store_error,unclassified,?,"(1x,a) Discretization (DIS6, DISV6, or DISU6) package not specified." +Model/ParticleTracking/prt.f90,1035,ftype_check,ParticleTracking,store_error,not-found-ref,K7,"(1x,a) Model input (MIP6) package not specified." +Model/ParticleTracking/prt.f90,1040,ftype_check,ParticleTracking,store_error,unclassified,?,"(1x,a) One or more required package(s) not specified." +Model/ParticleTracking/prt.f90,1104,prt_solve,ParticleTracking,store_warning,not-found-ref,K7,"(a,g0,a,g0,a,g0,a) Particle release point has release time after package stop time ; particle will not be released." +Model/SurfaceWaterFlow/swf-cdb.f90,456,cdb_bd_obs,SurfaceWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/SurfaceWaterFlow/swf-cdb.f90,517,cdb_bound_value,SurfaceWaterFlow,store_error,not-input,-,Programming error. CDB bound value requested column ''outside range of ncolbnd (1). +Model/SurfaceWaterFlow/swf-cxs.f90,240,source_dimensions,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in DIMENSIONS block: NSECTIONS not found. +Model/SurfaceWaterFlow/swf-cxs.f90,246,source_dimensions,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in DIMENSIONS block: NPOINTS not found. +Model/SurfaceWaterFlow/swf-cxs.f90,337,source_packagedata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in PACKAGEDATA block: IFNO not found. +Model/SurfaceWaterFlow/swf-cxs.f90,343,source_packagedata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in PACKAGEDATA block: NXSPOINTS not found. +Model/SurfaceWaterFlow/swf-cxs.f90,384,check_packagedata,SurfaceWaterFlow,store_error,sign-value,K2,"(a, i0, a) IFNO values must be greater than 0 and less than NSECTIONS. Found ." +Model/SurfaceWaterFlow/swf-cxs.f90,394,check_packagedata,SurfaceWaterFlow,store_error,sign-value,K2,"(a, i0, a, i0, a) NXSPOINTS values must be greater than 1 for each cross section. Found for cross section ." +Model/SurfaceWaterFlow/swf-cxs.f90,457,source_crosssectiondata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in CROSSSECTIONDATA block: IFNO not found. +Model/SurfaceWaterFlow/swf-cxs.f90,464,source_crosssectiondata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in CROSSSECTIONDATA block: xfraction not found. +Model/SurfaceWaterFlow/swf-cxs.f90,471,source_crosssectiondata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in CROSSSECTIONDATA block: HEIGHT not found. +Model/SurfaceWaterFlow/swf-cxs.f90,478,source_crosssectiondata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in CROSSSECTIONDATA block: MANFRACTION not found. +Model/SurfaceWaterFlow/swf-cxs.f90,515,check_crosssectiondata,SurfaceWaterFlow,store_error,range-count,K3,"(a, i0, a, i0, a, i0, a) IFNO value in CROSSSECTIONDATA does not match the cross section implied by PACKAGEDATA. Found IFNO = at point but expected ." +Model/SurfaceWaterFlow/swf-dfw.f90,357,source_options,SurfaceWaterFlow,store_error,block-structure,K1,Multiple OBS6 keywords detected in OPTIONS block. Only one OBS6 entry allowed. +Model/SurfaceWaterFlow/swf-dfw.f90,464,source_griddata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in GRIDDATA block: MANNINGSN not found. +Model/SurfaceWaterFlow/swf-dfw.f90,1350,calc_velocity,SurfaceWaterFlow,store_error,geometry,K10,Error. ANGLDEGX not provided in discretization file. ANGLDEGX required for calculation of velocity. +Model/SurfaceWaterFlow/swf-dfw.f90,1673,dfwobsidprocessor,SurfaceWaterFlow,store_error,not-input,-,Error reading data from ID string +Model/SurfaceWaterFlow/swf-dfw.f90,1704,dfw_bd_obs,SurfaceWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/SurfaceWaterFlow/swf-dfw.f90,1746,dfw_rp_obs,SurfaceWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) reach must be greater than 0 and less than or equal to (specified value is )" +Model/SurfaceWaterFlow/swf-dfw.f90,1752,dfw_rp_obs,SurfaceWaterFlow,store_error,not-input,-,Programming error in dfw_rp_obs +Model/SurfaceWaterFlow/swf-dfw.f90,1764,dfw_rp_obs,SurfaceWaterFlow,store_error,sign-value,K2,"(a,1x,a,1x,i0,1x,a,1x,i0,a) reach must be greater than 0 and less than or equal to (specified value is )" +Model/SurfaceWaterFlow/swf-evp.f90,229,evp_source_dimensions,SurfaceWaterFlow,store_error,sign-value,K2,(a) MAXBOUND must be an integer greater than zero. +Model/SurfaceWaterFlow/swf-evp.f90,296,evp_ck,SurfaceWaterFlow,store_error,unclassified,?, +Model/SurfaceWaterFlow/swf-evp.f90,620,evp_bound_value,SurfaceWaterFlow,store_error,not-input,-,Programming error. EVP bound value requested column ''outside range of ncolbnd (1). +Model/SurfaceWaterFlow/swf-flw.f90,370,flw_bd_obs,SurfaceWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/SurfaceWaterFlow/swf-flw.f90,445,flw_bound_value,SurfaceWaterFlow,store_error,not-input,-,Programming error. FLW bound value requested column ''outside range of ncolbnd (1). +Model/SurfaceWaterFlow/swf-ic.f90,156,source_griddata,SurfaceWaterFlow,store_error,block-structure,K1,(a) Error in GRIDDATA block: STRT not found. +Model/SurfaceWaterFlow/swf-obs.f90,118,swf_obs_bd,SurfaceWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/SurfaceWaterFlow/swf-obs.f90,187,swf_process_stage_obs_id,SurfaceWaterFlow,store_error,not-input,-,Error reading data from ID string +Model/SurfaceWaterFlow/swf-obs.f90,220,swf_process_intercell_obs_id,SurfaceWaterFlow,store_error,not-input,-,Error reading data from ID string: +Model/SurfaceWaterFlow/swf-obs.f90,231,swf_process_intercell_obs_id,SurfaceWaterFlow,store_error,not-input,-,Error reading data from ID string: +Model/SurfaceWaterFlow/swf-obs.f90,238,swf_process_intercell_obs_id,SurfaceWaterFlow,store_error,unclassified,?, +Model/SurfaceWaterFlow/swf-pcp.f90,218,pcp_source_dimensions,SurfaceWaterFlow,store_error,sign-value,K2,(a) MAXBOUND must be an integer greater than zero. +Model/SurfaceWaterFlow/swf-pcp.f90,285,pcp_ck,SurfaceWaterFlow,store_error,unclassified,?, +Model/SurfaceWaterFlow/swf-pcp.f90,512,pcp_bound_value,SurfaceWaterFlow,store_error,not-input,-,Programming error. PCP bound value requested column ''outside range of ncolbnd (1). +Model/SurfaceWaterFlow/swf-sto.f90,179,sto_rp,SurfaceWaterFlow,store_error,unknown-keyword,K1,"(a,a) Unknown STORAGE data tag:" +Model/SurfaceWaterFlow/swf-sto.f90,228,sto_fc,SurfaceWaterFlow,store_error,sign-value,K2,"('Detected time step length of zero. SWF Storage Package cannot be ', 'used unless delt is non-zero.')" +Model/SurfaceWaterFlow/swf-zdg.f90,477,zdg_bd_obs,SurfaceWaterFlow,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/SurfaceWaterFlow/swf-zdg.f90,542,zdg_bound_value,SurfaceWaterFlow,store_error,not-input,-,Programming error. ZDG bound value requested column ''outside range of ncolbnd (1). +Model/SurfaceWaterFlow/swf.f90,881,package_create,SurfaceWaterFlow,store_error,type-parse,K1,Invalid package type: +Model/SurfaceWaterFlow/swf.f90,892,package_create,SurfaceWaterFlow,store_error,duplicate,K6,"(a,a) Cannot create package. Package name already exists:" +Model/SurfaceWaterFlow/swf.f90,912,ftype_check,SurfaceWaterFlow,store_error,not-found-ref,K7,(a) Discretization Package (DISV1D6 or DIS2D6) not specified. +Model/SurfaceWaterFlow/swf.f90,918,ftype_check,SurfaceWaterFlow,store_error,option-coherence,K5,(a) Initial Conditions (IC6) must be specified if the Diffusive Wave (DFW) package is used. +Model/SurfaceWaterFlow/swf.f90,923,ftype_check,SurfaceWaterFlow,store_error,option-coherence,K5,"(1x,a) DFW6 Package must be specified." +Model/SurfaceWaterFlow/swf.f90,931,ftype_check,SurfaceWaterFlow,store_error,option-coherence,K5,"(1x,a) CXS6 Package can only be used for one-dimensional discretization package (DISV1D). Remove CXS6 Package from model or convert discretization package to DISV1D." +Model/SurfaceWaterFlow/swf.f90,937,ftype_check,SurfaceWaterFlow,store_error,unclassified,?,(a) One or more required package(s) not specified and/or there are package incompatibilities. +Model/SurfaceWaterFlow/swf.f90,1138,steady_period_check,SurfaceWaterFlow,store_warning,not-found-ref,K7,"(a,a,a,i0,a) SWF Model ( ) is steady state for period and adaptive time stepping is active. Adaptive time stepping may not work properly for steady-state conditions." +Model/TransportModel/Isotherm/IsothermFactory.f90,42,create_isotherm,TransportModel,store_error,not-input,-,Sorption type not implemented. +Model/TransportModel/tsp-adv.f90,148,adv_ar,TransportModel,store_error,unclassified,?,Unknown advection scheme +Model/TransportModel/tsp-adv.f90,369,source_options,TransportModel,store_error,unclassified,?,"(a, a) Unknown scheme, must be ""UPSTREAM"", ""CENTRAL"", ""TVD"" or ""UTVD""" +Model/TransportModel/tsp-apt.f90,364,apt_ar,TransportModel,store_error,not-found-ref,K7,Could not find auxiliary variable in flow package +Model/TransportModel/tsp-apt.f90,461,apt_rp,TransportModel,store_error,unknown-keyword,K1,"(a,a) Unknown status keyword: ." +Model/TransportModel/tsp-apt.f90,547,apt_check_valid,TransportModel,store_error,sign-value,K2,"(a,1x,i6,1x,a,1x,i6) Featureno must be > 0 and <=" +Model/TransportModel/tsp-apt.f90,840,pak_fc_expanded,TransportModel,store_error,not-input,-,Program error: pak_fc_expanded not implemented. +Model/TransportModel/tsp-apt.f90,1299,find_apt_package,TransportModel,store_error,not-input,-,Program error: pak_solve not implemented. +Model/TransportModel/tsp-apt.f90,1430,apt_source_dimensions,TransportModel,store_error,unclassified,?,(a) Number of control volumes could not be determined correctly. +Model/TransportModel/tsp-apt.f90,1670,pak_solve,TransportModel,store_error,not-input,-,Program error: pak_solve not implemented. +Model/TransportModel/tsp-apt.f90,1791,pak_get_nbudterms,TransportModel,store_error,not-input,-,Program error: pak_get_nbudterms not implemented. +Model/TransportModel/tsp-apt.f90,1994,pak_setup_budobj,TransportModel,store_error,not-input,-,Program error: pak_setup_budobj not implemented. +Model/TransportModel/tsp-apt.f90,2151,pak_fill_budobj,TransportModel,store_error,not-input,-,Program error: pak_fill_budobj not implemented. +Model/TransportModel/tsp-apt.f90,2330,pak_df_obs,TransportModel,store_error,not-input,-,Program error: pak_df_obs not implemented. +Model/TransportModel/tsp-apt.f90,2346,pak_rp_obs,TransportModel,store_error,not-input,-,Program error: pak_rp_obs not implemented. +Model/TransportModel/tsp-apt.f90,2376,rp_obs_byfeature,TransportModel,store_error,type-parse,K1,"('Boundary ', a, ' for observation ', a, ' is invalid in package ', a)" +Model/TransportModel/tsp-apt.f90,2387,rp_obs_byfeature,TransportModel,store_error,sign-value,K2,"(7a, i0, a, i0, a) Observation of type in package was assigned ID = . ID must be >= 1 and <= ." +Model/TransportModel/tsp-apt.f90,2424,rp_obs_budterm,TransportModel,store_error,type-parse,K1,"('Boundary ', a, ' for observation ', a, ' is invalid in package ', a)" +Model/TransportModel/tsp-apt.f90,2435,rp_obs_budterm,TransportModel,store_error,sign-value,K2,"(7a, i0, a, i0, a) Observation of type in package was assigned ID = . ID must be >= 1 and <= ." +Model/TransportModel/tsp-apt.f90,2453,rp_obs_budterm,TransportModel,store_error,not-found-ref,K7,"(7a, i0, a, i0, a) Observation of type in package specifies iconn = , but this is not a valid connection for ID ." +Model/TransportModel/tsp-apt.f90,2460,rp_obs_budterm,TransportModel,store_error,not-found-ref,K7,"(7a, i0, a, i0, a) Observation of type in package specifies iconn = , but this is not a valid connection for ID ." +Model/TransportModel/tsp-apt.f90,2495,rp_obs_flowjaface,TransportModel,store_error,type-parse,K1,"('Boundary ', a, ' for observation ', a, ' is invalid in package ', a)" +Model/TransportModel/tsp-apt.f90,2506,rp_obs_flowjaface,TransportModel,store_error,sign-value,K2,"(7a, i0, a, i0, a) Observation of type in package was assigned ID = . ID must be >= 1 and <= ." +Model/TransportModel/tsp-apt.f90,2517,rp_obs_flowjaface,TransportModel,store_error,sign-value,K2,"(7a, i0, a, i0, a) Observation of type in package was assigned ID2 = . ID must be >= 1 and <= ." +Model/TransportModel/tsp-apt.f90,2534,rp_obs_flowjaface,TransportModel,store_error,not-found-ref,K7,"(7a, i0, a, i0, a) Observation of type in package specifies a connection between feature feature , but these features are not connected." +Model/TransportModel/tsp-apt.f90,2567,apt_rp_obs,TransportModel,store_error,duplicate,K6,"(a, a, a, a) for observation must be assigned to a feature with a unique boundname." +Model/TransportModel/tsp-apt.f90,2582,apt_rp_obs,TransportModel,store_error,geometry,K10,(7a) Observation of type in package cannot be processed because there are no flow connections. +Model/TransportModel/tsp-apt.f90,2602,apt_rp_obs,TransportModel,store_error,unknown-keyword,K1,"Unrecognized observation type "" "" for package" +Model/TransportModel/tsp-apt.f90,2693,apt_bd_obs,TransportModel,store_error,unknown-keyword,K1,"Unrecognized observation type "" "" for package" +Model/TransportModel/tsp-apt.f90,2806,apt_process_obsID12,TransportModel,store_error,option-coherence,K5,"(a,1x,a,a,1x,a,1x,a) For observation type , ID given as an integer and not as boundname, but ID2 is missing. Either change ID to valid boundname or supply valid entry for ID2." +Model/TransportModel/tsp-fmi.f90,123,fmi_rp,TransportModel,store_error,not-found-ref,K7,(a) GWF water mover is active but the GWT MVT package has not been specified. activate GWT MVT package. +Model/TransportModel/tsp-fmi.f90,129,fmi_rp,TransportModel,store_error,binding-context,K8,(a) GWF water mover terms are not available but the GWT MVT package has been activated. Activate GWF-GWT exchange or specify GWFMOVER in FMI PACKAGEDATA. +Model/TransportModel/tsp-fmi.f90,676,initialize_gwfterms_from_bfr,TransportModel,store_error,not-found-ref,K7,(a) Specific discharge not found in budget file. SAVE_SPECIFIC_DISCHARGE and SAVE_FLOWS must be activated in the NPF package. +Model/TransportModel/tsp-fmi.f90,682,initialize_gwfterms_from_bfr,TransportModel,store_error,not-found-ref,K7,(a) Saturation not found in budget file. SAVE_SATURATION and SAVE_FLOWS must be activated in the NPF package. +Model/TransportModel/tsp-fmi.f90,688,initialize_gwfterms_from_bfr,TransportModel,store_error,not-found-ref,K7,(a) FLOWJA not found in budget file. SAVE_FLOWS must be activated in the NPF package. +Model/TransportModel/tsp-mvt.f90,641,read_options,TransportModel,store_error,unclassified,?,Optional BUDGET keyword must be followed by FILEOUT +Model/TransportModel/tsp-mvt.f90,654,read_options,TransportModel,store_error,unclassified,?,Optional BUDGETCSV keyword must be followed by FILEOUT +Model/TransportModel/tsp-mvt.f90,660,read_options,TransportModel,store_error,unknown-keyword,K1,"(a,a) Unknown MVT option:" +Model/TransportModel/tsp-obs.f90,127,tsp_obs_bd,TransportModel,store_error,unknown-keyword,K1,Unrecognized observation type: +Model/TransportModel/tsp-obs.f90,199,tsp_process_obs_id,TransportModel,store_error,not-input,-,Error reading data from ID string +Model/TransportModel/tsp-obs.f90,234,tsp_process_intercell_obs_id,TransportModel,store_error,not-input,-,Error reading data from ID string: +Model/TransportModel/tsp-obs.f90,245,tsp_process_intercell_obs_id,TransportModel,store_error,not-input,-,Error reading data from ID string: +Model/TransportModel/tsp-obs.f90,252,tsp_process_intercell_obs_id,TransportModel,store_error,unclassified,?, +Model/TransportModel/tsp-ssm.f90,160,ssm_ar,TransportModel,store_error,binding-context,K8,"(a) SSM package does not detect any boundary flows that require SSM terms. Activate GWF-GWT (or GWF-GWE, as appropriate) exchange or activate FMI package and provide a budget file that contains boundary flows. If no boundary flows are present in corresponding GWF model then this SSM package should b" +Model/TransportModel/tsp-ssm.f90,810,source_sources,TransportModel,store_error,unclassified,?,"(a,a) Flow package cannot be found:" +Model/TransportModel/tsp-ssm.f90,820,source_sources,TransportModel,store_error,block-structure,K1,"(a, a) A package cannot be specified more than once in the SSM SOURCES block. The following package was specified more than once:" +Model/TransportModel/tsp-ssm.f90,833,source_sources,TransportModel,store_error,enum,K2,"(a, a) SRCTYPE must be AUX or AUXMIXED. Found:" +Model/TransportModel/tsp-ssm.f90,908,source_fileinput,TransportModel,store_error,unclassified,?,"(a,a) Flow package cannot be found:" +Model/TransportModel/tsp-ssm.f90,918,source_fileinput,TransportModel,store_error,block-structure,K1,"(a, a) A package cannot be specified more than once in the SSM SOURCES block. The following package was specified more than once:" +Model/TransportModel/tsp-ssm.f90,928,source_fileinput,TransportModel,store_error,unclassified,?,"(a,a) SRCTYPE must be SPC6. Found:" +Model/TransportModel/tsp-ssm.f90,936,source_fileinput,TransportModel,store_error,unclassified,?,"SPC6 keyword must be followed by ""FILEIN"" then by filename and optionally by ." +Model/TransportModel/tsp-ssm.f90,995,set_iauxpak,TransportModel,store_error,unclassified,?,"(a, a) Auxiliary name cannot be found:" +Model/TransportModel/tsp.f90,606,ftype_check,TransportModel,store_error,unclassified,?,(a) Initial conditions (IC6) package not specified. +Model/TransportModel/tsp.f90,611,ftype_check,TransportModel,store_error,unclassified,?,(a) Discretization (DIS6 or DISU6) package not specified. +Model/TransportModel/tsp.f90,616,ftype_check,TransportModel,store_error,unclassified,?,(a) Mass storage and transfer (MST6) package not specified. +Model/TransportModel/tsp.f90,621,ftype_check,TransportModel,store_error,unclassified,?,(a) Required package(s) not specified. +Distributed/DistributedSim.f90,157,get_load_balance,distributed,store_warning,block-structure,K1,Ignoring PARTITIONS block in HPC file when running a serial process +Distributed/DistributedSim.f90,236,set_load_balance_from_input,distributed,store_error,unclassified,?,HPC input error: undefined model name ( ) +Distributed/DistributedSim.f90,246,set_load_balance_from_input,distributed,store_error,type-parse,K1,"(a,i0,2a) HPC input error: invalid target rank ( ) for model" +Distributed/DistributedSim.f90,265,set_load_balance_from_input,distributed,store_error,unclassified,?,(2a) HPC input error: no target rank for model +Distributed/DistributedSim.f90,284,set_load_balance_from_input,distributed,store_error,unclassified,?,"(a,i0,a) HPC input error: rank has no models assigned" +Distributed/DistributedSim.f90,346,set_load_balance_default,distributed,store_error,not-found-ref,K7,Model type not supported in parallel mode. +Distributed/DistributedSim.f90,473,validate_load_balance,distributed,store_error,unclassified,?,(7a) HPC input error: models and with a coupling have to be assigned to the same rank +Distributed/MpiWorld.f90,125,CHECK_MPI,distributed,store_error,not-input,-,Internal error: +Exchange/DisConnExchange.f90,368,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/DisConnExchange.f90,380,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/DisConnExchange.f90,390,source_data,exchange,store_error,binding-context,K8,Errors encountered in exchange input file. +Exchange/GhostNode.f90,124,gnc_df,exchange,store_error,binding-context,K8,(a) GNC is implicit but models are in different solutions. +Exchange/GhostNode.f90,208,gnc_mc,exchange,store_error,binding-context,K8,"('GHOST NODE ERROR. Cell ', i0, ' in model ', a, ' is not connected to cell ', i0, ' in model ', a)" +Exchange/GhostNode.f90,678,read_options,exchange,store_error,unknown-keyword,K1,"(a,a) Unknown GNC option:" +Exchange/GhostNode.f90,725,read_dimensions,exchange,store_error,unclassified,?,"(a,a) Unknown GNC dimension:" +Exchange/GhostNode.f90,731,read_dimensions,exchange,store_error,block-structure,K1,Required DIMENSIONS block not found. +Exchange/GhostNode.f90,821,read_data,exchange,store_error,unclassified,?,Cell is outside active grid domain: +Exchange/GhostNode.f90,831,read_data,exchange,store_error,unclassified,?,Cell is outside active grid domain: +Exchange/GhostNode.f90,842,read_data,exchange,store_error,unclassified,?,Cell is outside active grid domain: +Exchange/GhostNode.f90,851,read_data,exchange,store_error,unclassified,?,Errors encountered in GNC input file. +Exchange/GhostNode.f90,858,read_data,exchange,store_error,block-structure,K1,(a) Required GNCDATA block not found. +Exchange/GhostNode.f90,883,nodeu_to_noder,exchange,store_error,sign-value,K2,node number < 0 or > model nodes: +Exchange/GwfExchangeMover.f90,80,xmvr_check_packages,exchange,store_error,block-structure,K1,"(a, a, a) ERROR. MODEL AND PACKAGE "" "" DOES NOT HAVE MOVER SPECIFIED IN OPTIONS BLOCK." +Exchange/GwfExchangeMover.f90,93,xmvr_check_packages,exchange,store_error,binding-context,K8,"(a, a, a) The input for an exchange mover lists a model name ("" "") in the PACKAGES input block that is not a member of the exchange." +Exchange/exg-chfgwf.f90,77,chf_gwf_df,exchange,store_error,binding-context,K8,Two models are connected in a SWF-GWF exchange but they are in different solutions. Models must be in same solution: +Exchange/exg-chfgwf.f90,260,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/exg-chfgwf.f90,272,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/exg-chfgwf.f90,282,source_data,exchange,store_error,binding-context,K8,Errors encountered in exchange input file. +Exchange/exg-gwegwe.f90,179,gweexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWE-GWE exchange . First specified GWE Model does not appear to be of the correct type. +Exchange/exg-gwegwe.f90,187,gweexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWE-GWE exchange . Second specified GWE Model does not appear to be of the correct type. +Exchange/exg-gwegwe.f90,212,gwe_gwe_df,exchange,store_error,binding-context,K8,Two models are connect in a GWE exchange but they are in different solutions. GWE models must be in same solution: +Exchange/exg-gwegwe.f90,265,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWE-GWE exchange requires that GWFMODELNAME1 be entered in the OPTIONS block. +Exchange/exg-gwegwe.f90,271,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWE-GWE exchange requires that GWFMODELNAME2 be entered in the OPTIONS block. +Exchange/exg-gwegwe.f90,280,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWE-GWE exchange is a periodic boundary condition which cannot be configured with XT3D +Exchange/exg-gwegwe.f90,294,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWE-GWE exchange requires that ANGLDEGX be specified as an auxiliary variable because dispersion was specified in one or both transport models. +Exchange/exg-gwegwe.f90,303,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWE-GWE exchange requires that ANGLDEGX be specified as an auxiliary variable because XT3D is enabled +Exchange/exg-gwegwe.f90,750,source_options,exchange,store_error,unknown-keyword,K1,Unrecognized input value for ADV_SCHEME option. +Exchange/exg-gwegwe.f90,763,source_options,exchange,store_error,geometry,K10,CND_XT3D_OFF and CND_XT3D_RHS cannot both be set as options. +Exchange/exg-gwegwe.f90,1014,gwe_gwe_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwegwe.f90,1028,gwe_gwe_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwegwe.f90,1125,gwe_gwe_save_simvals,exchange,store_error,unknown-keyword,K1,Unrecognized observation type: +Exchange/exg-gwfgwe.f90,111,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWE exchange . Specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfgwe.f90,118,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWE exchange . Specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfgwe.f90,162,exg_df,exchange,store_error,binding-context,K8,"(3a) Problem with GWF-GWE exchange . The GWF model must be solved by a different IMS than the GWE model. Furthermore, the IMS specified for GWF must be listed in mfsim.nam before the IMS for GWE." +Exchange/exg-gwfgwe.f90,228,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWE Models do not have the same discretization for exchange ',a,'. GWF Model has ', i0, ' user nodes and ', i0, ' reduced nodes. GWE Model has ', i0, ' user nodes and ', i0, ' reduced nodes. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwe.f90,238,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWE Models do not have the same discretization for exchange ',a,'. GWF Model and GWE Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwe.f90,246,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWE Models do not have the same discretization for exchange ',a,'. GWF Model and GWE Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwe.f90,254,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWE Models do not have the same discretization for exchange ',a,'. GWF Model and GWE Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwe.f90,450,gwfconn2gweconn,exchange,store_error,binding-context,K8,(/6a) Missing GWF-GWF exchange when connecting GWE model with exchange to GWF model +Exchange/exg-gwfgwf.f90,187,gwfexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWF exchange . First specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfgwf.f90,195,gwfexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWF exchange . Second specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfgwf.f90,220,gwf_gwf_df,exchange,store_error,binding-context,K8,Two models are connected in a GWF exchange but they are in different solutions. GWF models must be in same solution: +Exchange/exg-gwfgwf.f90,285,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange is a periodic boundary condition which cannot be configured with XT3D +Exchange/exg-gwfgwf.f90,294,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange requires that ANGLDEGX be specified as an auxiliary variable because XT3D is enabled +Exchange/exg-gwfgwf.f90,323,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange requires that ANGLDEGX be specified as an auxiliary variable because K22 was specified in one or both groundwater models. +Exchange/exg-gwfgwf.f90,337,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange requires that ANGLDEGX be specified as an auxiliary variable because specific discharge is being calculated in one or both groundwater models. +Exchange/exg-gwfgwf.f90,345,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange requires that CDIST be specified as an auxiliary variable because specific discharge is being calculated in one or both groundwater models. +Exchange/exg-gwfgwf.f90,355,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWF-GWF exchange requires that the Viscosity Package is inactive in both of the connected models. +Exchange/exg-gwfgwf.f90,814,gwf_gwf_set_flow_to_npf,exchange,store_error,unclassified,?,error in gwf_gwf_cq +Exchange/exg-gwfgwf.f90,829,gwf_gwf_set_flow_to_npf,exchange,store_error,unclassified,?,error in gwf_gwf_cq +Exchange/exg-gwfgwf.f90,842,gwf_gwf_set_flow_to_npf,exchange,store_error,unclassified,?,error in gwf_gwf_cq +Exchange/exg-gwfgwf.f90,1264,source_options,exchange,store_error,unknown-keyword,K1,Unrecognized input value for CELL_AVERAGING option. +Exchange/exg-gwfgwf.f90,1353,read_gnc,exchange,store_error,binding-context,K8,"GNC is explicit, but GWF exchange has active newton." +Exchange/exg-gwfgwf.f90,1354,read_gnc,exchange,store_error,binding-context,K8,Add implicit option to GNC or remove NEWTON from GWF exchange. +Exchange/exg-gwfgwf.f90,1361,read_gnc,exchange,store_error,binding-context,K8,Number of exchanges does not match number of GNCs +Exchange/exg-gwfgwf.f90,1374,read_gnc,exchange,store_error,binding-context,K8,"('EXCHANGE NODES ', i0, ' AND ', i0, ' NOT CONSISTENT WITH GNC NODES ', i0, ' AND ', i0)" +Exchange/exg-gwfgwf.f90,1866,gwf_gwf_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwfgwf.f90,1880,gwf_gwf_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwfgwf.f90,2022,gwf_gwf_save_simvals,exchange,store_error,unknown-keyword,K1,Unrecognized observation type: +Exchange/exg-gwfgwt.f90,114,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWT exchange . Specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfgwt.f90,121,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-GWT exchange . Specified GWT Model does not appear to be of the correct type. +Exchange/exg-gwfgwt.f90,165,exg_df,exchange,store_error,binding-context,K8,"(3a) Problem with GWF-GWT exchange . The GWF model must be solved by a different IMS than the GWT model. Furthermore, the IMS specified for GWF must be listed in mfsim.nam before the IMS for GWT." +Exchange/exg-gwfgwt.f90,231,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWT Models do not have the same discretization for exchange ',a,'. GWF Model has ', i0, ' user nodes and ', i0, ' reduced nodes. GWT Model has ', i0, ' user nodes and ', i0, ' reduced nodes. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwt.f90,241,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWT Models do not have the same discretization for exchange ',a,'. GWF Model and GWT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwt.f90,249,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWT Models do not have the same discretization for exchange ',a,'. GWF Model and GWT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwt.f90,257,exg_ar,exchange,store_error,binding-context,K8,"('GWF and GWT Models do not have the same discretization for exchange ',a,'. GWF Model and GWT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfgwt.f90,447,gwfconn2gwtconn,exchange,store_error,binding-context,K8,Cannot find GWF-GWF exchange when connecting GWT model with exchange to GWF model . Note: GWF-GWF and GWT-GWT need identical exchange data (both in value and order) for the match to succeed. +Exchange/exg-gwfprt.f90,101,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-PRT exchange . Specified GWF Model does not appear to be of the correct type. +Exchange/exg-gwfprt.f90,108,set_model_pointers,exchange,store_error,binding-context,K8,(3a) Problem with GWF-PRT exchange . Specified PRT Model does not appear to be of the correct type. +Exchange/exg-gwfprt.f90,153,exg_df,exchange,store_error,binding-context,K8,(3a) Problem with GWF-PRT exchange . The GWF model must be solved by a different solution than the PRT model. The IMS specified for GWF must be listed in mfsim.nam before the EMS for PRT. +Exchange/exg-gwfprt.f90,224,exg_ar,exchange,store_error,binding-context,K8,"('GWF and PRT Models do not have the same discretization for exchange ',a,'. GWF Model has ', i0, ' user nodes and ', i0, ' reduced nodes. PRT Model has ', i0, ' user nodes and ', i0, ' reduced nodes. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfprt.f90,234,exg_ar,exchange,store_error,binding-context,K8,"('GWF and PRT Models do not have the same discretization for exchange ',a,'. GWF Model and PRT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfprt.f90,242,exg_ar,exchange,store_error,binding-context,K8,"('GWF and PRT Models do not have the same discretization for exchange ',a,'. GWF Model and PRT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwfprt.f90,250,exg_ar,exchange,store_error,binding-context,K8,"('GWF and PRT Models do not have the same discretization for exchange ',a,'. GWF Model and PRT Model have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Exchange/exg-gwtgwt.f90,177,gwtexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWT-GWT exchange . First specified GWT Model does not appear to be of the correct type. +Exchange/exg-gwtgwt.f90,185,gwtexchange_create,exchange,store_error,binding-context,K8,(3a) Problem with GWT-GWT exchange . Second specified GWT Model does not appear to be of the correct type. +Exchange/exg-gwtgwt.f90,210,gwt_gwt_df,exchange,store_error,binding-context,K8,Two models are connected in a GWT exchange but they are in different solutions. GWT models must be in same solution: +Exchange/exg-gwtgwt.f90,261,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWT-GWT exchange requires that GWFMODELNAME1 be entered in the OPTIONS block. +Exchange/exg-gwtgwt.f90,267,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWT-GWT exchange requires that GWFMODELNAME2 be entered in the OPTIONS block. +Exchange/exg-gwtgwt.f90,276,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWT-GWT exchange is a periodic boundary condition which cannot be configured with XT3D +Exchange/exg-gwtgwt.f90,290,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWT-GWT exchange requires that ANGLDEGX be specified as an auxiliary variable because dispersion was specified in one or both transport models. +Exchange/exg-gwtgwt.f90,299,validate_exchange,exchange,store_error,binding-context,K8,(3a) GWT-GWT exchange requires that ANGLDEGX be specified as an auxiliary variable because XT3D is enabled +Exchange/exg-gwtgwt.f90,746,source_options,exchange,store_error,unknown-keyword,K1,Unrecognized input value for ADV_SCHEME option. +Exchange/exg-gwtgwt.f90,759,source_options,exchange,store_error,geometry,K10,DSP_XT3D_OFF and DSP_XT3D_RHS cannot both be set as options. +Exchange/exg-gwtgwt.f90,1010,gwt_gwt_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwtgwt.f90,1024,gwt_gwt_rp_obs,exchange,store_error,unclassified,?, +Exchange/exg-gwtgwt.f90,1121,gwt_gwt_save_simvals,exchange,store_error,unknown-keyword,K1,Unrecognized observation type: +Exchange/exg-olfgwf.f90,77,olf_gwf_df,exchange,store_error,binding-context,K8,Two models are connected in a SWF-GWF exchange but they are in different solutions. Models must be in same solution: +Exchange/exg-olfgwf.f90,260,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/exg-olfgwf.f90,272,source_data,exchange,store_error,unclassified,?,Cell is outside active grid domain ( ). +Exchange/exg-olfgwf.f90,282,source_data,exchange,store_error,binding-context,K8,Errors encountered in exchange input file. +Exchange/exg-swfgwf.f90,158,initialize,exchange,store_error,binding-context,K8,(7a) Problem with exchange . Specified model does not appear to be of the correct type. +Exchange/exg-swfgwf.f90,166,initialize,exchange,store_error,binding-context,K8,(3a) Problem with SWF-GWF exchange . Specified GWF model does not appear to be of the correct type. +Exchange/exg-swfgwf.f90,994,swf_gwf_save_simvals,exchange,store_error,unknown-keyword,K1,Unrecognized observation type: +Idm/selector/IdmChfDfnSelector.f90,191,chf_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""CHF"" , subcomponent="" ""." +Idm/selector/IdmChfDfnSelector.f90,229,chf_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""CHF"" , subcomponent="" ""." +Idm/selector/IdmDfnSelector.f90,142,idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector component not found; component="" "", subcomponent="" ""." +Idm/selector/IdmDfnSelector.f90,173,idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector component not found; component="" "", subcomponent="" ""." +Idm/selector/IdmDfnSelector.f90,204,idm_subpackages,idm,store_error,not-found-ref,K7,"Idm selector component not found; component="" "", subcomponent="" ""." +Idm/selector/IdmExgDfnSelector.f90,146,exg_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""EXG"" , subcomponent="" ""." +Idm/selector/IdmExgDfnSelector.f90,174,exg_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""EXG"" , subcomponent="" ""." +Idm/selector/IdmGweDfnSelector.f90,227,gwe_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWE"" , subcomponent="" ""." +Idm/selector/IdmGweDfnSelector.f90,273,gwe_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWE"" , subcomponent="" ""." +Idm/selector/IdmGwfDfnSelector.f90,353,gwf_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWF"" , subcomponent="" ""." +Idm/selector/IdmGwfDfnSelector.f90,427,gwf_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWF"" , subcomponent="" ""." +Idm/selector/IdmGwtDfnSelector.f90,245,gwt_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWT"" , subcomponent="" ""." +Idm/selector/IdmGwtDfnSelector.f90,295,gwt_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""GWT"" , subcomponent="" ""." +Idm/selector/IdmOlfDfnSelector.f90,191,olf_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""OLF"" , subcomponent="" ""." +Idm/selector/IdmOlfDfnSelector.f90,229,olf_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""OLF"" , subcomponent="" ""." +Idm/selector/IdmPrtDfnSelector.f90,137,prt_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""PRT"" , subcomponent="" ""." +Idm/selector/IdmPrtDfnSelector.f90,163,prt_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""PRT"" , subcomponent="" ""." +Idm/selector/IdmSimDfnSelector.f90,92,sim_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""SIM"" , subcomponent="" ""." +Idm/selector/IdmSimDfnSelector.f90,108,sim_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""SIM"" , subcomponent="" ""." +Idm/selector/IdmUtlDfnSelector.f90,146,utl_idm_multi_package,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""UTL"" , subcomponent="" ""." +Idm/selector/IdmUtlDfnSelector.f90,174,utl_idm_is_advanced,idm,store_error,not-found-ref,K7,"Idm selector subcomponent not found; component=""UTL"" , subcomponent="" ""." +Utilities/Idm/DefinitionSelect.f90,131,get_param_definition_type,idm,store_error,not-input,-,"(a,a,a,a,a) Input file tag not found: "" "" in block "" ""." +Utilities/Idm/DefinitionSelect.f90,174,get_aggregate_definition_type,idm,store_error,not-found-ref,K7,"(a,a,a,a,a,a,a) Idm aggregate definition not found: . Component="" "", subcomponent="" ""." +Utilities/Idm/IdmLoad.f90,362,load_exchanges,idm,store_error,binding-context,K8,Exchange has invalid (unrecognized) model name(s): +Utilities/Idm/IdmLoad.f90,416,load_exchanges,idm,store_error,binding-context,K8,IDM unimplemented. Dynamic Exchanges not supported. +Utilities/Idm/IdmLoad.f90,595,allocate_simnam_int,idm,store_error,unclassified,?,"(a,a) Idm SIMNAM Load default value setting ''is unhandled for this variable:" +Utilities/Idm/IdmLoad.f90,638,allocate_simnam_param,idm,store_error,not-input,-,"(a,a) IdmLoad allocate simnam param unhandled datatype:" +Utilities/Idm/IdmLogger.f90,421,idm_export_int1d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported int1d export_dim=" +Utilities/Idm/IdmLogger.f90,478,idm_export_int2d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported int2d export_dim=" +Utilities/Idm/IdmLogger.f90,521,idm_export_int3d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported int3d export_dim=" +Utilities/Idm/IdmLogger.f90,611,idm_export_dbl1d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported dbl1d export_dim=" +Utilities/Idm/IdmLogger.f90,668,idm_export_dbl2d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported dbl2d export_dim=" +Utilities/Idm/IdmLogger.f90,712,idm_export_dbl3d,idm,store_error,enum,K2,"(a,i0) EXPORT unsupported dbl3d export_dim=" +Utilities/Idm/InputLoadType.f90,346,build_subpkg_prefix,idm,store_error,block-structure,K1,"Internal IDM error: subpackage load cannot identify package "" "" in model name file packages block." +Utilities/Idm/InputLoadType.f90,476,create_subpkg_list,idm,store_error,not-found-ref,K7,"Identified subpackage is not IDM integrated. Remove dfn subpackage tagline for package "" ""." +Utilities/Idm/InputLoadType.f90,538,dynamic_init,idm,store_error,not-input,-,"(a,a) Programming error. (IDM) PERIOD block not found in ''dynamic package input block dfns:" +Utilities/Idm/LoadContext.f90,203,resolve_context,idm,store_error,unknown-keyword,K1,LoadContext unrecognized load_scope for mempath: +Utilities/Idm/LoadContext.f90,541,in_scope,idm,store_error,not-input,-,LoadContext in_scope needs new case for: / +Utilities/Idm/LoadContext.f90,719,resolve_nfeatures,idm,store_error,range-count,K3,"(a,1x,a,1x,i0,1x,a,1x,i0,a) IDM dimension mismatch: does not match the PACKAGEDATA row count ." +Utilities/Idm/LoadContext.f90,766,resolve_item_nfeatures,idm,store_error,unclassified,?,"(a,1x,a,1x,a) item : DIMENSION is not defined." +Utilities/Idm/ModelPackageInputs.f90,246,modelpkgs_init,idm,store_error,block-structure,K1,"(3a) Models block model type "" "" is not valid." +Utilities/Idm/ModelPackageInputs.f90,319,modelpkgs_create,idm,store_error,not-found-ref,K7,"(a,a,a,a,a) Model package type not supported [model= , type= ]." +Utilities/Idm/ModelPackageInputs.f90,329,modelpkgs_create,idm,store_error,unclassified,?,"(3a) Discretization package not specified for model "" ""." +Utilities/Idm/ModelPackageInputs.f90,434,modelpkgs_pkgcount,idm,store_error,duplicate,K6,"(a,a,a,a,a) Multiple instances specified for model base package type [model= , type= ]." +Utilities/Idm/SourceCommon.f90,79,idm_component_type,idm,store_error,unknown-keyword,K1,"(a) IDP input error, unrecognized component: "" """ +Utilities/Idm/SourceCommon.f90,291,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NLAY"" not found." +Utilities/Idm/SourceCommon.f90,297,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NROW"" not found." +Utilities/Idm/SourceCommon.f90,303,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NCOL"" not found." +Utilities/Idm/SourceCommon.f90,324,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NROW"" not found." +Utilities/Idm/SourceCommon.f90,330,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NCOL"" not found." +Utilities/Idm/SourceCommon.f90,350,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NLAY"" not found." +Utilities/Idm/SourceCommon.f90,356,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NCPL"" not found." +Utilities/Idm/SourceCommon.f90,373,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NODES"" not found." +Utilities/Idm/SourceCommon.f90,396,set_model_shape,idm,store_error,block-structure,K1,"(a) Required input dimension ""NODES"" not found." +Utilities/Idm/SourceCommon.f90,406,set_model_shape,idm,store_error,unclassified,?,"Unknown discretization type. IDM cannot set shape for "" '" +Utilities/Idm/SourceCommon.f90,470,filein_fname,idm,store_error,block-structure,K1,"Multiple FILEIN keywords detected for tag "" "" in OPTIONS block. Only one entry allowed." +Utilities/Idm/SourceCommon.f90,499,inlen_check,idm,store_error,range-count,K3,"(a,i0,a) Input name "" "" exceeds maximum allowed length ( ) for ." +Utilities/Idm/SourceLoad.F90,100,package_loader,idm,store_error,not-found-ref,K7,"(a) Simulation package input source type "" "" not currently supported." +Utilities/Idm/SourceLoad.F90,220,load_simtdis,idm,store_error,not-input,-,"(a) Simulation TIMING input file "" "" does not exist." +Utilities/Idm/SourceLoad.F90,292,remote_model_ndim,idm,store_error,unclassified,?,"(a) Unknown discretization type "" "" not currently supported." +Utilities/Idm/SourceLoad.F90,318,export_cr,idm,store_error,not-found-ref,K7,(a) Model namefile NETCDF_STUCTURED or NETCDF_MESH2D option configured but NetCDF libraries are not available. +Utilities/Idm/SourceLoad.F90,398,netcdf_context,idm,store_error,not-found-ref,K7,(a) Cannot load model packages. NetCDF keyword specified in input file but NetCDF libraries are not available. +Utilities/Idm/mf6blockfile/IdmMf6File.f90,249,dynamic_read_ionper,idm,store_error,block-structure,K1,"(a, i0, a, i0, a, i0, a) Error in stress period . Period numbers not increasing. Found but last period block was assigned ." +Utilities/Idm/mf6blockfile/IdmMf6File.f90,261,dynamic_read_ionper,idm,store_error,enum,K2,"('Looking for BEGIN PERIOD iper. Found ', a, ' instead.')" +Utilities/Idm/mf6blockfile/IdmMf6File.f90,364,open_mf6file,idm,store_error,unclassified,?,"(a,a,a) File unspecified, cannot load model or package type "" ""." +Utilities/Idm/mf6blockfile/LoadMf6File.f90,419,parse_record_tag,idm,store_error,block-structure,K1,"(5a) "" "" block input record that includes keyword "" "" is not properly formed." +Utilities/Idm/mf6blockfile/LoadMf6File.f90,448,parse_record_tag,idm,store_error,unclassified,?,"(5a) Expecting record input tag "" "" but instead found "" ""." +Utilities/Idm/mf6blockfile/LoadMf6File.f90,522,load_tag,idm,store_error,unclassified,?,"(a,a) Failure reading data for tag:" +Utilities/Idm/mf6blockfile/LoadMf6File.f90,646,parse_structarray_block,idm,store_error,block-structure,K1,"(3a) Required dimension "" "" not found." +Utilities/Idm/mf6blockfile/LoadMf6File.f90,667,parse_structarray_block,idm,store_error,range-count,K3,"(5a) Shape variable "" "" rank does not match its declared shape "" ""." +Utilities/Idm/mf6blockfile/LoadMf6File.f90,1294,read_control_record,idm,store_error,not-input,-,"('Specified OPEN/CLOSE file ',(A),' does not exist')" +Utilities/Idm/mf6blockfile/LoadMf6File.f90,1295,read_control_record,idm,store_error,not-input,-,Specified OPEN/CLOSE file does not exist +Utilities/Idm/mf6blockfile/LoadMf6File.f90,1395,alloc_tslink,idm,store_error,block-structure,K1,"(a) Time series link identifier not defined for block "" ""." +Utilities/Idm/mf6blockfile/LoadNCInput.F90,40,error_and_exit,idm,store_error,unclassified,?,"(a) Cannot load NetCDF array for variable "" "". NetCDF libraries are not available. Ensure Extended MODFLOW 6 in use." +Utilities/Idm/mf6blockfile/Mf6FileGridArray.f90,261,param_load,idm,store_error,range-count,K3,"(a,i0,a) Input error: number of defined (non-DNODATA) cells exceeds MAXBOUND= ." +Utilities/Idm/mf6blockfile/Mf6FileGridArray.f90,302,param_load,idm,store_error,range-count,K3,"(a,i0,a) Input error: number of defined (non-DNODATA) cells exceeds MAXBOUND= ." +Utilities/Idm/mf6blockfile/Mf6FileGridArray.f90,315,param_load,idm,store_error,unclassified,?,IDM unimplemented. GridArrayLoad::param_load datatype= +Utilities/Idm/mf6blockfile/Mf6FileKeystring.f90,381,valid_address,idm,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,a) on row must be greater than 0 and less than or equal to ." +Utilities/Idm/mf6blockfile/Mf6FileKeystring.f90,751,resolve_subindex_address,idm,store_error,range-count,K3,"(a,1x,a,1x,i0,1x,a,1x,i0,1x,a,1x,a,1x,i0,a) index must be between 1 and for ." +Utilities/Idm/mf6blockfile/Mf6FileLayerArray.f90,358,param_load,idm,store_error,unclassified,?,IDM unimplemented. LayerArrayLoad::param_load datatype= +Utilities/Idm/mf6blockfile/Mf6FileList.f90,317,apply_persistent_settings,idm,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,i0,1x,a,1x,i0,a) on row must be greater than 0 and less than or equal to ." +Utilities/Idm/mf6blockfile/Mf6FileSetting.f90,174,set_idts,idm,store_error,not-found-ref,K7,Internal IDM error: trailing setting param not found +Utilities/Idm/mf6blockfile/StructArray.f90,191,mem_create_vector,idm,store_error,not-input,-,IDM unimplemented. StructArray::mem_create_vector type= +Utilities/Idm/mf6blockfile/StructArray.f90,247,idm_input_varname,idm,store_error,range-count,K3,"(*(G0)) IDM mf6internal "" "" exceeds characters (IDM reserves for the internal input-array suffix "" "")." +Utilities/Idm/mf6blockfile/StructArray.f90,503,allocate_dbl1d_type,idm,store_error,not-input,-,"IDM unimplemented. StructArray::allocate_dbl1d_type unsupported shape "" ""." +Utilities/Idm/mf6blockfile/StructArray.f90,659,load_deferred_vector,idm,store_error,not-input,-,StructArray::load_deferred_vector intvector reallocate unimplemented. +Utilities/Idm/mf6blockfile/StructArray.f90,664,load_deferred_vector,idm,store_error,not-input,-,StructArray::load_deferred_vector int2d reallocate unimplemented. +Utilities/Idm/mf6blockfile/StructArray.f90,683,load_deferred_vector,idm,store_error,not-input,-,StructArray::load_deferred_vector dbl2d reallocate unimplemented. +Utilities/Idm/mf6blockfile/StructArray.f90,909,check_reallocate,idm,store_error,not-input,-,IDM unimplemented. StructArray::check_reallocate unsupported memtype. +Utilities/Idm/mf6blockfile/StructArray.f90,932,read_param,idm,store_error,unclassified,?,"(a,i0) IDM read_param called for MTYPE_UNDEF metadata vector at column" +Utilities/Idm/mf6blockfile/StructArray.f90,1015,read_param,idm,store_error,cellid-grid,K7,"(a,a,a) CELLID must be an integer or the keyword NONE, got: ." +Utilities/Idm/mf6blockfile/StructArray.f90,1114,read_from_parser,idm,store_error,range-count,K3,"(a,i0,a) Input error: line count exceeds input dimension. Expected rows= ." +Utilities/Idm/mf6blockfile/StructArray.f90,1198,read_from_parser_keystring,idm,store_error,not-input,-,"(a,i0,a) Input error: keystring row count exceeds pre-allocated maxbound= ." +Utilities/Idm/mf6blockfile/StructArray.f90,1237,read_from_parser_keystring,idm,store_error,unknown-keyword,K1,"(a,a,a) Unrecognized keystring keyword "" "" in PERIOD block." +Utilities/Idm/mf6blockfile/StructArray.f90,1315,read_from_binary,idm,store_error,option-coherence,K5,IDM unimplemented. StructArray::read_from_binary deferred shape not supported for binary inputs. +Utilities/Idm/mf6blockfile/StructArray.f90,1334,read_from_binary,idm,store_error,option-coherence,K5,"List style binary inputs not supported for text columns, tag= ." +Utilities/Idm/mf6blockfile/StructArray.f90,1340,read_from_binary,idm,store_error,option-coherence,K5,"List style binary inputs not supported for self-sizing (ragged) columns, tag= ." +Utilities/Idm/mf6blockfile/StructArray.f90,1378,read_from_binary,idm,store_error,not-input,-,"('Error reading LIST from file: ',1x,a,1x,' on UNIT: ',I0)" +Utilities/Idm/netcdf/NCArrayReader.f90,318,load_integer1d_spd,idm,store_error,not-found-ref,K7,"(a,a,a) Timeseries netcdf input read not supported for full grid int1d type ( )." +Utilities/Idm/netcdf/NCArrayReader.f90,440,load_integer2d_layered,idm,store_error,option-coherence,K5,"(a,a,a) Layered netcdf read not supported for DIS int2d type ( )." +Utilities/Idm/netcdf/NCArrayReader.f90,731,load_double2d_layered,idm,store_error,option-coherence,K5,"(a,a,a) Layered netcdf read not supported for DIS dbl2d type ( )." +Utilities/Idm/netcdf/NCContextBuild.f90,41,open_ncfile,idm,store_error,not-input,-,"(a,a,a) Specified NetCDF input file does not exist [file= ]." +Utilities/Idm/netcdf/NCContextBuild.f90,112,add_package_var,idm,store_error,type-parse,K1,"NetCDF variable invalid modflow_input attribute: "" ""." +Utilities/Idm/netcdf/NCContextBuild.f90,152,verify_global_attr,idm,store_error,enum,K2,"NetCDF unsupported mesh type: "" ""." +Utilities/Idm/netcdf/NCContextBuild.f90,160,verify_global_attr,idm,store_warning,unclassified,?,"Verify ""modflow_grid"" and ""modflow_mesh"" global attributes in file:" +Utilities/Idm/netcdf/NCContextBuild.f90,165,verify_global_attr,idm,store_error,not-input,-,"NetCDF input file global attribute ""modflow_grid"" not found." +Utilities/Idm/netcdf/NCFileVars.f90,118,ncvars_varid,idm,store_error,unclassified,?,(a) . +Utilities/Idm/netcdf/NCFileVars.f90,124,ncvars_varid,idm,store_error,not-found-ref,K7,"(a) NetCDF variable not found, tagname="" "". NetCDF input not provided for model "" ""." +Utilities/Idm/netcdf/NetCDFCommon.f90,105,nf_verify,idm,store_error,not-input,-,"(a,i0,a) NetCDF library error [exit code= ]." +Solution/LinearMethods/ImsLinear.f90,185,imslinear_ar,solution,store_error,sign-value,K2,(A) IMSLINEAR7AR ISCL MUST BE <= 2 +Solution/LinearMethods/ImsLinear.f90,190,imslinear_ar,solution,store_error,sign-value,K2,(A) IMSLINEAR7AR IORD MUST BE <= 2 +Solution/LinearMethods/ImsLinear.f90,194,imslinear_ar,solution,store_error,unclassified,?,(A) IMSLINEAR7AR NORTH MUST >= 0 +Solution/LinearMethods/ImsLinear.f90,199,imslinear_ar,solution,store_error,sign-value,K2,(A) IMSLINEAR7AR RCLOSE MUST > 0.0 +Solution/LinearMethods/ImsLinear.f90,204,imslinear_ar,solution,store_error,sign-value,K2,(A) IMSLINEAR7AR RELAX MUST BE >= 0.0 +Solution/LinearMethods/ImsLinear.f90,208,imslinear_ar,solution,store_error,sign-value,K2,(A) IMSLINEAR7AR RELAX MUST BE <= 1.0 +Solution/LinearMethods/ImsLinearBase.f90,595,ims_base_calc_order,solution,store_error,unclassified,?,"(A,1X,A) IMSLINEARSUB_CALC_ORDER error creating minimum degree order permutation" +Solution/LinearMethods/ImsLinearBase.f90,835,ims_base_pcu,solution,store_error,unclassified,?,"(a,1x,a) ILUT:" +Solution/LinearMethods/ImsLinearSettings.f90,183,read_from_file,solution,store_error,unclassified,?,(3a) Unknown IMSLINEAR LINEAR_ACCELERATION method ( ). +Solution/LinearMethods/ImsLinearSettings.f90,197,read_from_file,solution,store_error,unclassified,?,(3a) Unknown IMSLINEAR SCALING_METHOD ( ). +Solution/LinearMethods/ImsLinearSettings.f90,214,read_from_file,solution,store_error,unclassified,?,(3a) Unknown IMSLINEAR REORDERING_METHOD ( ). +Solution/LinearMethods/ImsLinearSettings.f90,227,read_from_file,solution,store_error,sign-value,K2,"(a,1x,a) IMSLINEAR PRECONDITIONER_LEVELS must be greater than or equal to zero" +Solution/LinearMethods/ImsLinearSettings.f90,235,read_from_file,solution,store_error,sign-value,K2,"(a,1x,a) IMSLINEAR PRECONDITIONER_DROP_TOLERANCE must be greater than or equal to zero" +Solution/LinearMethods/ImsLinearSettings.f90,242,read_from_file,solution,store_error,unknown-keyword,K1,(3a) Unknown IMSLINEAR keyword ( ). +Solution/LinearMethods/ImsLinearSettings.f90,249,read_from_file,solution,store_error,block-structure,K1,(a) NO LINEAR block detected. +Solution/LinearMethods/ImsLinearSettings.f90,265,check_settings,solution,store_warning,sign-value,K2,(a) PRECONDITIONER_DROP_TOLERANCE is ignored because PRECONDITIONER_LEVELS equals zero. +Solution/NumericalSolution.f90,599,sln_ar,solution,store_error,unknown-keyword,K1,(3a) Unknown IMS print option ( ). +Solution/NumericalSolution.f90,615,sln_ar,solution,store_error,unknown-keyword,K1,(3a) Unknown IMS COMPLEXITY option ( ). +Solution/NumericalSolution.f90,631,sln_ar,solution,store_error,unclassified,?,(a) Optional CSV_OUTER_OUTPUT keyword must be followed by FILEOUT +Solution/NumericalSolution.f90,647,sln_ar,solution,store_error,unclassified,?,(a) Optional CSV_INNER_OUTPUT keyword must be followed by FILEOUT +Solution/NumericalSolution.f90,670,sln_ar,solution,store_error,range-count,K3,"(a,G0) Value for ATS_OUTER_MAXIMUM_FRAC must be between 0 and 0.5. Found" +Solution/NumericalSolution.f90,701,sln_ar,solution,store_error,unclassified,?,(a) Optional PTC_OUTPUT keyword must be followed by FILEOUT +Solution/NumericalSolution.f90,715,sln_ar,solution,store_error,sign-value,K2,(a) PTC_EXPONENT must be > 0. +Solution/NumericalSolution.f90,727,sln_ar,solution,store_error,sign-value,K2,(a) IMS sln_ar: PTC_DEL0 must be > 0. +Solution/NumericalSolution.f90,737,sln_ar,solution,store_error,unknown-keyword,K1,"(a,2(1x,a)) Unknown IMS option ( )." +Solution/NumericalSolution.f90,787,sln_ar,solution,store_error,enum,K2,(3a) Unknown UNDER_RELAXATION specified ( ). +Solution/NumericalSolution.f90,799,sln_ar,solution,store_error,enum,K2,(3a) Unknown LINEAR_SOLVER specified ( ). +Solution/NumericalSolution.f90,822,sln_ar,solution,store_error,unknown-keyword,K1,(3a) Unknown IMS NONLINEAR keyword ( ). +Solution/NumericalSolution.f90,829,sln_ar,solution,store_error,block-structure,K1,(a) NO IMS NONLINEAR block detected. +Solution/NumericalSolution.f90,845,sln_ar,solution,store_error,sign-value,K2,(a) Outer iteration number must be > 0. +Solution/NumericalSolution.f90,858,sln_ar,solution,store_error,unclassified,?,(a) Incorrect value for variable NONMETH was specified. +Solution/NumericalSolution.f90,866,sln_ar,solution,store_error,sign-value,K2,(a) GAMMA must be greater than zero if SIMPLE under-relaxation is used. +Solution/NumericalSolution.f90,904,sln_ar,solution,store_error,binding-context,K8,(a) Incorrect value for linear solution method specified. +Solution/NumericalSolution.f90,923,sln_ar,solution,store_error,geometry,K10,"(a,' **',a,'** PRODUCES AN ASYMMETRIC COEFFICIENT MATRIX, BUT THE CONJUGATE GRADIENT METHOD WAS SELECTED. USE BICGSTAB INSTEAD. ') MODEL" +Solution/NumericalSolution.f90,932,sln_ar,solution,store_error,binding-context,K8,"(a,' **',a,'** PRODUCES AN ASYMMETRIC COEFFICIENT MATRIX, BUT THE CONJUGATE GRADIENT METHOD WAS SELECTED. USE BICGSTAB INSTEAD. ') EXCHANGE" +Solution/NumericalSolution.f90,952,sln_ar,solution,store_error,binding-context,K8,"(2(a,1x)) dependent_variable_scaling must be specified for all models in the solution and can only be used with GWT and GWE models." +Solution/PETSc/PetscSolver.F90,80,create_petsc_solver,solution,store_error,not-input,-,"(a,a) PETSc solver not supported for run mode:" +Solution/PETSc/PetscSolver.F90,139,petsc_check_settings,solution,store_error,not-input,-,"(a,a) PETSc: unknown linear solver method in" +Solution/PETSc/PetscSolver.F90,146,petsc_check_settings,solution,store_warning,not-input,-,(a) PETSc: IMS reordering not supported +Solution/PETSc/PetscSolver.F90,151,petsc_check_settings,solution,store_warning,not-input,-,(a) PETSc: IMS matrix scaling not supported +Solution/PETSc/PetscSolver.F90,156,petsc_check_settings,solution,store_warning,not-input,-,(a) PETSc: IMS orthogonalization not supported +Solution/PETSc/PetscSolver.F90,378,petsc_solve,solution,store_error,not-input,-,"(1x,3a,i0) PETSc convergence failure in :" +Timing/ats.f90,233,ats_read_options,timing,store_error,unknown-keyword,K1,"(a,a) Unknown ATS option:" +Timing/ats.f90,275,ats_read_dimensions,timing,store_error,temporal,K9,"(a,a) Unknown ATS dimension:" +Timing/ats.f90,282,ats_read_dimensions,timing,store_error,block-structure,K1,(a) Required DIMENSIONS block not found. +Timing/ats.f90,332,ats_read_timing,timing,store_error,block-structure,K1,(a) Required PERIODDATA block not found. +Timing/ats.f90,428,ats_check_timing,timing,store_error,sign-value,K2,"(a, i0, a, i0) IPERATS must be greater than zero. Found for ATS PERIODDATA record" +Timing/ats.f90,434,ats_check_timing,timing,store_warning,temporal,K9,"(a, i0, a, i0) IPERATS greater than NPER. Found for ATS PERIODDATA record" +Timing/ats.f90,442,ats_check_timing,timing,store_error,sign-value,K2,"(a, g15.7, a, i0) DT0 must be >= zero. Found for ATS PERIODDATA record" +Timing/ats.f90,450,ats_check_timing,timing,store_error,sign-value,K2,"(a, g15.7, a, i0) DTMIN must be > zero. Found for ATS PERIODDATA record" +Timing/ats.f90,458,ats_check_timing,timing,store_error,sign-value,K2,"(a, g15.7, a, i0) DTMAX must be > zero. Found for ATS PERIODDATA record" +Timing/ats.f90,466,ats_check_timing,timing,store_error,range-count,K3,"(a, 2g15.7, a, i0) DTMIN must be < dtmax. Found for ATS PERIODDATA record" +Timing/ats.f90,474,ats_check_timing,timing,store_error,temporal,K9,"(a, g15.7, a, i0) DTADJ must be 0 or >= 1.0. Found for ATS PERIODDATA record" +Timing/ats.f90,482,ats_check_timing,timing,store_error,temporal,K9,"(a, g15.7, a, i0) DTFAILADJ must be 0 or >= 1.0. Found for ATS PERIODDATA record" +Timing/tdis.f90,414,tdis_source_options,timing,store_error,unknown-keyword,K1,Unrecognized input value for TIME_UNITS option. +Timing/tdis.f90,632,check_tdis_timing,timing,store_error,temporal,K9,"(A,' for stress period ', I0) Number of time steps less than one" +Timing/tdis.f90,645,check_tdis_timing,timing,store_error,sign-value,K2,"(A,' for stress period ', I0) TSMULT must be greater than 0.0" +Timing/tdis.f90,652,check_tdis_timing,timing,store_error,sign-value,K2,"(A,' for stress period ', I0) PERLEN cannot be less than 0.0" +Timing/tdis.f90,671,check_tdis_timing,timing,store_error,geometry,K10,"('Time step length of ', G0, ' is too small in period ', I0, ' and time step ', I0)" +Model/NumericalPackage.f90,89,set_names,top,store_error,unclassified,?,Package name contains spaces: +Model/NumericalPackage.f90,91,set_names,top,store_error,unclassified,?,Remove spaces from name. +Model/NumericalPackage.f90,218,read_check_ionper,top,store_error,block-structure,K1,"(a, i0, a, i0, a, i0, a) Error in stress period . Period numbers not increasing. Found but last period block was assigned ." +Model/NumericalPackage.f90,283,get_block_data,top,store_error,unknown-keyword,K1,"(a,a) Unknown GRIDDATA tag:" +RunControlFactory.F90,32,create_run_control,top,store_error,not-input,-,(a) Can not run parallel mode with this executable: no MPI +SimulationCreate.f90,132,options_create,top,store_error,unclassified,?, +SimulationCreate.f90,191,timing_create,top,store_error,block-structure,K1,TIMING block variable TDIS6 is unset in simulation control input. +SimulationCreate.f90,343,models_create,top,store_error,unclassified,?,"(a,a) Unknown simulation model type:" +SimulationCreate.f90,354,models_create,top,store_error,unclassified,?,"(a, i0) No MODELS assigned to process" +SimulationCreate.f90,435,exchanges_create,top,store_error,not-found-ref,K7,"('Error in simulation control ', 'file. Could not find model: ', a)" +SimulationCreate.f90,440,exchanges_create,top,store_error,not-found-ref,K7,"('Error in simulation control ', 'file. Could not find model: ', a)" +SimulationCreate.f90,501,exchanges_create,top,store_error,binding-context,K8,"(a,a) Unknown simulation exchange type:" +SimulationCreate.f90,533,solution_group_check,top,store_error,binding-context,K8,"(a,i0) There are no solutions for solution group" +SimulationCreate.f90,539,solution_group_check,top,store_error,binding-context,K8,"('MXITER is set to ', i0, ' but there is only one solution', ' in SOLUTION GROUP ', i0, '. Set MXITER to 1 in simulation control', ' file.')" +SimulationCreate.f90,636,solution_groups_create,top,store_error,binding-context,K8,"(a,i0,a,i0,a) Solution group blocks are not listed consecutively. Found when looking for ." +SimulationCreate.f90,668,solution_groups_create,top,store_error,not-found-ref,K7,"(a,a) Invalid model name:" +SimulationCreate.f90,688,solution_groups_create,top,store_error,binding-context,K8,"(4a) Model "" "" is an explicit model and cannot be added to an IMS6 solution. Explicit models require EMS6." +SimulationCreate.f90,720,solution_groups_create,top,store_error,not-found-ref,K7,"(a,a) Invalid model name:" +SimulationCreate.f90,740,solution_groups_create,top,store_error,binding-context,K8,"(4a) Model "" "" is a numerical model and cannot be added to an EMS6 solution. Numerical models require IMS6." +SimulationCreate.f90,767,solution_groups_create,top,store_error,binding-context,K8,There are no solution groups. +SimulationCreate.f90,784,check_model_assignment,top,store_error,binding-context,K8,"(a,a) Model was not assigned to a solution:" +SimulationCreate.f90,847,check_model_name,top,store_error,not-found-ref,K7,"(a,a) Invalid model name:" +SimulationCreate.f90,851,check_model_name,top,store_error,range-count,K3,"(a,i0,a,i0) Name length of exceeds maximum length of" +SimulationCreate.f90,856,check_model_name,top,store_error,not-found-ref,K7,"(a,a) Invalid model name:" +SimulationCreate.f90,859,check_model_name,top,store_error,unclassified,?,(a) Model name cannot have spaces within it. +SimulationCreate.f90,864,check_model_name,top,store_error,not-found-ref,K7,"(a,a) Invalid model name:" +SimulationCreate.f90,867,check_model_name,top,store_error,duplicate,K6,(a) Model names must be unique. +Model/ModelUtilities/BoundaryPackage.f90,338,bnd_rp,utilities,store_error,enum,K2,"('Looking for BEGIN PERIOD iper. Found ', a, ' instead.')" +Model/ModelUtilities/BoundaryPackage.f90,1242,bnd_read_options,utilities,store_error,unclassified,?,"TS6 keyword must be followed by ""FILEIN"" then by filename." +Model/ModelUtilities/BoundaryPackage.f90,1252,bnd_read_options,utilities,store_error,option-coherence,K5,TAS6 FILE cannot be used with selected discretization type. +Model/ModelUtilities/BoundaryPackage.f90,1257,bnd_read_options,utilities,store_error,not-found-ref,K7,The package does not support TIMEARRAYSERIESFILE +Model/ModelUtilities/BoundaryPackage.f90,1264,bnd_read_options,utilities,store_error,unclassified,?,"TAS6 keyword must be followed by ""FILEIN"" then by filename." +Model/ModelUtilities/BoundaryPackage.f90,1280,bnd_read_options,utilities,store_error,unclassified,?,"OBS6 keyword must be followed by ""FILEIN"" then by filename." +Model/ModelUtilities/BoundaryPackage.f90,1286,bnd_read_options,utilities,store_error,block-structure,K1,Multiple OBS6 keywords detected in OPTIONS block. Only one OBS6 entry allowed for a package. +Model/ModelUtilities/BoundaryPackage.f90,1312,bnd_read_options,utilities,store_error,unknown-keyword,K1,"(a,3(1x,a)) UNKNOWN OPTION:" +Model/ModelUtilities/BoundaryPackage.f90,1331,bnd_read_options,utilities,store_error,unclassified,?,"(a,2(1x,a)) AUXMULTNAME was specified as but no AUX variables specified." +Model/ModelUtilities/BoundaryPackage.f90,1348,bnd_read_options,utilities,store_error,unclassified,?,"(a,2(1x,a)) AUXMULTNAME was specified as but no AUX variable found with this name." +Model/ModelUtilities/BoundaryPackage.f90,1392,bnd_read_dimensions,utilities,store_error,unclassified,?,"(a,3(1x,a)) Unknown dimension:" +Model/ModelUtilities/BoundaryPackage.f90,1399,bnd_read_dimensions,utilities,store_error,block-structure,K1,Required DIMENSIONS block not found. +Model/ModelUtilities/BoundaryPackage.f90,1406,bnd_read_dimensions,utilities,store_error,sign-value,K2,(a) MAXBOUND must be an integer greater than zero. +Model/ModelUtilities/BoundaryPackageExt.f90,244,validate_ifno,utilities,store_error,sign-value,K2,"(a,1x,a,1x,a,i0,a,1x,i0,a) ( ) must be greater than 0 and less than or equal to ." +Model/ModelUtilities/BoundaryPackageExt.f90,276,report_ifno_coverage,utilities,store_error,unclassified,?,"(a,1x,a,1x,a,1x,i0,a) no data specified for ." +Model/ModelUtilities/BoundaryPackageExt.f90,281,report_ifno_coverage,utilities,store_error,unclassified,?,"(a,1x,a,1x,a,1x,i0,1x,a,1x,i0,1x,a) data for specified times." +Model/ModelUtilities/BoundaryPackageExt.f90,518,source_options,utilities,store_error,unclassified,?,"(a,2(1x,a)) AUXMULTNAME was specified as but no AUX variables specified." +Model/ModelUtilities/BoundaryPackageExt.f90,535,source_options,utilities,store_error,unclassified,?,"(a,2(1x,a)) AUXMULTNAME was specified as but no AUX variable found with this name." +Model/ModelUtilities/BoundaryPackageExt.f90,543,source_options,utilities,store_error,option-coherence,K5,READASARRAYS option is not compatible with selected discretization type. +Model/ModelUtilities/BoundaryPackageExt.f90,647,source_dimensions,utilities,store_error,sign-value,K2,(a) MAXBOUND must be an integer greater than zero. +Model/ModelUtilities/BoundaryPackageExt.f90,699,cellid_to_nlist,utilities,store_error,unclassified,?,Cell is outside active grid domain: +Model/ModelUtilities/BoundaryPackageExt.f90,710,cellid_to_nlist,utilities,store_error,unclassified,?,errors encountered. +Model/ModelUtilities/BoundaryPackageExt.f90,856,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Model/ModelUtilities/BoundaryPackageExt.f90,866,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Model/ModelUtilities/BoundaryPackageExt.f90,877,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Model/ModelUtilities/Connections.f90,212,con_finalize,utilities,store_error,unclassified,?,"('Error in array: ',a,'.', ' First value for cell : ',i0,' must equal ',i0,'.', ' Found ',i0,' instead.')" +Model/ModelUtilities/Connections.f90,219,con_finalize,utilities,store_error,unclassified,?,"('Error in array: ',a,'.', ' Entries not sorted for row: ',i0,'.', ' Offending entries are: ',i0,' and ',i0)" +Model/ModelUtilities/Connections.f90,237,con_finalize,utilities,store_error,geometry,K10,"('Error in array: ',a,'.', ' Array does not have symmetric counterpart in position ',i0, ' for cell ',i0,' connected to cell ',i0)" +Model/ModelUtilities/Connections.f90,255,con_finalize,utilities,store_error,geometry,K10,"(/,'Error in array: ',(a),/, 'Array is not symmetric in positions: ',2i9,/, 'Values in these positions are: ', 2(1pg15.6))" +Model/ModelUtilities/Connections.f90,289,con_finalize,utilities,store_error,geometry,K10,"(/,'Error in array: ',(a),/, 'Array is not symmetric in positions: ',2i9,/, 'Values in these positions are: ', 2(1pg15.6))" +Model/ModelUtilities/Connections.f90,293,con_finalize,utilities,store_error,geometry,K10,"(/,'IHC must be 0, 1, or 2. Found: ',i0)" +Model/ModelUtilities/Connections.f90,381,read_connectivity_from_block,utilities,store_error,unknown-keyword,K1,"(a,a) Unknown CONNECTIONDATA tag:" +Model/ModelUtilities/Connections.f90,387,read_connectivity_from_block,utilities,store_error,block-structure,K1,Required CONNECTIONDATA block not found. +Model/ModelUtilities/Connections.f90,396,read_connectivity_from_block,utilities,store_error,unclassified,?,"(a,a) Required input was not specified:" +Model/ModelUtilities/DiscretizationBase.f90,133,dis_df,utilities,store_error,not-input,-,Programmer error: dis_df must be overridden +Model/ModelUtilities/DiscretizationBase.f90,205,write_grb,utilities,store_error,not-input,-,Programmer error: write_grb must be overridden +Model/ModelUtilities/DiscretizationBase.f90,257,nodeu_to_string,utilities,store_error,not-input,-,Programmer error: nodeu_to_string must be overridden +Model/ModelUtilities/DiscretizationBase.f90,267,nodeu_to_array,utilities,store_error,not-input,-,Programmer error: nodeu_to_array must be overridden +Model/ModelUtilities/DiscretizationBase.f90,291,get_nodenumber_idx1,utilities,store_error,not-input,-,Programmer error: get_nodenumber_idx1 must be overridden +Model/ModelUtilities/DiscretizationBase.f90,302,get_nodenumber_idx2,utilities,store_error,not-input,-,Programmer error: get_nodenumber_idx2 must be overridden +Model/ModelUtilities/DiscretizationBase.f90,313,get_nodenumber_idx3,utilities,store_error,not-input,-,Programmer error: get_nodenumber_idx3 must be overridden +Model/ModelUtilities/DiscretizationBase.f90,330,connection_normal,utilities,store_error,not-input,-,Programmer error: connection_normal must be overridden +Model/ModelUtilities/DiscretizationBase.f90,351,connection_vector,utilities,store_error,not-input,-,Programmer error: connection_vector must be overridden +Model/ModelUtilities/DiscretizationBase.f90,388,get_dis_type,utilities,store_error,not-input,-,Programmer error: get_dis_type must be overridden +Model/ModelUtilities/DiscretizationBase.f90,399,get_dis_enum,utilities,store_error,not-input,-,Programmer error: get_dis_enum must be overridden +Model/ModelUtilities/DiscretizationBase.f90,511,nodeu_from_string,utilities,store_error,not-input,-,Programmer error: nodeu_from_string must be overridden +Model/ModelUtilities/DiscretizationBase.f90,535,nodeu_from_cellid,utilities,store_error,not-input,-,Programmer error: nodeu_from_cellid must be overridden +Model/ModelUtilities/DiscretizationBase.f90,582,noder_from_string,utilities,store_error,unclassified,?,Cell is outside active grid domain: +Model/ModelUtilities/DiscretizationBase.f90,636,noder_from_cellid,utilities,store_error,unclassified,?,Cell is outside active grid domain: +Model/ModelUtilities/DiscretizationBase.f90,644,noder_from_cellid,utilities,store_error,not-input,-,Programmer error: supports_layers must be overridden +Model/ModelUtilities/DiscretizationBase.f90,654,get_ncpl,utilities,store_error,not-input,-,Programmer error: get_ncpl must be overridden +Model/ModelUtilities/DiscretizationBase.f90,689,get_polyverts,utilities,store_error,not-input,-,Programmer error: get_polyverts must be overridden +Model/ModelUtilities/DiscretizationBase.f90,700,get_npolyverts,utilities,store_error,not-input,-,Programmer error: get_npolyverts must be overridden +Model/ModelUtilities/DiscretizationBase.f90,710,get_max_npolyverts,utilities,store_error,not-input,-,Programmer error: get_max_npolyverts must be overridden +Model/ModelUtilities/DiscretizationBase.f90,728,read_int_array,utilities,store_error,not-input,-,Programmer error: read_int_array must be overridden +Model/ModelUtilities/DiscretizationBase.f90,746,read_dbl_array,utilities,store_error,not-input,-,Programmer error: read_dbl_array must be overridden +Model/ModelUtilities/DiscretizationBase.f90,919,read_list,utilities,store_error,unclassified,?,Cell is outside active grid domain: +Model/ModelUtilities/DiscretizationBase.f90,927,read_list,utilities,store_error,unclassified,?,errors encountered. +Model/ModelUtilities/DiscretizationBase.f90,952,read_layer_array,utilities,store_error,not-input,-,Programmer error: read_layer_array must be overridden +Model/ModelUtilities/DiscretizationBase.f90,975,record_array,utilities,store_error,not-input,-,Programmer error: record_array must be overridden +Model/ModelUtilities/DiscretizationBase.f90,1038,record_srcdst_list_header,utilities,store_error,not-input,-,Programmer error: record_srcdst_list_header must be overridden +Model/ModelUtilities/DiscretizationBase.f90,1097,nlarray_to_nodelist,utilities,store_error,not-input,-,Programmer error: nlarray_to_nodelist must be overridden +Model/ModelUtilities/FlowModelInterface.f90,378,source_packagedata,utilities,store_error,not-found-ref,K7,Could not find file +Model/ModelUtilities/FlowModelInterface.f90,383,source_packagedata,utilities,store_error,type-parse,K1,"Unexpected packagedata input keyword read: "" ""." +Model/ModelUtilities/FlowModelInterface.f90,439,source_packagedata_other,utilities,store_error,unclassified,?,"(a,3(1x,a)) UNKNOWN PACKAGEDATA:" +Model/ModelUtilities/FlowModelInterface.f90,486,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,496,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,506,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,515,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,525,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,530,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,540,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,549,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,559,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,564,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,574,read_grid,utilities,store_error,binding-context,K8,"('Error in ',a,': Models do not have the same discretization. GWF model has ', i0, ' user nodes, this model has ', i0, '. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,579,read_grid,utilities,store_error,cellid-grid,K7,"('Error in ',a,': models do not have the same discretization. Models have different IDOMAIN arrays. Ensure discretization packages, including IDOMAIN, are identical.')" +Model/ModelUtilities/FlowModelInterface.f90,649,advance_bfr,utilities,store_error,not-input,-,"(4x,a) REACHED END OF GWF BUDGET FILE BEFORE READING SUFFICIENT BUDGET INFORMATION FOR THIS GWT SIMULATION." +Model/ModelUtilities/FlowModelInterface.f90,667,advance_bfr,utilities,store_error,unclassified,?,"(4x,a) GWF BUDGET READ NOT SUCCESSFUL" +Model/ModelUtilities/FlowModelInterface.f90,674,advance_bfr,utilities,store_error,option-coherence,K5,"(4x, 'TIME DISCRETIZATION IN BUDGET FILE IS INCOMPATIBLE WITH TIME DISCRETIZATION IN COUPLED MODEL. IF THERE IS MORE THAN ONE TIME STEP IN THE BUDGET FILE FOR A GIVEN STRESS PERIOD, BUDGET FILE TIME STEPS MUST MATCH THE COUPLED MODEL TIME STEPS ONE-FOR-ONE IN THAT STRESS PERIOD.')" +Model/ModelUtilities/FlowModelInterface.f90,681,advance_bfr,utilities,store_error,option-coherence,K5,"(4x, 'TIME DISCRETIZATION IN BUDGET FILE IS INCOMPATIBLE WITH TIME DISCRETIZATION IN COUPLED MODEL. IF THERE IS MORE THAN ONE TIME STEP IN THE BUDGET FILE FOR A GIVEN STRESS PERIOD, BUDGET FILE TIME STEPS MUST MATCH THE COUPLED MODEL TIME STEPS ONE-FOR-ONE IN THAT STRESS PERIOD.')" +Model/ModelUtilities/FlowModelInterface.f90,746,advance_bfr,utilities,store_error,option-coherence,K5,"(4x, 'TIME DISCRETIZATION IN BUDGET FILE IS INCOMPATIBLE WITH TIME DISCRETIZATION IN COUPLED MODEL. IF THERE IS MORE THAN ONE TIME STEP IN THE BUDGET FILE FOR A GIVEN STRESS PERIOD, BUDGET FILE TIME STEPS MUST MATCH THE COUPLED MODEL TIME STEPS ONE-FOR-ONE IN THAT STRESS PERIOD.')" +Model/ModelUtilities/FlowModelInterface.f90,808,advance_hfr,utilities,store_error,not-input,-,"(4x,a) REACHED END OF GWF HEAD FILE BEFORE READING SUFFICIENT HEAD INFORMATION FOR THIS GWT SIMULATION." +Model/ModelUtilities/FlowModelInterface.f90,826,advance_hfr,utilities,store_error,unclassified,?,"(4x,a) GWF HEAD READ NOT SUCCESSFUL" +Model/ModelUtilities/FlowModelInterface.f90,837,advance_hfr,utilities,store_error,range-count,K3,"(4x,a) PERIOD NUMBER IN HEAD FILE DOES NOT MATCH PERIOD NUMBER IN TRANSPORT MODEL. IF THERE IS MORE THAN ONE TIME STEP IN THE HEAD FILE FOR A GIVEN STRESS PERIOD, HEAD FILE TIME STEPS MUST MATCH GWT MODEL TIME STEPS ONE-FOR-ONE IN THAT STRESS PERIOD." +Model/ModelUtilities/FlowModelInterface.f90,848,advance_hfr,utilities,store_error,range-count,K3,"(4x,a) TIME STEP NUMBER IN HEAD FILE DOES NOT MATCH TIME STEP NUMBER IN TRANSPORT MODEL. IF THERE IS MORE THAN ONE TIME STEP IN THE HEAD FILE FOR A GIVEN STRESS PERIOD, HEAD FILE TIME STEPS MUST MATCH GWT MODEL TIME STEPS ONE-FOR-ONE IN THAT STRESS PERIOD." +Model/ModelUtilities/FlowModelInterface.f90,953,initialize_gwfterms_from_bfr,utilities,store_error,not-found-ref,K7,"(4x,a) SPECIFIC DISCHARGE NOT FOUND IN BUDGET FILE. SAVE_SPECIFIC_DISCHARGE AND SAVE_FLOWS MUST BE ACTIVATED IN THE NPF PACKAGE." +Model/ModelUtilities/FlowModelInterface.f90,959,initialize_gwfterms_from_bfr,utilities,store_error,not-found-ref,K7,"(4x,a) SATURATION NOT FOUND IN BUDGET FILE. SAVE_SATURATION AND SAVE_FLOWS MUST BE ACTIVATED IN THE NPF PACKAGE." +Model/ModelUtilities/FlowModelInterface.f90,965,initialize_gwfterms_from_bfr,utilities,store_error,not-found-ref,K7,"(4x,a) FLOWJA NOT FOUND IN BUDGET FILE. SAVE_FLOWS MUST BE ACTIVATED IN THE NPF PACKAGE." +Model/ModelUtilities/FlowModelInterface.f90,1099,get_package_index,utilities,store_error,not-found-ref,K7,Error in get_package_index. Could not find +Model/ModelUtilities/GwfMvrPeriodData.f90,104,read_from_parser,utilities,store_error,range-count,K3,"(a,a) Movers exceed MAXMVR on line:" +Model/ModelUtilities/GwfMvrPeriodData.f90,138,read_from_parser,utilities,store_error,type-parse,K1,Invalid mover type: +Model/ModelUtilities/Mover.f90,112,prepare,utilities,store_error,unclassified,?,Provider and receiver are the same: : +Model/ModelUtilities/Mover.f90,129,prepare,utilities,store_error,option-coherence,K5,Mover capability not activated in +Model/ModelUtilities/Mover.f90,130,prepare,utilities,store_error,block-structure,K1,"Add ""MOVER"" keyword to package options block." +Model/ModelUtilities/Mover.f90,142,prepare,utilities,store_error,option-coherence,K5,Mover capability not activated in +Model/ModelUtilities/Mover.f90,143,prepare,utilities,store_error,block-structure,K1,"Add ""MOVER"" keyword to package options block." +Model/ModelUtilities/Mover.f90,154,prepare,utilities,store_error,unclassified,?,Provider ID < 1 or greater than package size +Model/ModelUtilities/Mover.f90,157,prepare,utilities,store_error,unclassified,?,"(a,i0,a,i0) Provider ID = ; Package size =" +Model/ModelUtilities/Mover.f90,176,prepare,utilities,store_error,unclassified,?,Receiver ID < 1 or greater than package size +Model/ModelUtilities/Mover.f90,179,prepare,utilities,store_error,unclassified,?,"(a,i0,a,i0) Receiver ID = ; package size =" +Model/ModelUtilities/SfrCrossSectionManager.f90,173,source_table,utilities,store_error,sign-value,K2,"(a,1x,i0,3a) TABLE NROW MUST BE > 0 FOR REACH (' ')" +Model/ModelUtilities/SfrCrossSectionManager.f90,179,source_table,utilities,store_error,sign-value,K2,"(a,1x,i0,3a) TABLE NCOL MUST BE >= 2 FOR REACH (' ')" +Model/ModelUtilities/SfrCrossSectionManager.f90,305,read_table,utilities,store_error,sign-value,K2,"(a,1x,a) TABLE NROW MUST BE > 0 FOR" +Model/ModelUtilities/SfrCrossSectionManager.f90,313,read_table,utilities,store_error,sign-value,K2,"(a,1x,i0,1x,a,1x,a) TABLE NCOL MUST BE >= FOR" +Model/ModelUtilities/SfrCrossSectionManager.f90,319,read_table,utilities,store_error,unknown-keyword,K1,"(a,a) UNKNOWN DIMENSIONS keyword:" +Model/ModelUtilities/SfrCrossSectionManager.f90,327,read_table,utilities,store_error,block-structure,K1,Required DIMENSIONS block not found. +Model/ModelUtilities/SfrCrossSectionManager.f90,334,read_table,utilities,store_error,block-structure,K1,(a) NROW not specified in the table DIMENSIONS block +Model/ModelUtilities/SfrCrossSectionManager.f90,339,read_table,utilities,store_error,block-structure,K1,(a) NCOL not specified in the table DIMENSIONS block +Model/ModelUtilities/SfrCrossSectionManager.f90,401,read_table,utilities,store_error,block-structure,K1,Required TABLE block not found. +Model/ModelUtilities/SfrCrossSectionManager.f90,408,read_table,utilities,store_error,unclassified,?,"(a,1x,i0,1x,a,1x,i0,1x,a) NROW set to but rows were read" +Model/ModelUtilities/SfrCrossSectionManager.f90,497,validate,utilities,store_error,sign-value,K2,"(3a,1x,i0,1x,a) All xfraction data in ' ' for reach must be greater than or equal to zero." +Model/ModelUtilities/SfrCrossSectionManager.f90,503,validate,utilities,store_error,sign-value,K2,"(3a,1x,i0,1x,a) All height data in ' ' for reach must be greater than or equal to zero." +Model/ModelUtilities/SfrCrossSectionManager.f90,509,validate,utilities,store_error,sign-value,K2,"(3a,1x,i0,1x,a) At least one height data value in ' ' for reach must be equal to zero." +Model/ModelUtilities/SfrCrossSectionManager.f90,515,validate,utilities,store_error,sign-value,K2,"(3a,1x,i0,1x,a) All manfraction data in ' ' for reach must be greater than zero." +Model/ModelUtilities/SfrCrossSectionManager.f90,722,output,utilities,store_warning,duplicate,K6,"(a,1x,i0,7(1x,a)) Cross-section data for reaches include one or more points that result in a non-unique depth-conveyance relation. This occurs when there are horizontal sections at non-zero heights (for example, flat overbank sections). This can usually be resolved by adding a small slope to these f" +Model/ModelUtilities/TimeStepSelect.f90,141,read,utilities,store_error,enum,K2,"(2a) Looking for ALL, STEPS, FIRST, LAST, OR FREQUENCY. Found:" +Model/ModelUtilities/TspSpc.f90,347,check_flow_package,utilities,store_error,unclassified,?,"(a,a,a,i0,a,i0,a) The SPC Package corresponding to flow package has MAXBOUND set less than the number of boundaries active in this package. Found MAXBOUND equal and number of flow boundaries (NBOUND) equal . Increase MAXBOUND in the SPC input file for this package." +Model/ModelUtilities/TspSpc.f90,360,check_flow_package,utilities,store_error,option-coherence,K5,"(a,a,a) Array-based recharge must be used with array-based stress package concentrations. GWF Package is being used with list-based SPC6 input. Use array-based SPC6 input instead." +Model/ModelUtilities/TspSpc.f90,370,check_flow_package,utilities,store_error,option-coherence,K5,"(a,a,a) Array-based evapotranspiration must be used with array-based stress package concentrations. GWF Package is being used with list-based SPC6 input. Use array-based SPC6 input instead." +Model/ModelUtilities/TspSpc.f90,380,check_flow_package,utilities,store_error,option-coherence,K5,"(a,a,a) List-based packages must be used with list-based stress package concentrations. GWF Package is being used with array-based SPC6 input. Use list-based SPC6 input instead." +Model/ModelUtilities/Xt3dInterface.f90,347,xt3d_ar,utilities,store_error,geometry,K10,"Vertices not specified for discretization package, but XT3D is active: . Vertices must be specified in discretization package in order to use XT3D." +Model/ModelUtilities/Xt3dInterface.f90,355,xt3d_ar,utilities,store_error,geometry,K10,"ANGLDEGX is not specified in the DIS package, but XT3D is active: . ANGLDEGX must be provided in discretization package in order to use XT3D." +Model/ModelUtilities/Xt3dInterface.f90,1380,xt3d_areas,utilities,store_error,not-input,-,Program error: xt3d_areas called for a Newton horizontal connection without the saturated flow qsat. +Utilities/ArrayRead/ArrayReaderBase.f90,97,read_control_record,utilities,store_error,not-input,-,"Error reading control record for . Use CONSTANT, INTERNAL, or OPEN/CLOSE." +Utilities/ArrayRead/ArrayReaderBase.f90,122,set_constant,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/ArrayReaderBase.f90,128,fill_constant,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/ArrayReaderBase.f90,154,read_ascii,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/ArrayReaderBase.f90,160,read_binary,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/ArrayReaderBase.f90,166,set_factor,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/ArrayReaderBase.f90,172,apply_factor,utilities,store_error,not-input,-,Programming error in ArrayReader +Utilities/ArrayRead/Double1dReader.f90,83,read_ascii,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Double1dReader.f90,102,read_binary,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Double2dReader.f90,87,read_ascii,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Double2dReader.f90,107,read_binary,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Integer1dReader.f90,115,read_ascii,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Integer1dReader.f90,134,read_binary,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Integer2dReader.f90,87,read_ascii,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayRead/Integer2dReader.f90,107,read_binary,utilities,store_error,not-input,-,Error reading data for array . +Utilities/ArrayReaders.f90,109,read_array_int1d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,132,read_array_int1d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,206,read_array_int2d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,228,read_array_int2d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,346,read_array_dbl1d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,369,read_array_dbl1d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,444,read_array_dbl2d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,466,read_array_dbl2d,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,674,read_control_1,utilities,store_error,unclassified,?,"READING CONTROL RECORD FOR '. Use CONSTANT, INTERNAL, or OPEN/CLOSE." +Utilities/ArrayReaders.f90,709,read_control_2,utilities,store_error,unclassified,?,"""(BINARY)"" option for array input is valid only if OPEN/CLOSE is also specified." +Utilities/ArrayReaders.f90,722,read_control_2,utilities,store_error,unknown-keyword,K1,"Invalid option found in array-control record: "" """ +Utilities/ArrayReaders.f90,975,print_array_int,utilities,store_error,not-input,-,Program error printing array : ii > 1 when prowcolnum is false. +Utilities/ArrayReaders.f90,1021,print_array_dbl,utilities,store_error,not-input,-,Program error printing array : ii > 1 when prowcolnum is false. +Utilities/ArrayReaders.f90,1056,read_binary_header,utilities,store_error,not-input,-,Error reading data for array ' '. +Utilities/ArrayReaders.f90,1084,check_binary_filesize,utilities,store_error,type-parse,K1,"(a,i0,a,i0,a) Unexpected file size for binary input array . Expected= /Found= bytes." +Utilities/ArrayReaders.f90,1117,check_binary_size,utilities,store_error,not-input,-,"(a,i0,a,1x,a,1x,a,i0,a,1x,i0,3(1x,a)) The size of the data array calculated from the binary header ( ) will exceed the remainder of the data array ( ) array by elements. This is usually caused by incorrect assignment of (m1,m2,m3) in the binary header. See the mf6io.pdf document for information on a" +Utilities/BlockParser.f90,225,GetNextLine,utilities,store_error,not-input,-,Unexpected end of file reached. +Utilities/BlockParser.f90,365,ReadScalarError,utilities,store_error,unclassified,?,(3a) '. +Utilities/BlockParser.f90,457,terminateblock,utilities,store_error,enum,K2,LOOKING FOR 'END '. FOUND: ' '. +Utilities/BlockParser.f90,631,uget_block,utilities,store_error,not-input,-,"Required block "" "" not found. Found end of file instead." +Utilities/BlockParser.f90,673,uget_block,utilities,store_error,block-structure,K1,"Error: Required block "" "" not found. Found block "" "" instead." +Utilities/BlockParser.f90,686,uget_block,utilities,store_error,enum,K2,Error: Looking for BEGIN but found END instead. +Utilities/BlockParser.f90,747,uget_any_block,utilities,store_error,block-structure,K1,Block name missing in file. +Utilities/BlockParser.f90,792,uterminate_block,utilities,store_error,unclassified,?, +Utilities/BlockParser.f90,804,uterminate_block,utilities,store_error,unclassified,?, +Utilities/Budget.f90,409,add_single_entry,utilities,store_error,unclassified,?,"('Error in MODFLOW 6.', 'Entries do not match: ', (a), (a) )" +Utilities/Budget.f90,475,add_multi_entry,utilities,store_error,unclassified,?,"('Error in MODFLOW 6.', 'Entries do not match: ', (a), (a) )" +Utilities/Budget.f90,492,add_multi_entry,utilities,store_error,unclassified,?,Could not add multi-entry +Utilities/BudgetFileReader.f90,175,read_header,utilities,store_error,not-input,-,"(a, a) ERROR READING:" +Utilities/BudgetFileReader.f90,177,read_header,utilities,store_error,type-parse,K1,"(a, i0) INVALID METHOD CODE DETECTED:" +Utilities/Export/DisNCStructured.f90,155,dis_export_init,utilities,store_warning,unclassified,?,"(a) Ignoring user provided NetCDF chunking parameters. Define chunk_time, chunk_x, chunk_y and chunk_z input parameters to see an effect in file "" ""." +Utilities/Export/DisNCStructured.f90,167,dis_export_init,utilities,store_warning,unclassified,?,"(a) Ignoring user provided NetCDF wkt/crs_wkt parameter(s) as longitude and latitude arrays have been provided. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,181,dis_export_init,utilities,store_warning,unclassified,?,"(a) Structured rotated grid with user-provided latitude/longitude: MF6 cannot verify these values correctly account for grid rotation (angrot). They must be computed from each cell''s true rotated position, not from grid-local x/y. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,196,dis_export_init,utilities,store_warning,unclassified,?,"(a) CRS parameter set with structured rotated grid. The x/y coordinate variables have grid-local, not real-world, values. A CRS-aware consumer must apply the full crs_wkt (which encodes rotation when a valid WKT2 PROJCRS was supplied) via proper reprojection to resolve the true position; deriving re" +Utilities/Export/DisNCStructured.f90,998,define_gridmap,utilities,store_warning,unclassified,?,"(a) Structured rotated grid with no CRS_WKT input: no rotation-positioning information will be written to the output CRS; x/y coordinate values will not reflect the true rotated position. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1010,define_gridmap,utilities,store_warning,unclassified,?,"(a) Structured rotated grid: wrapped CRS_WKT in a derived CRS to encode grid rotation ( ). Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1018,define_gridmap,utilities,store_warning,unclassified,?,"(a) Structured rotated grid: CRS_WKT is not a recognized WKT2 PROJCRS; no rotation-positioning information will be written to the output CRS; x/y coordinate values will not reflect the true rotated position. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1075,define_gridmap,utilities,store_warning,unclassified,?,"(a) Structured grid: unable to extract CF transverse_mercator grid_mapping parameters from CRS_WKT; some netCDF readers may not correctly position this grid. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1113,define_gridmap,utilities,store_warning,option-coherence,K5,"(a) Structured grid: 1SP lambert_conformal_conic variants are not supported for CF grid_mapping parameter extraction (CF has no scale_factor attribute for this method); some netCDF readers may not correctly position this grid. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1119,define_gridmap,utilities,store_warning,unclassified,?,"(a) Structured grid: unable to extract CF lambert_conformal_conic grid_mapping parameters from CRS_WKT; some netCDF readers may not correctly position this grid. Applies to file "" ""." +Utilities/Export/DisNCStructured.f90,1156,define_gridmap,utilities,store_warning,geometry,K10,"(a) Structured grid: unable to extract CF albers_conical_equal_area grid_mapping parameters from CRS_WKT; some netCDF readers may not correctly position this grid. Applies to file "" ""." +Utilities/Export/MeshNCModel.f90,133,mesh_init,utilities,store_warning,option-coherence,K5,"(a) LATITUDE and LONGITUDE griddata are not supported for a NETCDF_MESH2D export and will be ignored. Applies to file "" ""." +Utilities/Export/MeshNCModel.f90,154,mesh_init,utilities,store_warning,unclassified,?,"(a) Ignoring user provided NetCDF chunking parameter. Define chunk_time and chunk_face input parameters to see an effect in file "" ""." +Utilities/Export/ModelExport.f90,130,modelexports_create,utilities,store_error,unclassified,?,"NetCDF output file name must use "".nc"" extension. Filename="" ""." +Utilities/Export/NCExportCreate.f90,130,create_nc_export,utilities,store_error,option-coherence,K5,DISV model discretization only supported as UGRID NetCDF export. Model= . +Utilities/Export/NCExportCreate.f90,136,create_nc_export,utilities,store_error,enum,K2,Unsupported discretization for NetCDF model export. Model= . +Utilities/Export/NCModel.f90,245,set,utilities,store_error,option-coherence,K5,models not supported for NetCDF export. +Utilities/Export/NCModel.f90,342,export_init,utilities,store_error,option-coherence,K5,models not supported for NetCDF export. +Utilities/FeatureFlags.f90,33,developmode,utilities,store_error,unclassified,?, +Utilities/FeatureFlags.f90,36,developmode,utilities,store_error,unclassified,?, +Utilities/GridFileReader.f90,109,read_header_meta,utilities,store_error,block-structure,K1,"Binary grid file must begin with ""GRID"". Found:" +Utilities/GridFileReader.f90,241,lookup,utilities,store_error,not-found-ref,K7,(a) Variable not found +Utilities/GridFileReader.f90,245,lookup,utilities,store_error,unclassified,?,(a) Variable is not +Utilities/GridFileReader.f90,312,read_int_1d_into,utilities,store_error,range-count,K3,"(a,i0,a,i0) Array size mismatch for : expected , got" +Utilities/GridFileReader.f90,355,read_dbl_1d_into,utilities,store_error,range-count,K3,"(a,i0,a,i0) Array size mismatch for : expected , got" +Utilities/InputOutput.f90,129,openfile,utilities,store_error,geometry,K10,"(a,1x,a) STOP EXECUTION in subroutine openfile()." +Utilities/InputOutput.f90,191,assign_iounit,utilities,store_error,unclassified,?,"(a,1x,i0) already assigned at unit:" +Utilities/InputOutput.f90,577,URWORD,utilities,store_error,unclassified,?,"('Keyboard input: Error converting """"',a, '"""" to ',A,' in following line:')" +Utilities/InputOutput.f90,578,URWORD,utilities,store_error,unclassified,?, +Utilities/InputOutput.f90,1257,read_line,utilities,store_error,not-input,-,"('Error reading from file """"',a,'"""" opened on unit ',i0, ' in read_line.')" +Utilities/InputOutput.f90,1361,urdaux,utilities,store_error,block-structure,K1,(a) Auxiliary variables already specified. Auxiliary variables must be specified on one line in the options block. +Utilities/InputOutput.f90,1374,urdaux,utilities,store_error,unclassified,?,"(a, a, a, i0, a, i0, a) Found auxiliary variable ( ) with a name of size . Auxiliary variable names must be len than or equal to characters." +Utilities/InputOutput.f90,1432,print_format,utilities,store_error,unclassified,?,Could not build PRINT_FORMAT from line +Utilities/InputOutput.f90,1435,print_format,utilities,store_error,unclassified,?,Syntax is: COLUMNS WIDTH DIGITS +Utilities/InputOutput.f90,1454,print_format,utilities,store_error,unclassified,?,Error setting PRINT_FORMAT. Syntax is incorrect in line: COLUMNS WIDTH +Utilities/InputOutput.f90,1455,print_format,utilities,store_error,unclassified,?, +Utilities/InputOutput.f90,1458,print_format,utilities,store_error,unclassified,?,Syntax is: COLUMNS WIDTH DIGITS +Utilities/InputOutput.f90,1496,print_format,utilities,store_error,unknown-keyword,K1,Error in format specification. Unrecognized option: +Utilities/InputOutput.f90,1498,print_format,utilities,store_error,unclassified,?,"Valid values are EXPONENTIAL, FIXED, GENERAL, or SCIENTIFIC." +Utilities/InputOutput.f90,1506,print_format,utilities,store_error,unclassified,?, +Utilities/InputOutput.f90,1507,print_format,utilities,store_error,unclassified,?, +Utilities/InputOutput.f90,1787,u9rdcom,utilities,store_error,type-parse,K1,u9rdcom: Could not read from unit: +Utilities/InputOutput.f90,1920,get_line,utilities,store_error,unclassified,?,"(a,1x,5a) Line: "" | """ +Utilities/Iunit.f90,85,addfile,utilities,store_error,not-found-ref,K7,"(a,a) Package type not supported:" +Utilities/ListReader.f90,190,set_openclose,utilities,store_error,not-input,-,"('Specified OPEN/CLOSE file ',(A),' does not exist')" +Utilities/ListReader.f90,191,set_openclose,utilities,store_error,not-input,-,Specified OPEN/CLOSE file does not exist +Utilities/ListReader.f90,299,read_binary,utilities,store_error,not-input,-,"('Error reading list. The number of records encountered exceeds the maximum number of records. Number of records found is ',I0,' but MAXBOUND is ', I0, '. Try increasing MAXBOUND for this list. Error occurred reading the following line: ', a, 5x, '>>> ', a)" +Utilities/ListReader.f90,320,read_binary,utilities,store_error,not-input,-,"('ERROR READING LIST FROM FILE: ',1x,a,1x,' ON UNIT: ',I0)" +Utilities/ListReader.f90,334,read_binary,utilities,store_error,not-input,-,"('ERROR READING LIST FROM FILE: ',1x,a,1x,' ON UNIT: ',I0)" +Utilities/ListReader.f90,416,read_ascii,utilities,store_error,not-input,-,"('Error reading list. The number of records encountered exceeds the maximum number of records. Number of records found is ',I0,' but MAXBOUND is ', I0, '. Try increasing MAXBOUND for this list. Error occurred reading the following line: ', a, 5x, '>>> ', a) A" +Utilities/ListReader.f90,506,read_ascii,utilities,store_error,block-structure,K1,"(a,i0,a) Auxiliary data or time series name expected but not found in period block "" ""." +Utilities/ListReader.f90,562,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Utilities/ListReader.f90,570,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Utilities/ListReader.f90,579,check_cellid,utilities,store_error,cellid-grid,K7,"('List entry ',i0,' contains cellid ',a,' but this cellid is invalid for model with shape ', a)" +Utilities/LongLineReader.f90,106,bkspc,utilities,store_error,not-input,-,Programming error in LongLineReaderType%bkspc(). Backspace called more than once for an open file. +Utilities/Memory/MemoryHelper.f90,125,split_mem_path,utilities,store_error,not-input,-,"(*(G0)) Fatal error in Memory Manager, cannot split invalid memory path:" +Utilities/Memory/MemoryHelper.f90,208,mem_check_length,utilities,store_error,not-input,-,"(*(G0)) Fatal error in Memory Manager, length of must be characters or less: (len= )" +Utilities/Memory/MemoryManager.f90,327,get_from_memorystore,utilities,store_error,not-input,-,Programming error in memory manager. Variable ' ' in ' ' cannot be assigned because it does not exist in memory manager. +Utilities/Memory/MemoryManager.f90,355,allocate_error,utilities,store_error,not-input,-,Error trying to allocate memory. Path ' ' variable name ' ' size ' '. Error message is ' '. Status code is . +Utilities/Memory/MemoryManager.f90,455,allocate_str,utilities,store_error,not-input,-,Programming error in allocate_str. ILEN must be greater than 0. +Utilities/Memory/MemoryManager.f90,512,allocate_str1d,utilities,store_error,not-input,-,Programming error in allocate_str1d. ILEN must be greater than 0. +Utilities/Memory/MemoryManager.f90,1267,reallocate_str1d,utilities,store_error,not-input,-,"Programming error, variable ' ' from ' ' is not defined in the memory manager. Use mem_allocate instead." +Utilities/Memory/MemoryManager.f90,1361,reallocate_charstr1d,utilities,store_error,not-input,-,"Programming error, variable ' ' from ' ' is not defined in the memory manager. Use mem_allocate instead." +Utilities/Memory/MemoryManager.f90,2717,mem_da,utilities,store_error,unclassified,?,Could not clear memory list. +Utilities/Memory/MemoryManager.f90,2734,mem_da_check,utilities,store_error,unclassified,?, +Utilities/Memory/MemoryManager.f90,2741,mem_da_check,utilities,store_error,unclassified,?, +Utilities/Memory/MemoryManager.f90,2750,mem_da_check,utilities,store_error,unclassified,?, +Utilities/Memory/MemoryManagerExt.f90,206,mem_set_value_logical1d,utilities,store_error,not-input,-,"mem_set_value() size mismatch logical1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,244,mem_set_value_logical1d_mapped,utilities,store_error,not-input,-,"mem_set_value() size mismatch logical1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,276,mem_set_value_int1d,utilities,store_error,not-input,-,"mem_set_value() size mismatch int1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,314,mem_set_value_int1d_mapped,utilities,store_error,not-input,-,"mem_set_value() size mismatch int1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,347,mem_set_value_int2d,utilities,store_error,not-input,-,"mem_set_value() size mismatch int2d, varname=" +Utilities/Memory/MemoryManagerExt.f90,382,mem_set_value_int3d,utilities,store_error,not-input,-,"mem_set_value() size mismatch int3d, varname=" +Utilities/Memory/MemoryManagerExt.f90,441,mem_set_value_dbl1d,utilities,store_error,not-input,-,"mem_set_value() size mismatch dbl1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,479,mem_set_value_dbl1d_mapped,utilities,store_error,not-input,-,"mem_set_value() size mismatch dbl1d, varname=" +Utilities/Memory/MemoryManagerExt.f90,512,mem_set_value_dbl2d,utilities,store_error,not-input,-,"mem_set_value() size mismatch dbl2d, varname=" +Utilities/Memory/MemoryManagerExt.f90,547,mem_set_value_dbl3d,utilities,store_error,not-input,-,"mem_set_value() size mismatch dbl3d, varname=" +Utilities/Observation/Obs.f90,277,DefaultObsIdProcessor,utilities,store_error,not-input,-,Error reading data from ID string +Utilities/Observation/Obs.f90,476,StoreObsType,utilities,store_error,not-input,-,Programmer error: Invalid argument in store_obs_type. +Utilities/Observation/Obs.f90,491,StoreObsType,utilities,store_error,not-input,-,Size of obsData array is insufficient; need to increase MAXOBSTYPES. +Utilities/Observation/Obs.f90,636,read_obs_options,utilities,store_error,unclassified,?,"End-of-file encountered while searching for OPTIONS in OBS input file "" """ +Utilities/Observation/Obs.f90,656,read_obs_options,utilities,store_error,duplicate,K6,Error in OBS input: DIGITS has already been defined +Utilities/Observation/Obs.f90,671,read_obs_options,utilities,store_error,type-parse,K1,Error in OBS input: Invalid value for DIGITS option +Utilities/Observation/Obs.f90,684,read_obs_options,utilities,store_error,unknown-keyword,K1,Error in OBS input: Unrecognized option: +Utilities/Observation/Obs.f90,874,get_obs_datum,utilities,store_error,not-found-ref,K7,Observation type not found: +Utilities/Observation/Obs.f90,984,read_obs_blocks,utilities,store_error,unclassified,?,"CONTINUOUS keyword must be followed by ""FILEOUT"" then by filename." +Utilities/Observation/Obs.f90,995,read_obs_blocks,utilities,store_error,unclassified,?, +Utilities/Observation/Obs.f90,1000,read_obs_blocks,utilities,store_error,duplicate,K6,"OBS outfile "" "" is provided more than once." +Utilities/Observation/Obs.f90,1055,read_obs_blocks,utilities,store_error,unknown-keyword,K1,Error: Observation block type not recognized: +Utilities/OutputControl/OutputControl.f90,129,oc_rp,utilities,store_error,block-structure,K1,"Input OC period block rtype not found: "" ""." +Utilities/OutputControl/OutputControl.f90,308,set_ocfile,utilities,store_error,not-input,-,"OC internal error: oc data type not found for name "" ""." +Utilities/OutputControl/OutputControlData.f90,73,ocd_rp_check,utilities,store_error,option-coherence,K5,"(1X,'REQUESTING TO SAVE ',A,' BUT ',A,' SAVE FILE NOT SPECIFIED. ', A,' SAVE FILE MUST BE SPECIFIED IN OUTPUT CONTROL OPTIONS.')" +Utilities/OutputControl/PrintSaveManager.f90,108,read,utilities,store_error,enum,K2,Looking for PRINT or SAVE. Found: +Utilities/OutputControl/PrintSaveManager.f90,131,set,utilities,store_error,enum,K2,Looking for PRINT or SAVE. Found: +Utilities/PtrHashTable.f90,57,add,utilities,store_warning,unclassified,?,Already existing variable being added to the HashTable - +Utilities/Sim.f90,461,initial_message,utilities,store_warning,unclassified,?,Running parallel MODFLOW on only 1 process +Utilities/Table.f90,212,initialize_column,utilities,store_error,unclassified,?,"(a,a,a,i0,a,1x,a,1x,a,a,a,1x,i0,1x,a) Trying to add column "" "" ( ) in the table ("" "") that only has columns." +Utilities/Table.f90,534,add_error,utilities,store_error,unclassified,?,"(a,1x,i0,5(1x,a),1x,i0,1x,a) Trying to add data to column in the table ( ) that only has columns." +Utilities/TimeSeries/TimeArray.f90,67,ConstructTimeArray,utilities,store_error,option-coherence,K5,Time array series is not supported for discretization type +Utilities/TimeSeries/TimeArraySeries.f90,77,ConstructTimeArraySeries,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,128,tas_init,utilities,store_error,block-structure,K1,Error: Attributes block not found in file: +Utilities/TimeSeries/TimeArraySeries.f90,155,tas_init,utilities,store_error,unclassified,?,"Unknown interpolation method: "" """ +Utilities/TimeSeries/TimeArraySeries.f90,165,tas_init,utilities,store_error,not-input,-,"Error reading numeric SFAC value from "" """ +Utilities/TimeSeries/TimeArraySeries.f90,171,tas_init,utilities,store_error,unknown-keyword,K1,"Unknown option found in ATTRIBUTES block: "" """ +Utilities/TimeSeries/TimeArraySeries.f90,180,tas_init,utilities,store_error,temporal,K9,Name not specified for time array series in file: +Utilities/TimeSeries/TimeArraySeries.f90,186,tas_init,utilities,store_error,temporal,K9,Interpolation method not specified for time array series in file: +Utilities/TimeSeries/TimeArraySeries.f90,194,tas_init,utilities,store_error,temporal,K9,Error(s) encountered initializing time array series from file: +Utilities/TimeSeries/TimeArraySeries.f90,202,tas_init,utilities,store_error,temporal,K9,Error encountered reading time-array data from file: +Utilities/TimeSeries/TimeArraySeries.f90,378,get_surrounding_records,utilities,store_error,option-coherence,K5,Time array series is not supported for selected discretization type. +Utilities/TimeSeries/TimeArraySeries.f90,491,get_values_at_time,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,537,get_integrated_values,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,586,get_integrated_values,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,587,get_integrated_values,utilities,store_error,not-input,-,(Probable programming error) +Utilities/TimeSeries/TimeArraySeries.f90,602,get_integrated_values,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,610,get_integrated_values,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeArraySeries.f90,611,get_integrated_values,utilities,store_error,not-input,-,(Probable programming error) +Utilities/TimeSeries/TimeArraySeries.f90,683,get_latest_preceding_node,utilities,store_error,not-input,-,probable programming error in get_latest_preceding_node +Utilities/TimeSeries/TimeArraySeriesManager.f90,173,tasmgr_ad,utilities,store_error,range-count,K3,Size mismatch between boundary and multiplier arrays using time-array series: +Utilities/TimeSeries/TimeArraySeriesManager.f90,304,MakeTasLink,utilities,store_error,not-found-ref,K7,"Error: Time-array series "" "" not found." +Utilities/TimeSeries/TimeArraySeriesManager.f90,369,tasmgr_convert_flux,utilities,store_error,not-input,-,Programming error. Cannot convert flux. Verify that ''a valid DIS instance and nodelist were provided. +Utilities/TimeSeries/TimeSeries.f90,264,initialize_time_series,utilities,store_error,temporal,K9,Name not specified for time series. +Utilities/TimeSeries/TimeSeries.f90,552,get_value_at_time,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeSeries.f90,601,get_integrated_value,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeSeries.f90,673,get_integrated_value,utilities,store_error,unclassified,?, +Utilities/TimeSeries/TimeSeries.f90,746,get_latest_preceding_node,utilities,store_error,not-input,-,probable programming error in get_latest_preceding_node +Utilities/TimeSeries/TimeSeries.f90,1107,Initializetsfile,utilities,store_error,temporal,K9,"End-of-file encountered while searching for ATTRIBUTES in time-series input file "" """ +Utilities/TimeSeries/TimeSeries.f90,1112,Initializetsfile,utilities,store_error,block-structure,K1,"ATTRIBUTES block not found in time-series tsfile input file "" """ +Utilities/TimeSeries/TimeSeries.f90,1152,Initializetsfile,utilities,store_error,unclassified,?,Error: NAME attribute not provided before METHOD in file: +Utilities/TimeSeries/TimeSeries.f90,1164,Initializetsfile,utilities,store_error,unclassified,?,"Unknown interpolation method: "" """ +Utilities/TimeSeries/TimeSeries.f90,1174,Initializetsfile,utilities,store_error,unclassified,?,Error: NAME attribute not provided before METHODS in file: +Utilities/TimeSeries/TimeSeries.f90,1182,Initializetsfile,utilities,store_error,temporal,K9,METHODS attribute does not list a method for all time series. +Utilities/TimeSeries/TimeSeries.f90,1196,Initializetsfile,utilities,store_error,unclassified,?,"Unknown interpolation method: "" """ +Utilities/TimeSeries/TimeSeries.f90,1204,Initializetsfile,utilities,store_error,unclassified,?,NAME attribute not provided before SFAC in file: +Utilities/TimeSeries/TimeSeries.f90,1210,Initializetsfile,utilities,store_error,not-input,-,"Error reading numeric value from: "" """ +Utilities/TimeSeries/TimeSeries.f90,1219,Initializetsfile,utilities,store_error,unclassified,?,NAME attribute not provided before SFACS in file: +Utilities/TimeSeries/TimeSeries.f90,1233,Initializetsfile,utilities,store_error,unknown-keyword,K1,"Unknown option found in ATTRIBUTES block: "" """ +Utilities/TimeSeries/TimeSeries.f90,1246,Initializetsfile,utilities,store_error,temporal,K9,Error: No time-series data contained in file: +Utilities/TimeSeries/TimeSeries.f90,1253,Initializetsfile,utilities,store_error,block-structure,K1,Interpolation method was not set. METHOD or METHODS must be specified in the ATTRIBUTES block for this time series file. +Utilities/TimeSeries/TimeSeriesManager.f90,113,add_tsfile,utilities,store_error,duplicate,K6,Found duplicate time-series file name: +Utilities/TimeSeries/TimeSeriesManager.f90,402,make_link,utilities,store_error,not-input,-,programmer error in make_link +Utilities/TimeSeries/TimeSeriesManager.f90,564,read_value_or_time_series,utilities,store_error,temporal,K9,"Error in list input. Expected numeric value or time-series name, but found ' '." +Utilities/TimeSeries/TimeSeriesManager.f90,652,read_value_or_time_series_adv,utilities,store_error,temporal,K9,"Error in list input. Expected numeric value or time-series name, but found ' '." +Utilities/comarg.f90,191,GetCommandLineArguments,utilities,store_error,unclassified,?,"(2a,1x,a) : illegal STDOUT level option -" +Utilities/comarg.f90,212,GetCommandLineArguments,utilities,store_error,unclassified,?,: illegal MODFLOW 6 simulation mode option - +Utilities/comarg.f90,226,GetCommandLineArguments,utilities,store_error,unclassified,?,"(2a,1x,a) : illegal option -" +Utilities/comarg.f90,238,GetCommandLineArguments,utilities,store_error,unclassified,?,"(2a,2(1x,a)) : is not present in working directory." +Utilities/sort.f90,124,qsort_int1d,utilities,store_error,unclassified,?,"(a,3(1x,a)) JSTACK > NSTACK IN SortModule::qsort" +Utilities/sort.f90,252,qsort_dbl1d,utilities,store_error,unclassified,?,"(a,3(1x,a)) JSTACK > NSTACK IN SortModule::qsort" diff --git a/constraint-audit/mf6-sample.csv b/constraint-audit/mf6-sample.csv new file mode 100644 index 00000000..f6fb66d1 --- /dev/null +++ b/constraint-audit/mf6-sample.csv @@ -0,0 +1,167 @@ +site,msg,kind,level,arity,cd,trigger,dfn,ddl,sev,when +LoadMf6File.f90:419,record with keyword not properly formed,K1,layout,row,data,uncond,yes,none:syntax,e,load +StructArray.f90:1237,unrecognized keystring keyword in PERIOD,K1,layout,row,data,uncond,yes,none:syntax,e,load +SourceCommon.f90:470,multiple FILEIN for one tag,K1,layout,field,config,uncond,no,none:syntax,e,load +SourceCommon.f90:499,input name exceeds max length,K2,logical,field,data,uncond,no,CHECK(length),e,load +StructArray.f90:1015,CELLID must be integer or NONE,K1,layout,field,data,uncond,partly,none:syntax,e,load +IdmMf6File.f90:249,period numbers not increasing,K9,layout,component,data,uncond,partly,none:block order,e,load +StructArray.f90:1114,line count exceeds input dimension,K3,logical,table,cfg-data,dim,yes,CHECK(count(*)<=dim),e,load +StructArray.f90:1198,keystring rows exceed MAXBOUND,K3,logical,table,cfg-data,dim,yes,CHECK(count(*)<=dim),e,load +Mf6FileGridArray.f90:261,defined cells exceed MAXBOUND,K3,logical,table,cfg-data,dim,partly,CHECK(count(*)<=dim),e,load +LoadContext.f90:719,dimension does not match PACKAGEDATA row count,K3,logical,table,cfg-data,dim,partly,CHECK(count(*)=dim),e,load +Mf6FileKeystring.f90:381,feature number on row out of 1..n,K7,logical,row,data,dim,yes,FK,e,load +SourceCommon.f90:291,required dimension NLAY not found,K4,logical,field,config,uncond,yes,NOT NULL,e,load +ModelPackageInputs.f90:329,DIS package not specified for model,K8,logical,multi,config,uncond,no,assertion:exists,e,load +ModelPackageInputs.f90:434,multiple instances of base package type,K8,logical,multi,config,uncond,yes,"UNIQUE(model,ftype)",e,load +ModelPackageInputs.f90:246,models block model type not valid,K2,logical,field,config,uncond,partly,CHECK(in),e,load +IdmLoad.f90:362,exchange has unrecognized model names,K7,logical,multi,config,uncond,yes,FK,e,load +NCArrayReader.f90:440,layered netcdf read not supported for DIS int2d,L,layout,field,data,opt:netcdf(layout),no,none:encoding domain,e,load +StructArray.f90:1334,binary list input not supported for text columns,L,layout,table,data,opt:binary(layout),no,none:encoding domain,e,load +SourceLoad.F90:398,NetCDF keyword but no NetCDF libraries,-,runtime,sim,config,opt:netcdf(layout),no,none:build,e,load +SimulationCreate.f90:867,model names must be unique,K6,logical,table,config,uncond,yes,PK,e,load +SimulationCreate.f90:859,model name cannot have spaces,K2,logical,field,config,uncond,no,CHECK(pattern),e,load +SimulationCreate.f90:784,model not assigned to a solution,K8,logical,multi,config,uncond,no,assertion:total,e,load +SimulationCreate.f90:688,explicit model cannot be in IMS,K8,logical,multi,config,binding,no,assertion:type match,e,load +SimulationCreate.f90:539,MXITER>1 with one solution in group,K5,logical,multi,config,binding,no,assertion,e,load +SimulationCreate.f90:636,solution group blocks not consecutive,K1,layout,component,config,uncond,no,none:block order,e,load +SimulationCreate.f90:191,TDIS6 unset,K4,logical,field,config,uncond,yes,NOT NULL,e,load +tdis.f90:632,NSTP less than one,K2,logical,field,data,uncond,no,CHECK,e,load +tdis.f90:652,PERLEN cannot be negative,K2,logical,field,data,uncond,no,CHECK,e,load +tdis.f90:671,time step length too small,K10,runtime,row,data,uncond,no,none:numerics,e,ar +ats.f90:428,IPERATS must be > 0,K9,logical,field,data,uncond,no,FK(period),e,load +ats.f90:434,IPERATS greater than NPER,K9,logical,field,data,dim,no,FK(period),w,load +ats.f90:466,DTMIN must be < DTMAX,K2,logical,row,data,uncond,no,CHECK(row),e,load +ImsLinear.f90:199,RCLOSE must be > 0,K2,logical,field,config,uncond,no,CHECK,e,load +ImsLinear.f90:185,ISCL must be <= 2,K2,logical,field,config,uncond,no,CHECK(in),e,load +Dis.f90:264,NLAY not specified or <=0,K2,logical,field,config,uncond,partly,CHECK(>0),e,load +Dis.f90:418,no active nodes (IDOMAIN),K10,logical,table,data,uncond,no,assertion:exists,e,load +Dis.f90:800,layer out of 1..nlay,K7,logical,field,data,dim,partly,FK(grid),e,load +Dis.f90:1620,MAXBOUND too small (array->list),K3,logical,table,cfg-data,dim,partly,CHECK(count(*)<=dim),e,rp +Disu.f90:413,first connection must be self,K1,layout,row,data,uncond,no,none:CSR encoding,e,ar +Disu.f90:422,invalid connection in JA,K7,logical,row,data,dim,no,FK(node),e,ar +Disu.f90:438,invalid IDOMAIN value,K2,logical,field,data,uncond,no,CHECK(in),e,ar +Disu.f90:479,top above bottom of overlying cell (offset tol),K10,logical,multi,data,opt:vertical_offset_tolerance(output),no,assertion:join,e,ar +Disu.f90:1135,ICELL2D input sequence violation,K1,layout,table,data,uncond,no,none:row order,e,load +Disu.f90:1460,unit vectors need VERTEX data,K4,logical,component,data,presence:vertices,no,assertion:exists,e,ar +doc:gwf/disv.tex,cell vertices in clockwise order,K10,logical,row,data,uncond,no,none:geometry predicate,e,load +Disu(npf.f90:391),ANGLDEGX inconsistent across faces,K10,logical,multi,data,uncond,no,assertion:join,e,ar +gwf-npf.f90:280,XT3D cannot be used with GNC,K8,logical,multi,config,presence:gnc,no,assertion:exclusion,e,ar +gwf-npf.f90:1715,NEWTON cannot be used with PERCHED,K5,logical,multi,config,opt:newton(logical),no,CHECK(cross-table),e,load +gwf-npf.f90:1742,ALTERNATIVE_CELL_AVERAGING not with XT3D,K5,logical,component,config,opt:xt3d(logical),no,CHECK(row),e,load +gwf-npf.f90:1733,HIGHEST_CELL_SATURATION ignored without NEWTON,K5,logical,multi,config,opt:newton(logical),no,CHECK(cross-table),w,load +gwf-npf.f90:1862,K not found in GRIDDATA,K4,logical,field,data,uncond,yes,NOT NULL,e,load +gwf-npf.f90:1874,K22OVERK specified but K22 not,K4,layout,component,cfg-data,opt:k22overk(layout),no,none:layout flag needs data,e,load +gwf-npf.f90:1942,K <= 0 for cell,K2,logical,field,data,uncond,no,CHECK(>0),e,ar +gwf-npf.f90:1981,K22 given but no ANGLDEGX in DIS,K4,logical,multi,data,presence:k22,no,assertion:cross-component,e,ar +gwf-npf.f90:2032,ANGLE2 requires ANGLE1,K4,logical,component,data,presence:angle1,no,CHECK(row: a2 -> a1),e,ar +gwf-npf.f90:2195,negative cell thickness (top-bot),K10,logical,multi,data,uncond,no,assertion:join,e,ar +gwf-npf.f90:2466,CHD cell went dry,-,runtime,cell,data,uncond,no,none:state,e,ad +gwf-sto.f90:1160,SS < 0,K2,logical,field,data,uncond,no,CHECK(>=0),e,ar +gwf-sto.f90:281,zero DELT with STO,K10,logical,multi,data,presence:sto,no,assertion:cross-component,e,ad +gwf-ic.f90:159,STRT not found,K4,logical,field,data,uncond,yes,NOT NULL,e,load +gwf-tvk.f90:126,CELLID not in active domain,K7,logical,row,data,binding,partly,FK(active cell),e,rp +gwf-tvk.f90:236,changed K <= 0,K2,logical,field,data,uncond,no,CHECK(>0),e,ad +BoundaryPackage.f90:1406,MAXBOUND must be > 0,K2,logical,field,config,uncond,no,CHECK(>0),e,load +BoundaryPackage.f90:1286,multiple OBS6 entries,K8,logical,component,config,uncond,no,UNIQUE,e,load +BoundaryPackage.f90:1252,TAS6 not for this discretization,L,layout,multi,config,binding,no,none:encoding domain,e,load +BoundaryPackage.f90:1331,AUXMULTNAME given but no AUX,K7,logical,component,config,opt:auxmultname(logical/schema),partly,FK(options.auxiliary),e,load +BoundaryPackage.f90:338,expecting BEGIN PERIOD,K1,layout,component,data,uncond,yes,none:syntax,e,rp +gwf-chd.f90:154,cell already a constant head,K6,logical,multi,data,uncond,no,UNIQUE(cell) across packages,e,rp +gwf-wel.f90:231,FLOW_REDUCTION_LENGTH without AUTO_FLOW_REDUCE ignored,K5,layout,component,config,opt:flow_reduction_length(layout),no,CHECK(row),w,load +gwf-wel.f90:280,AUTO_FLOW_REDUCE_AUXNAME not in AUX,K7,logical,component,config,opt:auxiliary(logical),no,FK(options.auxiliary),e,load +gwf-evt.f90:202,READASARRAYS not with SURF_RATE_SPECIFIED,L,layout,component,config,opt:readasarrays(layout),no,none:encoding domain,e,load +gwf-evt.f90:291,NSEG>1 not with READASARRAYS,L,layout,component,config,opt:readasarrays(layout),no,none:encoding domain,e,load +gwf-evt.f90:383,"PXDP must be in (0,1)",K2,logical,field,data,uncond,no,CHECK,e,rp +gwf-evt.f90:395,PXDP not monotonically increasing,K2,logical,row,data,dim,no,"CHECK(row, ordered columns)",e,rp +doc:gwf/evt.tex,list input required for DISU,L,layout,multi,config,binding,no,none:encoding domain,e,load +gwf-sfr.f90:909,"STORAGE_WEIGHT in [0.5,1]",K2,logical,field,config,uncond,no,CHECK,e,load +gwf-sfr.f90:1730,ATS_COURANT requires STORAGE,K5,logical,component,config,opt:ats_courant(logical),no,CHECK(row),e,load +gwf-sfr.f90:1742,ATS_COURANT but no ATS in TDIS,K8,logical,multi,config,presence:ats,no,assertion:exists,w,load +gwf-sfr.f90:4170,UNIT_CONVERSION with LENGTH/TIME_CONVERSION,K5,layout,component,config,opt:unit_conversion(layout),no,CHECK(row),e,load +gwf-sfr.f90:1346,connections given but NCON differs,K3,logical,table,data,dim,no,CHECK(count per key = column),e,load +gwf-sfr.f90:1365,reach number exceeds NREACHES,K7,logical,row,data,dim,yes,FK,e,load +gwf-sfr.f90:1513,DIVERSIONS block without diversions,K4,logical,component,data,dim,no,assertion,e,load +gwf-sfr.f90:1593,reach diversion specified more than once,K6,logical,table,data,uncond,partly,PK,e,load +gwf-sfr.f90:1574,invalid cprior type,K2,logical,field,data,uncond,no,CHECK(in),e,load +gwf-sfr.f90:1908,"FRACTION divflow outside [0,1]",K2,logical,row,data,uncond,no,CHECK(row: cprior -> range),e,rp +gwf-sfr.f90:4269,reach length must be > 0,K2,logical,field,data,uncond,no,CHECK,e,load +gwf-sfr.f90:4434,connection not reciprocal,K10,logical,table,data,uncond,no,assertion:self-join,e,ar +gwf-sfr.f90:4961,upstream fractions do not sum to one,K10,logical,table,data,uncond,no,assertion:aggregate,e,ar +gwf-sfr.f90:1442,reach graph has a cycle (warning),K10,logical,table,data,uncond,no,none:recursive,w,ar +gwf-sfr.f90:1658,initial stage below reach top,K10,logical,row,data,uncond,no,CHECK(row),e,ar +gwf-sfr.f90:3369,obs boundname must be unique,K6,logical,multi,data,opt:boundnames(logical),no,UNIQUE(conditional),e,rp +gwf-lak.f90:792,IMPLICIT requires a GWF connection per lake,K5,logical,table,data,opt:implicit(logical),no,assertion:aggregate,e,load +gwf-lak.f90:1092,telev >= belev for horizontal connection,K10,logical,row,data,opt:ctype,no,CHECK(row),e,load +gwf-lak.f90:1097,belev >= cell bottom,K10,logical,multi,data,uncond,no,assertion:join,e,ar +gwf-lak.f90:1122,lake connection specified more than once,K6,logical,table,data,uncond,yes,PK,e,load +gwf-lak.f90:1171,NTABLES > 0 but no TABLES block,K4,logical,component,cfg-data,dim,no,CHECK(count(*)=dim),e,load +gwf-lak.f90:1387,table stage not increasing,K2,logical,table,data,uncond,no,none:ordered rows,e,load +gwf-lak.f90:3116,DEV_FORCE_LEGACY and DEV_FORCE_LEGACY_LAKE,K5,logical,component,config,opt:dev(logical),no,CHECK(row),e,load +gwf-lak.f90:6455,IMPLICIT not with BUY,K8,logical,multi,config,presence:buy,no,assertion:exclusion,e,load +gwf-maw.f90:1234,one connection per cell unless MEAN,K6,logical,table,data,opt:condeqn,no,UNIQUE(conditional),e,load +gwf-maw.f90:1311,ANGLEDATA block without NON_VERTICAL_WELLS,K4,logical,component,data,opt:non_vertical_wells(logical),no,conditional table,e,load +gwf-maw.f90:1193,screen radius <= well radius,K10,logical,multi,data,uncond,no,assertion:join,e,load +gwf-maw.f90:972,CONDEQN deprecated spelling,K1,layout,field,data,uncond,no,none:alias,w,load +gwf-uzf.f90:1823,THTS must be > THTR,K2,logical,row,data,uncond,no,CHECK(row),e,load +gwf-uzf.f90:1799,SURFDEP > cell thickness,K10,logical,multi,data,uncond,no,assertion:join,e,load +gwf-uzf.f90:2063,total UZF area exceeds cell area,K10,logical,multi,data,uncond,no,assertion:aggregate,e,ar +gwf-csub.f90:455,STO SS must be zero when CSUB used,K8,logical,multi,data,presence:csub,no,assertion:cross-component,e,ar +gwf-csub.f90:1610,NDELAYCELLS odd for effective stress,K5,logical,component,config,opt:head_based(mixed),no,CHECK(row),e,load +gwf-csub.f90:1474,BOUNDNAME required for entry (BOUNDNAMES on),K4,logical,row,data,opt:boundnames(logical),yes,NOT NULL(conditional),e,load +gwf-csub.f90:2511,only first/last period steady with interbeds,K9,logical,multi,data,presence:csub,no,assertion:cross-component,e,rp +gwf-csub.f90:3928,small/negative effective stress,-,runtime,cell,data,uncond,no,none:state,e,ad +gwf-mvr.f90:789,MODELNAMES only for exchange MVR,K5,layout,component,config,binding,partly,none:fixed by binding,e,load +gwf-mvr.f90:798,MODELNAMES required for exchange MVR,K5,layout,component,config,binding,partly,none:fixed by binding,e,load +gwf-mvr.f90:373,provider not listed in PACKAGES,K7,logical,row,data,uncond,yes,FK,e,load +gwf-mvr.f90:917,package listed twice in PACKAGES,K6,logical,table,data,uncond,yes,PK,e,load +gwf-mvr.f90:935,number of packages != MAXPACKAGES,K3,logical,table,cfg-data,dim,partly,CHECK(count(*)=dim),e,load +gwf-mvr.f90:963,package lacks MOVER option,K7,logical,multi,cfg-data,opt:mover(logical/binding),partly,FK(conditional role),e,load +Mover.f90:112,provider and receiver are the same,K10,logical,row,data,uncond,no,CHECK(row),e,rp +Mover.f90:154,provider id out of package size,K7,logical,multi,data,binding,partly,FK(role),e,rp +exg-gwfgwf.f90:187,first model not of correct type,K7,logical,multi,config,uncond,partly,FK(typed),e,load +exg-gwfgwf.f90:220,models in different solutions,K8,logical,multi,config,binding,no,assertion:join,e,load +exg-gwfgwf.f90:294,XT3D requires ANGLDEGX aux,K4,logical,multi,config,presence:xt3d,no,assertion:cross-component,e,load +exg-gwfgwf.f90:1361,number of exchanges != number of GNCs,K3,logical,multi,data,binding,no,assertion:cross-component,e,load +exg-gwfgwf.f90:1353,explicit GNC with Newton exchange,K5,logical,multi,config,opt:newton(logical),no,assertion:cross-component,e,load +exg-gwfgwt.f90:165,GWF and GWT in different solutions,K8,logical,multi,config,binding,no,assertion:join,e,load +exg-gwfgwt.f90:231,GWF and GWT discretizations differ,K8,logical,multi,data,binding,no,assertion:equal grids,e,ar +exg-gwfgwt.f90:447,GWT-GWT exchange without matching GWF-GWF,K8,logical,multi,data,binding,no,assertion:equal tables,e,load +gwf-buy.f90:264,BUY species without concentration,K7,logical,multi,config,binding,no,FK(model),e,ar +tsp-fmi.f90:123,GWF MVR active but no MVT,K8,logical,multi,config,presence:mvr,no,assertion:co-occurrence,e,load +tsp-fmi.f90:676,budget file lacks specific discharge,K8,logical,multi,config,presence:fmi file,no,none:external file content,e,ar +tsp-ssm.f90:160,no flows need SSM terms,K8,logical,multi,config,binding,no,assertion:exists,e,ar +tsp-ssm.f90:810,flow package cannot be found,K7,logical,multi,data,binding,partly,FK(rule flow_model),e,ar +tsp-ssm.f90:820,package more than once in SOURCES,K6,logical,table,data,uncond,no,PK,e,load +tsp-ssm.f90:995,auxiliary name cannot be found,K7,logical,multi,data,binding,no,FK(other component's column),e,ar +TspSpc.f90:347,SPC MAXBOUND < flow package bounds,K3,logical,multi,cfg-data,binding,no,assertion:cross-component,e,rp +TspSpc.f90:360,array RCH needs array SPC,K7,logical,multi,config,binding,partly,FK(key kind),e,load +gwt-mst.f90:1285,sorption active but no BULK_DENSITY,K4,logical,component,data,opt:sorption(logical/schema),no,NOT NULL(conditional),e,load +gwt-mst.f90:1306,BULK_DENSITY given without sorption,K4,logical,component,data,opt:sorption(logical/schema),no,NULL(conditional),w,load +gwe-ctp.f90:160,cell already constant temperature,K6,logical,multi,data,uncond,no,UNIQUE(cell) across packages,e,rp +doc:gwt/sft.tex,SFT name defaults to SFR name,K7,logical,multi,config,binding,partly,FK(default from name),e,load +doc:gwt/lkt.tex,LKT needs corresponding LAK,K8,logical,multi,config,binding,no,assertion:co-occurrence,e,load +prt-prp.f90:928,DRY_TRACKING_METHOD enum,K2,logical,field,config,uncond,partly,CHECK(in),e,load +prt-prp.f90:937,EXIT_SOLVE_TOLERANCE positive,K2,logical,field,config,uncond,no,CHECK,e,load +prt-prp.f90:585,release point not in cell,K10,logical,multi,data,uncond,no,none:geometry predicate,e,ar +prt-prp.f90:1175,release point in IDOMAIN<=0 cell,K7,logical,multi,data,uncond,partly,FK(active cell),e,load +prt-prp.f90:1182,"LOCAL_Z in [0,1]",K2,logical,field,data,opt:local_z(logical),no,CHECK(conditional),e,load +prt-prp.f90:1261,RELEASETIMES block required when NRELEASETIMES>0,K4,logical,component,cfg-data,dim,no,CHECK(count(*)=dim),e,load +prt-prp.f90:1279,release times strictly increase,K9,logical,table,data,uncond,no,none:ordered rows,e,load +prt-prp.f90:496,release time after end of simulation,K9,logical,multi,data,opt:extend_tracking(logical),no,assertion:cross-component,w,ar +swf-cxs.f90:384,IFNO out of 1..NSECTIONS,K7,logical,row,data,dim,yes,FK,e,load +swf-cxs.f90:394,NXSPOINTS > 1,K2,logical,field,data,uncond,no,CHECK,e,load +swf-cxs.f90:515,CROSSSECTIONDATA rows not grouped as PACKAGEDATA implies,K1,layout,table,data,uncond,no,none:row order,e,load +swf-dfw.f90:1350,velocity needs ANGLDEGX,K4,logical,multi,data,opt:save_velocity(output),no,assertion:cross-component,e,ar +doc:olf/dfw.tex,DFW required for each OLF model,K8,logical,multi,config,uncond,no,assertion:exists,e,load +TimeSeries.f90:1182,METHODS does not cover all series,K3,logical,row,config,uncond,partly,CHECK(count),e,load +TimeSeries.f90:552,series does not cover the integration interval,K9,logical,multi,data,binding,no,assertion:cross-component,e,ad +TimeSeriesManager.f90:113,duplicate time-series file name,K6,logical,table,config,uncond,no,UNIQUE,e,load +TimeSeriesManager.f90:564,value neither number nor TS name,K7,logical,field,data,binding,partly,FK(union: number|TS name),e,rp +TimeArraySeriesManager.f90:304,time-array series not found,K7,logical,multi,data,binding,partly,FK,e,load +doc:gwf/tsi.tex,a TAS file belongs to one package,K6,logical,multi,config,uncond,no,UNIQUE,e,load +Obs.f90:874,observation type not found,K2,logical,field,data,binding,yes,"CHECK(in, by target type)",e,load +Obs.f90:1000,OBS outfile provided more than once,K6,logical,table,config,uncond,no,UNIQUE,e,load +gwf-lak.f90:5208,integer ID but ID2 missing,K4,logical,row,data,opt:obstype,no,CHECK(row),e,load +gwf-csub.f90:7316,BOUNDNAME not allowed for this obstype,K4,logical,row,data,opt:obstype,partly,CHECK(row),e,load +OutputControlData.f90:73,SAVE requested but no save file,K5,output,component,config,opt:save(output),no,CHECK(row),e,load +doc:prt/fmi.tex,GWF OC must SAVE BUDGET ALL for PRT,K8,output,multi,config,binding,no,none:external run,e,load diff --git a/constraint-audit/mf6_sites.py b/constraint-audit/mf6_sites.py new file mode 100644 index 00000000..38d74974 --- /dev/null +++ b/constraint-audit/mf6_sites.py @@ -0,0 +1,184 @@ +"""Pass 2, step 1-2 of constraint-audit-plan.md: extract and cluster MF6 error/warning sites. + +Usage: python mf6_sites.py +Writes mf6-checks.csv (one row per store_error / store_warning site) and prints tallies. +store_error_filename / store_error_unit are follow-ups that attach a file to earlier errors; +they are counted but not treated as sites. +""" + +import collections +import csv +import os +import re +import sys + +src, out = sys.argv[1:3] + +CALL = re.compile(r"^\s*(?:if\s*\(.*\)\s*)?call\s+(store_error|store_warning)\s*\(", re.I) +FOLLOW = re.compile(r"call\s+(store_error_filename|store_error_unit)\b", re.I) +MSGSTART = re.compile(r"write\s*\(\s*(errmsg|warnmsg|msg|ermsg|errmsgr|wmsg)\b|^\s*(errmsg|warnmsg|msg|ermsg)\s*=", re.I) +ROUTINE = re.compile(r"^\s*(?:(?:recursive|pure|elemental|module)\s+)*(subroutine|function)\s+(\w+)", re.I) +STR = re.compile(r"'((?:[^']|'')*)'|\"((?:[^\"]|\"\")*)\"") + +# Normalised message -> cluster. First match wins; order matters. +CLUSTERS = [ + ("not-input", r"could not open|error opening|unit number|already open|file .*not (found|exist)|" + r"iostat|allocat|programming error|internal error|not implemented|" + r"memory manager|mem_|should not happen|developer|unknown memory|" + r"did not converge|failed to converge|convergence failure|nonconvergence|" + r"petsc|mpi|binary file|header|end of file|reached end|eof|program(mer)? error|" + r"error reading|netcdf library|read_line|insufficient|needs new case|aborted|went dry|" + r"adjusted theta|must be overridden|maxobstypes"), + ("binding-context", r"exchange|solution|\\brank\\b|connect(ed|ing)? .*model|model .* (connected|in different)|" + r"flow model|gwf model|gwfbudget|fmi"), + ("enum", r"must be (one of |either )?['\"]?[A-Z_]+['\"]?,? (or|,)|unknown .* specified|unsupported|" + r"must be 1 .* or 2|looking for .* found"), + ("unknown-keyword", r"unknown .*(keyword|option|block|tag|record)|unrecognized|not recognized|" + r"invalid (keyword|option|tag)|not a valid (keyword|option)"), + ("block-structure", r"block|begin|end .* expected|required .* not found|not found .* block"), + ("deprecated", r"deprecat|no longer supported|removed|obsolete"), + ("duplicate", r"duplicat|more than once|already (been )?(defined|specified|exists|used)|unique|" + r"multiple .* (same|specified)"), + ("not-found-ref", r"not found|does not exist|could not find|cannot find|no .* (named|with name)|" + r"invalid .* name|not (a )?valid .* name|unknown .* name|" + r"(model|package|boundname|time.?series|feature|reach|lake|well) .* not"), + ("cellid-grid", r"cell ?id|cellid|node .* (outside|out of|invalid|exceed)|outside (of )?(the )?(model|grid)|" + r"idomain|inactive|reduced node|user node"), + ("range-count", r"less than (1|one) or greater than|was not specified or was specified incorrectly|" + r"at most|exceed|greater than (the )?(number|max|nbound|maxbound)|maximum number|" + r"out of range|outside .* range|must be between|must be (<|less than or equal) .*(n|max)|" + r"number of .* (must|should) (equal|match|be)|does not (equal|match)|mismatch|" + r"too (many|few)|not enough|invalid number"), + ("sign-value", r"must be (greater|less|positive|negative|non-?negative|>|<|=|zero|between)|" + r"(cannot|can not|may not) be (negative|zero|less|greater)|" + r"less than (zero|0)|negative|<= *0|< *0|> *0|positive|non-?zero|zero"), + ("geometry", r"top|bot(tom)?|thickness|elevation|vertex|vertices|clockwise|area|length|" + r"symmetr|ihc|ja |\bja\b|connection|cl12|hwva|angldegx|cell2d"), + ("option-coherence", r"specified but .* not (specified|provided|active)|not (specified|provided) .* but|" + r"must be .* when|when using|missing|cannot be (used|specified|combined)|can not be (used|specified)|" + r"only (be )?(used|specified|valid|available|supported)|requires|required when|" + r"must (also )?be (specified|used|active|activated|on)|not compatible|incompatible|" + r"not (allowed|supported|permitted)|mutually exclusive|in conjunction|" + r"must be specified|is not active|not activated|is activated"), + ("temporal", r"period|kper|nper|stress period|time ?step|\btime\b|tsmult|perlen|delt|ats"), + ("type-parse", r"invalid|not (a )?valid|could not (read|parse|convert)|expected|" + r"must be an? (integer|real|number|string)"), +] +CLUSTERS = [(n, re.compile(p, re.I)) for n, p in CLUSTERS] + + +def family(path): + p = path.replace("\\", "/") + for key, fam in [("Idm/", "idm"), ("Utilities/", "utilities"), ("Solution/", "solution"), + ("Exchange/", "exchange"), ("Timing/", "timing"), ("Distributed/", "distributed")]: + if key in p: + return fam + m = re.search(r"Model/(\w+)/", p) + if m: + return m.group(1) + return "top" + + +def join(lines): + """Join Fortran continuation lines, including strings continued with &...&.""" + s = "\n".join(lines) + s = re.sub(r"&[ \t]*(?:![^\n]*)?\n[ \t]*&", "", s) # string continuation + s = re.sub(r"&[ \t]*(?:![^\n'\"]*)?\n", " ", s) # statement continuation + return s + + +def literals(lines, fmts=None): + s = join(lines) + parts = [m.group(1) or m.group(2) or "" for m in STR.finditer(s)] + if fmts: + for name in re.findall(r"write\s*\(\s*\w+\s*,\s*(\w+)\s*\)", s, re.I): + if name.lower() in fmts: + parts.insert(0, fmts[name.lower()]) + return re.sub(r"\s+", " ", " ".join(parts)).strip() + + +FMT = re.compile(r"character\s*\(\s*len\s*=\s*\*\s*\)\s*,\s*parameter\s*::\s*(\w+)\s*=", re.I) + + +def formats(text): + """Named format constants in a file: name -> literal text.""" + out = {} + for i, ln in enumerate(text): + m = FMT.search(ln) + if m: + j, block = i, [ln[m.end():]] + while block[-1].rstrip().endswith("&") and j + 1 < len(text): + j += 1 + block.append(text[j]) + out[m.group(1).lower()] = literals(block) + return out + + +KIND = {"not-input": "-", "binding-context": "K8", "enum": "K2", "unknown-keyword": "K1", + "block-structure": "K1", "deprecated": "K1", "duplicate": "K6", "not-found-ref": "K7", + "cellid-grid": "K7", "range-count": "K3", "sign-value": "K2", "geometry": "K10", + "option-coherence": "K5", "temporal": "K9", "type-parse": "K1", "unclassified": "?"} + +rows = [] +follow = collections.Counter() +for root, _, files in os.walk(src): + for fn in files: + if not fn.lower().endswith((".f90",)): + continue + path = os.path.join(root, fn) + rel = os.path.relpath(path, src) + text = open(path, errors="replace").read().splitlines() + routine = "" + fmts = formats(text) + for i, ln in enumerate(text): + code = ln.split("!")[0] if "'" not in ln and '"' not in ln else ln + m = ROUTINE.match(code) + if m and not re.match(r"^\s*end\b", code, re.I): + routine = m.group(2) + fm = FOLLOW.search(code) + if fm: + follow[fm.group(1).lower()] += 1 + continue + cm = CALL.match(code) + if not cm: + continue + # Gather the call statement (continuation lines). + j = i + stmt = [text[j]] + while stmt[-1].rstrip().endswith("&") and j + 1 < len(text): + j += 1 + stmt.append(text[j]) + msg = literals(stmt, fmts) + # Look back for the message-building statement if the call passes a variable. + if len(msg) < 15: + for k in range(i - 1, max(i - 25, -1), -1): + if CALL.match(text[k]) or ROUTINE.match(text[k]): + break + if MSGSTART.search(text[k]): + block = text[k:i] + msg = (literals(block, fmts) + " " + msg).strip() + break + cluster = "unclassified" + for name, rx in CLUSTERS: + if rx.search(msg): + cluster = name + break + rows.append(dict(file=rel, line=i + 1, routine=routine, family=family(rel), + call=cm.group(1).lower(), cluster=cluster, kind=KIND[cluster], message=msg[:300])) + +os.makedirs(out, exist_ok=True) +with open(os.path.join(out, "mf6-checks.csv"), "w", newline="") as fh: + w = csv.DictWriter(fh, fieldnames=list(rows[0])) + w.writeheader() + w.writerows(sorted(rows, key=lambda r: (r["family"], r["file"], r["line"]))) + +print(f"{len(rows)} sites; follow-ups: {dict(follow)}") +print("by call:", dict(collections.Counter(r["call"] for r in rows))) +print("empty message:", sum(1 for r in rows if not r["message"])) +print("\nby cluster:") +for c, n in collections.Counter(r["cluster"] for r in rows).most_common(): + print(f" {c:18} {n:5}") +print("\nby tentative kind:", dict(sorted(collections.Counter(r["kind"] for r in rows).items()))) +print("\nby family:") +for c, n in collections.Counter(r["family"] for r in rows).most_common(): + print(f" {c:18} {n:5}") diff --git a/constraint-audit/sample.py b/constraint-audit/sample.py new file mode 100644 index 00000000..f3f055a5 --- /dev/null +++ b/constraint-audit/sample.py @@ -0,0 +1,249 @@ +"""Pass 2 step 3 + pass 3 of constraint-audit-plan.md: hand-labelled sample of MF6 checks and +mf6io prose rules. Writes mf6-sample.csv and prints tallies. + +Usage: python sample.py + +Columns + site MF6 file:line, or doc: for prose-only rules + kind K1..K10, or L (lens domain: where a layout's partial lens is defined) + level layout | logical | runtime + arity field | row | table | component | multi | sim + cd config | data | cfg-data (relates configuration to data) + trigger uncond | opt:(