Merge branch 'main' into spec0-calendar

Ouranosinc · Sep 17, 2024 · 9245be5 · 9245be5
2 parents 20d45b3 + 08656c1
commit 9245be5
Show file tree

Hide file tree

Showing 9 changed files with 76 additions and 65 deletions.
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -23,6 +23,7 @@ New features and enhancements
 Bug fixes
 ^^^^^^^^^
 * Fixed a small inefficiency in ``_otc_adjust``, and the `standardize` method of `OTC/dOTC` is now applied on individual variable  (:pull:`1890`, :pull:`1896`).
+* Remove deprecated cells in the tutorial notebook `sdba.ipynb` (:pull:`1895`).
 
 Breaking changes
 ^^^^^^^^^^^^^^^^
@@ -49,6 +50,22 @@ Internal changes
 * ``xclim.core`` now uses absolute imports for clarity and some objects commonly used in the module have been moved to hidden submodules. (:issue:`1719`, :pull:`1881`).
 * ``xclim.core.indicator.Parameter`` has a new attribute ``compute_name`` while ``xclim.core.indicator.Indicator`` lost its ``_variable_mapping``. The translation from parameter (and variable) names in the indicator to the names on the compute function is handled by ``Indicator._get_compute_args``. (:pull:`1885`).
 
+v0.52.2 (2024-09-16)
+--------------------
+Contributors to this version: Pascal Bourgault (:user:`aulemahal`).
+
+Bug fixes
+^^^^^^^^^
+* Fixed ``decimal_year`` import, fixed functions ``rate2amount``, ``amount2rate``, ``time_bnds`` and ``stack_periods``  for `xarray` version 2024.09.0. Removed ``datetime_to_decimal_year`` as the mirrored `xarray` function was replaced by ``ds.time.dt.decimal_year``. (:pull:`1920`).
+
+v0.52.1 (2024-09-11)
+--------------------
+Contributors to this version: Trevor James Smith (:user:`Zeitsperre`).
+
+Bug fixes
+^^^^^^^^^
+* Adjusted the required base version of `pyarrow` to be `v10.0.1` to address an environment resolution error on conda-forge. (:pull:`1918`).
+
 v0.52.0 (2024-08-08)
 --------------------
 Contributors to this version: David Huard (:user:`huard`), Trevor James Smith (:user:`Zeitsperre`), Hui-Min Wang (:user:`Hem-W`), Éric Dupuis (:user:`coxipi`), Sarah Gammon (:user:`SarahG-579462`), Pascal Bourgault (:user:`aulemahal`), Juliette Lavoie (:user:`juliettelavoie`), Adrien Lamarche (:user:`LamAdr`).

diff --git a/docs/notebooks/sdba.ipynb b/docs/notebooks/sdba.ipynb
@@ -641,52 +641,30 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "dref[\"pr\"] = dref.pr.chunk({\"location\": 1})\n",
-    "dref[\"tasmax\"] = dref.tasmax.chunk({\"location\": 1})\n",
+    "from xclim.core.units import convert_units_to\n",
+    "from xclim.testing import open_dataset\n",
     "\n",
-    "dhist[\"pr\"] = dhist.pr.chunk({\"location\": 1})\n",
-    "dhist[\"tasmax\"] = dhist.tasmax.chunk({\"location\": 1})\n",
+    "dref = open_dataset(\n",
+    "    \"sdba/ahccd_1950-2013.nc\", chunks={\"location\": 1}, drop_variables=[\"lat\", \"lon\"]\n",
+    ").sel(time=slice(\"1981\", \"2010\"))\n",
     "\n",
-    "dsim[\"pr\"] = dsim.pr.chunk({\"location\": 1})\n",
-    "dsim[\"tasmax\"] = dsim.tasmax.chunk({\"location\": 1})"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "##### Perform an initial univariate adjustment."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# additive for tasmax\n",
-    "QDMtx = sdba.QuantileDeltaMapping.train(\n",
-    "    dref.tasmax, dhist.tasmax, nquantiles=20, kind=\"+\", group=\"time\"\n",
+    "# Fix the standard name of the `pr` variable.\n",
+    "# This allows the convert_units_to below to infer the correct CF transformation (precip rate to flux)\n",
+    "# see the \"Unit handling\" notebook\n",
+    "dref.pr.attrs[\"standard_name\"] = \"lwe_precipitation_rate\"\n",
+    "\n",
+    "dref = dref.assign(\n",
+    "    tasmax=convert_units_to(dref.tasmax, \"K\"),\n",
+    "    pr=convert_units_to(dref.pr, \"kg m-2 s-1\"),\n",
     ")\n",
-    "# Adjust both hist and sim, we'll feed both to the Npdf transform.\n",
-    "scenh_tx = QDMtx.adjust(dhist.tasmax)\n",
-    "scens_tx = QDMtx.adjust(dsim.tasmax)\n",
-    "\n",
-    "# remove == 0 values in pr:\n",
-    "dref[\"pr\"] = sdba.processing.jitter_under_thresh(dref.pr, \"0.01 mm d-1\")\n",
-    "dhist[\"pr\"] = sdba.processing.jitter_under_thresh(dhist.pr, \"0.01 mm d-1\")\n",
-    "dsim[\"pr\"] = sdba.processing.jitter_under_thresh(dsim.pr, \"0.01 mm d-1\")\n",
-    "\n",
-    "# multiplicative for pr\n",
-    "QDMpr = sdba.QuantileDeltaMapping.train(\n",
-    "    dref.pr, dhist.pr, nquantiles=20, kind=\"*\", group=\"time\"\n",
+    "dsim = open_dataset(\n",
+    "    \"sdba/CanESM2_1950-2100.nc\", chunks={\"location\": 1}, drop_variables=[\"lat\", \"lon\"]\n",
     ")\n",
-    "# Adjust both hist and sim, we'll feed both to the Npdf transform.\n",
-    "scenh_pr = QDMpr.adjust(dhist.pr)\n",
-    "scens_pr = QDMpr.adjust(dsim.pr)\n",
     "\n",
-    "# Stack variables : Dataset -> DataArray with `multivar` dimension\n",
-    "dref, dhist, dsim = (sdba.stack_variables(da) for da in (dref, dhist, dsim))"
+    "dhist = dsim.sel(time=slice(\"1981\", \"2010\"))\n",
+    "dsim = dsim.sel(time=slice(\"2041\", \"2070\"))\n",
+    "\n",
+    "dref"
    ]
   },
   {
@@ -702,9 +680,14 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Stack variables : Dataset -> DataArray with `multivar` dimension\n",
+    "ref = sdba.processing.stack_variables(dref)\n",
+    "hist = sdba.processing.stack_variables(dhist)\n",
+    "sim = sdba.processing.stack_variables(dsim)\n",
+    "\n",
     "ADJ = sdba.MBCn.train(\n",
-    "    dref,\n",
-    "    dhist,\n",
+    "    ref,\n",
+    "    hist,\n",
     "    base_kws={\"nquantiles\": 20, \"group\": \"time\"},\n",
     "    adj_kws={\"interp\": \"nearest\", \"extrapolation\": \"constant\"},\n",
     "    n_iter=20,  # perform 20 iteration\n",
@@ -714,8 +697,8 @@
     "scenh, scens = (\n",
     "    ADJ.adjust(\n",
     "        sim=ds,\n",
-    "        ref=dref,\n",
-    "        hist=dhist,\n",
+    "        ref=ref,\n",
+    "        hist=hist,\n",
     "        base=sdba.QuantileDeltaMapping,\n",
     "        base_kws_vars={\n",
     "            \"pr\": {\n",
@@ -725,9 +708,9 @@
     "            },\n",
     "            \"tasmax\": {\"kind\": \"+\"},\n",
     "        },\n",
-    "        adj_kws={\"interp\": \"nearest\", \"extrapolation\": \"constant\"},\n",
+    "        adj_kws={\"interp\": \"linear\", \"extrapolation\": \"constant\"},\n",
     "    )\n",
-    "    for ds in (dhist, dsim)\n",
+    "    for ds in (hist, sim)\n",
     ")"
    ]
   },
@@ -767,12 +750,12 @@
    "outputs": [],
    "source": [
     "fig, axs = plt.subplots(1, 2, figsize=(16, 4))\n",
-    "for da, label in zip((dref, scenh, dhist), (\"Reference\", \"Adjusted\", \"Simulated\")):\n",
+    "for da, label in zip((ref, scenh, hist), (\"Reference\", \"Adjusted\", \"Simulated\")):\n",
     "    ds = sdba.unstack_variables(da).isel(location=2)\n",
     "    # time series - tasmax\n",
     "    ds.tasmax.plot(ax=axs[0], label=label, alpha=0.65 if label == \"Adjusted\" else 1)\n",
     "    # scatter plot\n",
-    "    ds.plot.scatter(x=\"pr\", y=\"tasmax\", ax=axs[1], label=label)\n",
+    "    ds.plot.scatter(x=\"pr\", y=\"tasmax\", ax=axs[1], label=label, marker=\".\", s=40)\n",
     "axs[0].legend()\n",
     "axs[1].legend()"
    ]
@@ -808,7 +791,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.4"
+   "version": "3.12.5"
   },
   "toc": {
    "base_numbering": 1,

diff --git a/environment.yml b/environment.yml
@@ -52,6 +52,7 @@ dependencies:
   - nbsphinx
   - nbval >=0.11.0
   - nc-time-axis >=1.4.1
+  - netcdf4  # Required for some Jupyter notebooks
   - notebook
   - pandas-stubs >=2.2
   - pooch >=1.8.0

diff --git a/pyproject.toml b/pyproject.toml
@@ -48,7 +48,7 @@ dependencies = [
   "packaging >=24.0",
   "pandas >=2.2",
   "pint >=0.18",
-  "pyarrow >=15.0.0", # Strongly encouraged for pandas v2.2.0+
+  "pyarrow >=10.0.1", # Strongly encouraged for pandas v2.2.0+
   "pyyaml >=6.0.1",
   "scikit-learn >=1.1.0",
   "scipy >=1.9.0",
@@ -136,7 +136,7 @@ target-version = [
 ]
 
 [tool.bumpversion]
-current_version = "0.52.1-dev.5"
+current_version = "0.52.3-dev.0"
 commit = true
 commit_args = "--no-verify"
 tag = false

diff --git a/tests/test_testing_utils.py b/tests/test_testing_utils.py
@@ -37,9 +37,6 @@ def file_md5_checksum(f_name):
             hash_md5.update(f.read())
         return hash_md5.hexdigest()
 
-    @pytest.mark.skip(
-        "This test has been significantly modified. Will adjust when #1889 is merged."
-    )
     @pytest.mark.requires_internet
     def test_open_testdata(
         self,

diff --git a/xclim/__init__.py b/xclim/__init__.py
@@ -13,7 +13,7 @@
 
 __author__ = """Travis Logan"""
 __email__ = "[email protected]"
-__version__ = "0.52.1-dev.5"
+__version__ = "0.52.3-dev.0"
 
 
 with _resources.as_file(_resources.files("xclim.data")) as _module_data:

diff --git a/xclim/core/calendar.py b/xclim/core/calendar.py
@@ -773,7 +773,7 @@ def time_bnds(  # noqa: C901
         time = time.indexes[time.name]
     elif isinstance(time, (DataArrayResample, DatasetResample)):
         for grouper in time.groupers:
-            if "time" in grouper.dims:
+            if isinstance(grouper.grouper, xr.groupers.TimeResampler):
                 datetime = grouper.unique_coord.data
                 freq = freq or grouper.grouper.freq
                 if datetime.dtype == "O":
@@ -1354,13 +1354,13 @@ def stack_periods(
     for _, strd_slc in da.resample(time=strd_frq).groups.items():
         win_resamp = time2.isel(time=slice(strd_slc.start, None)).resample(time=win_frq)
         # Get slice for first group
-        win_slc = win_resamp._group_indices[0]
+        win_slc = list(win_resamp.groups.values())[0]
         if min_length < window:
             # If we ask for a min_length period instead is it complete ?
             min_resamp = time2.isel(time=slice(strd_slc.start, None)).resample(
                 time=minl_frq
             )
-            min_slc = min_resamp._group_indices[0]
+            min_slc = list(min_resamp.groups.values())[0]
             open_ended = min_slc.stop is None
         else:
             # The end of the group slice is None if no outside-group value was found after the last element
@@ -1545,7 +1545,7 @@ def _reconstruct_time(_time_as_delta, _start):
     periods = []
     for i, (start, length) in enumerate(zip(starts.values, lengths.values)):
         real_time = _reconstruct_time(time_as_delta, start)
-        slices = real_time.resample(time=strd_frq)._group_indices
+        slices = list(real_time.resample(time=strd_frq).groups.values())
         if i == 0:
             slc = slice(slices[0].start, min(slices[mid].stop, length))
         elif i == da.period.size - 1:

diff --git a/xclim/core/units.py b/xclim/core/units.py
@@ -17,6 +17,7 @@
 
 import cf_xarray.units
 import numpy as np
+import pandas as pd
 import pint
 import xarray as xr
 from boltons.funcutils import wraps
@@ -650,7 +651,10 @@ def _rate_and_amount_converter(
             start = time.indexes[dim][0]
             if not start_anchor:
                 # Anchor is on the end of the period, subtract 1 period.
-                start = start - xr.coding.cftime_offsets.to_offset(freq)
+                if isinstance(start, pd.Timestamp):
+                    start = start - pd.tseries.frequencies.to_offset(freq)
+                else:
+                    start = start - xr.coding.cftime_offsets.to_offset(freq)
                 # In the diff below, assign to upper label!
                 label = "upper"
             # We generate "time" with an extra element, so we do not need to repeat the last element below.

diff --git a/xclim/indices/helpers.py b/xclim/indices/helpers.py
@@ -16,9 +16,15 @@
 import numba as nb
 import numpy as np
 import xarray as xr
-from xarray.coding.calendar_ops import (
-    _datetime_to_decimal_year as datetime_to_decimal_year,
-)
+
+try:
+    from xarray.coding.calendar_ops import (
+        _datetime_to_decimal_year as datetime_to_decimal_year,
+    )
+except ImportError:
+    XR2409 = True
+else:
+    XR2409 = False
 
 from xclim.core import Quantified
 from xclim.core.calendar import ensure_cftime_array, get_calendar
@@ -70,7 +76,10 @@ def day_angle(time: xr.DataArray):
     the beginning of the year up to that timestep. Also called the "julian day fraction".
     See :py:func:`~xclim.core.calendar.datetime_to_decimal_year`.
     """
-    decimal_year = datetime_to_decimal_year(times=time, calendar=time.dt.calendar)
+    if XR2409:
+        decimal_year = time.dt.decimal_year
+    else:
+        decimal_year = datetime_to_decimal_year(times=time, calendar=time.dt.calendar)
     return ((decimal_year % 1) * 2 * np.pi).assign_attrs(units="rad")