Update xclim

.github/workflows/main.yml

@@ -45,7 +45,7 @@ jobs:
         shell: bash -l {0}
     strategy:
         matrix:
-            python-version: [ "3.9", "3.10", "3.11" ] # "3.12"
+            python-version: [ "3.10", "3.11", "3.12"]
     steps:
       - name: Harden Runner
         uses: step-security/harden-runner@0080882f6c36860b6ba35c610c98ce87d4e2f26f # v2.10.2

.pre-commit-config.yaml

@@ -6,7 +6,7 @@ repos:
     rev: v3.17.0
     hooks:
       - id: pyupgrade
-        args: [ '--py39-plus' ]
+        args: [ '--py310-plus' ]
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v5.0.0
     hooks:
@@ -30,7 +30,7 @@ repos:
     rev: 24.10.0
     hooks:
       - id: black
-        args: [ '--target-version=py39' ]
+        args: [ '--target-version=py310' ]
   - repo: https://github.com/pycqa/flake8
     rev: 7.1.1
     hooks:
@@ -46,10 +46,10 @@ repos:
     rev: 1.8.7
     hooks:
       - id: nbqa-black
-        args: [ '--target-version=py39' ]
+        args: [ '--target-version=py310' ]
         additional_dependencies: [ 'black==24.8.0' ]
       - id: nbqa-pyupgrade
-        args: [ '--py39-plus' ]
+        args: [ '--py310-plus' ]
         additional_dependencies: [ 'pyupgrade==3.17.0' ]
       - id: nbqa-isort
         args: [ '--settings-file=setup.cfg' ]

CHANGELOG.rst

@@ -10,9 +10,10 @@ v0.13.0 (unreleased)
 * Permissions have been set for all workflows.
 * Pinned `cf-xarray`` below v0.9.3.
 * Reformatted and renamed `CHANGES.rst` to `CHANGELOG.rst`.
-* Dropped support for Python 3.8. The supported versions are now Python 3.9, 3.10, and 3.11.
+* Dropped support for Python 3.8 and 3.9. The supported versions are now Python 3.10, 3.11 and 3.12.
 * `black` has been updated to v2024.8.0, and coding conventions have been updated to Python3.9+.
 * Several dependencies now are pinned to baseline versions, including `anyascii`, `dask`, `ipython`, `matplotlib`, `nbsphinx` `numpy`, and `sphinxcontrib-bibtex`.
+* Added `xscen` dependency. Now used in spatial averaging of ensemble processes.
 
 v0.12.1 (2024-06-25)
 --------------------

environment-docs.yml

@@ -2,8 +2,8 @@ name: finch
 channels:
   - conda-forge
 dependencies:
-  - python >=3.8,<3.12
-  - anyascii >=0.3.2
+  - python >=3.10,<3.13
+  - anyascii >=0.3.0
   - birdy >=0.8.1
   - ipython >=8.0.0
   - matplotlib-base >=3.5.0

environment.yml

@@ -2,11 +2,11 @@ name: finch
 channels:
   - conda-forge
 dependencies:
-  - python >=3.9,<3.12
+  - python >=3.9,<3.13
   - pip >=24.2.0
   - anyascii >=0.3.0
-  - cftime
-  - cf_xarray >=0.6.1,<0.9.3
+  - cftime >=1.4.1
+  - cf_xarray >=0.9.3
   - click >=8.0.0
   - clisops >=0.11.0
   - dask >=2023.5.1
@@ -15,16 +15,16 @@ dependencies:
   - h5netcdf
   - netcdf4 <=1.6.5
   - numpy >=1.23.0
-  - pandas >=1.5.3,<2.2.0
+  - pandas >=2.2.0
   - parse
   - psutil
   - python-slugify
   - pywps >=4.5.1
-  - pyyaml
-  - requests>=2.32.2
+  - pyyaml >=6.0.1
   - scipy >=1.9.0
   - sentry-sdk
   - siphon
-  - xarray >=2023.01.0,<2023.11.0
-  - xclim =0.43  # remember to match xclim version in requirements_docs.txt as well
-  - xesmf >=0.8.2
+  - xarray >=2023.11.0
+  - xclim =0.52.2  # remember to match xclim version in requirements_docs.txt as well
+  - xesmf >=0.8.2,!=0.8.8
+  - xscen

finch/processes/__init__.py

@@ -1,6 +1,5 @@
 # noqa: D104
 import logging
-from pathlib import Path
 
 from pywps.configuration import get_config_value
 from xclim.core.indicator import registry as xclim_registry
@@ -52,11 +51,20 @@ def filter_func(elem):
 
 not_implemented = [
     "DC",
+    "E_SAT",
     "FWI",
+    "HURS",
+    "HURS_FROMDEWPOINT",
+    "HUSS",
+    "HUSS_FROMDEWPOINT",
     "RH",
     "RH_FROMDEWPOINT",
-    "E_SAT",
-    "HUSS",
+    "SPI",
+    "SPEI",
+    "WIND_POWER_POTENTIAL",
+    "WIND_PROFILE",
+    "WIND_SPEED_FROM_VECTOR",
+    "WIND_VECTOR_FROM_SPEED",
 ]
 
 
@@ -135,7 +143,7 @@ def get_processes():
         SubsetPolygonProcess(),
         AveragePolygonProcess(),
         HourlyToDailyProcess(),
-        GeoseriesToNetcdfProcess(),
+        # GeoseriesToNetcdfProcess(),
     ]
 
     return processes

finch/processes/ensemble_utils.py

@@ -6,9 +6,11 @@
 from collections.abc import Iterable
 from copy import deepcopy
 from dataclasses import dataclass
+from datetime import datetime
 from pathlib import Path
 from typing import Optional, Union
 
+import geopandas as gpd
 import pandas as pd
 import xarray as xr
 from pandas.api.types import is_numeric_dtype
@@ -21,7 +23,9 @@
 from xclim.core.calendar import days_since_to_doy, doy_to_days_since, percentile_doy
 from xclim.core.indicator import Indicator
 from xclim.indicators.atmos import tg
+from xscen.aggregate import spatial_mean
 
+from . import wpsio
 from .subset import finch_subset_bbox, finch_subset_gridpoint, finch_subset_shape
 from .utils import (
     DatasetConfiguration,
@@ -71,9 +75,9 @@ class Dataset:  # noqa: D101
     model: str
     scenario: str
     frequency: str = "day"
-    realization: Optional[str] = None
-    date_start: Optional[str] = None
-    date_end: Optional[str] = None
+    realization: str | None = None
+    date_start: str | None = None
+    date_end: str | None = None
 
     @classmethod
     def from_filename(cls, filename, pattern):  # noqa: D102
@@ -86,10 +90,10 @@ def from_filename(cls, filename, pattern):  # noqa: D102
 def file_is_required(
     filename: str,
     pattern: str,
-    model_lists: Optional[dict[str, list[str]]] = None,
+    model_lists: dict[str, list[str]] | None = None,
     variables: list[str] = None,
     scenario: str = None,
-    models: list[Union[str, tuple[str, int]]] = None,
+    models: list[str | tuple[str, int]] = None,
 ):
     """Parse metadata and filter datasets."""
     file = Dataset.from_filename(filename, pattern)
@@ -177,9 +181,9 @@ def _make_resource_input(url: str, workdir: str, local: bool):
 def get_datasets(
     dsconf: DatasetConfiguration,
     workdir: str,
-    variables: Optional[list[str]] = None,
-    scenario: Optional[str] = None,
-    models: Optional[list[str]] = None,
+    variables: list[str] | None = None,
+    scenario: str | None = None,
+    models: list[str] | None = None,
 ) -> list[PywpsInput]:
     """Parse a directory to find files and filters the list to return only the needed ones, as resource inputs.
 
@@ -215,7 +219,7 @@ def get_datasets(
     return inputs
 
 
-def _formatted_coordinate(value) -> Optional[str]:
+def _formatted_coordinate(value) -> str | None:
     """Return the first float value.
 
     The value can be a comma separated list of floats or a single float.
@@ -350,22 +354,44 @@ def make_file_groups(files_list: list[Path], variables: set) -> list[dict[str, P
 
 
 def make_ensemble(
-    files: list[Path], percentiles: list[int], average_dims: Optional[tuple[str]] = None
+    files: list[Path],
+    percentiles: list[int],
+    spatavg: bool | None = False,
+    region: dict | None = None,
 ) -> None:  # noqa: D103
     ensemble = ensembles.create_ensemble(
         files, realizations=[file.stem for file in files]
     )
     # make sure we have data starting in 1950
     ensemble = ensemble.sel(time=(ensemble.time.dt.year >= 1950))
 
+    if ensemble.lon.size == 1 and ensemble.lat.size == 1 and spatavg:
+        ensemble.attrs["history"] = (
+            f"[{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}] "
+            f"spatial average flag is set to True but will be skipped as dataset contains a "
+            f"single point\n{ensemble.attrs.get('history', '')}"
+        )
+        spatavg = False
+
     # If data is in day of year, percentiles won't make sense.
     # Convert to "days since" (base will be the time coordinate)
     for v in ensemble.data_vars:
         if ensemble[v].attrs.get("is_dayofyear", 0) == 1:
             ensemble[v] = doy_to_days_since(ensemble[v])
 
-    if average_dims is not None:
-        ensemble = ensemble.mean(dim=average_dims)
+    if spatavg:
+        # ensemble = ensemble.mean(dim=average_dims)
+        if region is None:
+            method = "cos-lat"
+        else:
+            method = "xesmf"
+        ensemble = spatial_mean(
+            ds=ensemble,
+            method=method,
+            spatial_subset=False,
+            region=region,
+            kwargs={"skipna": True},
+        )
 
     if percentiles:
         ensemble_percentiles = ensembles.ensemble_percentiles(
@@ -540,13 +566,28 @@ def ensemble_common_handler(
     )
 
     if single_input_or_none(request.inputs, "average"):
+        spatavg = True
         if subset_function == finch_subset_gridpoint:
-            average_dims = ("region",)
+            region = None
+        elif subset_function == finch_subset_bbox:
+            lon0 = single_input_or_none(request.inputs, wpsio.lon0.identifier)
+            lat0 = single_input_or_none(request.inputs, wpsio.lat0.identifier)
+            lon1 = single_input_or_none(request.inputs, wpsio.lon1.identifier)
+            lat1 = single_input_or_none(request.inputs, wpsio.lat1.identifier)
+            bbox = dict(lat_bnds=[lat0, lat1], lon_bnds=[lon0, lon1])
+            region = dict(name="region", method="bbox", **bbox)
         else:
-            average_dims = ("lat", "lon")
+            shp = gpd.read_file(
+                Path(request.inputs[wpsio.shape.identifier][0].file)
+            ).to_crs("EPSG:4326")
+            shp["geometry"] = shp.make_valid()
+            region = dict(name="region", method="shape", shape=shp)
     else:
-        average_dims = None
-    write_log(process, f"Will average over {average_dims}")
+        # average_dims = None
+        region = None
+        spatavg = False
+
+    write_log(process, f"Will average over {region}")
 
     base_work_dir = Path(process.workdir)
     ensembles = []
@@ -625,7 +666,12 @@ def ensemble_common_handler(
         warnings.filterwarnings("default", category=FutureWarning)
         warnings.filterwarnings("default", category=UserWarning)
 
-        ensemble = make_ensemble(indices_files, ensemble_percentiles, average_dims)
+        ensemble = make_ensemble(
+            files=indices_files,
+            percentiles=ensemble_percentiles,
+            spatavg=spatavg,
+            region=region,
+        )
         ensemble.attrs["source_datasets"] = "\n".join(
             [dsinp.url for dsinp in netcdf_inputs]
         )
@@ -651,7 +697,7 @@ def ensemble_common_handler(
             ensemble = ensemble.round(prec)
             prec = 0
         df = dataset_to_dataframe(ensemble)
-        if average_dims is None:
+        if spatavg is None:
             dims = ["lat", "lon", "time"]
         else:
             dims = ["time"]

finch/processes/modules/humidex.yml

@@ -19,10 +19,10 @@ indicators:
         default: "YS"
       threshold:
         default: "30"
-        units: ""
+        units: "1"
 
 variables:
   HXmax:
-    canonical_units: ""
+    canonical_units: "1"
     cell_methods: "time: max"
     description: Daily Maximum Humidex

finch/processes/modules/streamflow.yml

@@ -3,7 +3,7 @@ doc:
 realm: land
 indicators:
     discharge_stats:
-        base: xclim.core.indicator.stats
+        base: stats
         cf_attrs:
             long_name: Daily flow statistics
             description: "{freq} {op} of daily flow ({indexer})."
@@ -12,7 +12,7 @@ indicators:
         input:
             da: discharge
     freq_analysis:
-        base: xclim.core.indicator.return_level
+        base: return_level
         cf_attrs:
             long_name: N-year return level discharge
             description: "Streamflow frequency analysis for the {mode} {indexer} {window}-day flow estimated using the {dist} distribution."
@@ -21,7 +21,7 @@ indicators:
         input:
             da: discharge
     discharge_distribution_fit:
-        base: xclim.core.indicator.fit
+        base: fit
         cf_attrs:
             var_name: params
         input: