Update IBTrACS to version 4.1 (#976)

* Update IBTrACS version

* Update CHANGELOG

* Update tc_tracks tests for new IBTrACS version

* Fix deprecation notice for datasets.dims -> datasets.sizes

* Fix pool problem for concatenating hazards after pool computations

* Update changelog

* Apply black

* Update CHANGELOG.md

---------

Co-authored-by: Chahan Kropf <[email protected]>
Co-authored-by: emanuel-schmid <[email protected]>
Co-authored-by: Chahan M. Kropf <[email protected]>

Author: 4 people

CHANGELOG.md

@@ -37,6 +37,9 @@ Code freeze date: YYYY-MM-DD
   - latitude and longitude column are no longer present there (the according arrays can be retrieved as properties of the Exposures object: `exp.latitude` instead of `exp.gdf.latitude.values`).
   - `Exposures.gdf` has been renamed to `Exposures.data` (it still works though, as it is a property now pointing to the latter)
   - the `check` method does not add a default "IMPF_" column to the GeoDataFrame anymore
+- Updated IBTrACS version from v4.0 to v4.1 ([#976](https://github.com/CLIMADA-project/climada_python/pull/976)
+    - Fix xarray future warning in TCTracks for .dims to .sizes
+    - Fix hazard.concatenate type test for pathos pools
 
 ### Fixed
 

climada/hazard/base.py

@@ -936,8 +936,7 @@ def append(self, *others):
                 "The hazards are incompatible and cannot be concatenated."
             )
         self.haz_type = haz_types.pop()
-
-        haz_classes = {type(haz) for haz in haz_list}
+        haz_classes = {haz.__class__.__name__ for haz in haz_list}
         if len(haz_classes) > 1:
             raise TypeError(
                 f"The given hazards are of different classes: {haz_classes}. "

climada/hazard/tc_tracks.py

@@ -84,13 +84,13 @@
 IBTRACS_URL = (
     "https://www.ncei.noaa.gov/data/"
     "international-best-track-archive-for-climate-stewardship-ibtracs/"
-    "v04r00/access/netcdf"
+    "v04r01/access/netcdf"
 )
 """Site of IBTrACS netcdf file containing all tracks v4.0,
 s. https://www.ncdc.noaa.gov/ibtracs/index.php?name=ib-v4-access"""
 
-IBTRACS_FILE = "IBTrACS.ALL.v04r00.nc"
-"""IBTrACS v4.0 file all"""
+IBTRACS_FILE = "IBTrACS.ALL.v04r01.nc"
+"""IBTrACS v4.1 file all"""
 
 IBTRACS_AGENCIES = [
     "usa",
@@ -376,7 +376,7 @@ def from_ibtracs_netcdf(
         correct_pres=False,
         discard_single_points=True,
         additional_variables=None,
-        file_name="IBTrACS.ALL.v04r00.nc",
+        file_name=IBTRACS_FILE,
     ):
         """Create new TCTracks object from IBTrACS databse.
 
@@ -485,7 +485,7 @@ def from_ibtracs_netcdf(
             compatiblity with other functions such as `equal_timesteps`. Default: True.
         file_name : str, optional
             Name of NetCDF file to be dowloaded or located at climada/data/system.
-            Default: 'IBTrACS.ALL.v04r00.nc'
+            Default: 'IBTrACS.ALL.v04r01.nc'
         additional_variables : list of str, optional
             If specified, additional IBTrACS data variables are extracted, such as "nature" or
             "storm_speed". Only variables that are not agency-specific are supported.
@@ -731,7 +731,7 @@ def from_ibtracs_netcdf(
                     )
                     ibtracs_ds = ibtracs_ds.sel(storm=valid_storms_mask)
 
-            if ibtracs_ds.dims["storm"] == 0:
+            if ibtracs_ds.sizes["storm"] == 0:
                 LOGGER.info(
                     "After discarding IBTrACS events without valid values by the selected "
                     "reporting agencies, there are no tracks left that match the specified "
@@ -2576,7 +2576,7 @@ def ibtracs_fit_param(explained, explanatory, year_range=(1980, 2019), order=1):
             raise KeyError("Unknown ibtracs variable: %s" % var)
 
     # load ibtracs dataset
-    fn_nc = SYSTEM_DIR.joinpath("IBTrACS.ALL.v04r00.nc")
+    fn_nc = SYSTEM_DIR.joinpath(IBTRACS_FILE)
     with xr.open_dataset(fn_nc) as ibtracs_ds:
         # choose specified year range
         years = ibtracs_ds.sid.str.slice(0, 4).astype(int)

climada/hazard/test/__init__.py

@@ -28,7 +28,7 @@
 
 
 def download_ibtracs():
-    """This makes sure a IBTrACS.ALL.v04r00.nc file is present in SYSTEM_DIR
+    """This makes sure a IBTrACS.ALL.v04r01.nc file is present in SYSTEM_DIR
     First, downloading from the original sources is attempted. If that fails an old version
     is downloaded from the CLIMADA Data API
     """
@@ -44,9 +44,9 @@ def download_ibtracs():
     ):  # plan b: download an old version of that file from the climada api
         client = Client()
         dsinfo = client.get_dataset_info(
-            name="IBTrACS", version="v04r00", status="external"
+            name="IBTrACS", version="v04r01", status="external"
         )
         [fileinfo] = [
-            fi for fi in dsinfo.files if fi.file_name == "IBTrACS.ALL.v04r00.nc"
+            fi for fi in dsinfo.files if fi.file_name == "IBTrACS.ALL.v04r01.nc"
         ]
         client._download_file(local_path=SYSTEM_DIR, fileinfo=fileinfo)

climada/hazard/test/test_tc_tracks.py

@@ -82,13 +82,35 @@ def test_penv_rmax_penv_pass(self):
             provider="usa", storm_id="1992230N11325"
         )
         penv_ref = np.ones(97) * 1010
-        penv_ref[26:36] = [1011, 1012, 1013, 1014, 1015, 1014, 1014, 1014, 1014, 1012]
+        penv_ref[26:36] = [1011, 1012, 1013, 1014, 1015, 1015, 1014, 1014, 1014, 1012]
+        rmax_ref = np.zeros(97)
+        rmax_ref[63:82] = [
+            10.0,
+            10.0,
+            10.625,
+            11.25,
+            11.875,
+            12.5,
+            13.125,
+            13.75,
+            14.375,
+            15.0,
+            15.625,
+            16.25,
+            16.875,
+            17.5,
+            18.125,
+            18.75,
+            19.375,
+            20.0,
+            20.0,
+        ]
 
-        self.assertTrue(
-            np.allclose(tc_track.get_track()["environmental_pressure"].values, penv_ref)
+        np.testing.assert_array_almost_equal(
+            tc_track.get_track()["environmental_pressure"].values, penv_ref, decimal=4
         )
-        self.assertTrue(
-            np.allclose(tc_track.get_track()["radius_max_wind"].values, np.zeros(97))
+        np.testing.assert_array_almost_equal(
+            tc_track.get_track()["radius_max_wind"].values, rmax_ref, decimal=4
         )
 
     def test_ibtracs_raw_pass(self):
@@ -278,7 +300,7 @@ def test_ibtracs_correct_pass(self):
             tc_try.data[0]["central_pressure"].values[0], 1013, places=0
         )
         self.assertAlmostEqual(
-            tc_try.data[0]["central_pressure"].values[5], 1008, places=0
+            tc_try.data[0]["central_pressure"].values[5], 1007, places=0
         )
         self.assertAlmostEqual(
             tc_try.data[0]["central_pressure"].values[-1], 1012, places=0
@@ -701,14 +723,14 @@ def test_get_extent(self):
         tc_track = tc.TCTracks.from_ibtracs_netcdf(
             storm_id=storms, provider=["usa", "bom"]
         )
-        bounds = (153.585022, -23.200001, 258.714996, 17.514986)
-        bounds_buf = (153.485022, -23.300001, 258.814996, 17.614986)
-        np.testing.assert_array_almost_equal(tc_track.bounds, bounds)
+        bounds = (153.6, -23.2, 258.7, 17.5)
+        bounds_buf = (153.5, -23.3, 258.8, 17.6)
+        np.testing.assert_array_almost_equal(tc_track.bounds, bounds, decimal=4)
         np.testing.assert_array_almost_equal(
-            tc_track.get_bounds(deg_buffer=0.1), bounds_buf
+            tc_track.get_bounds(deg_buffer=0.1), bounds_buf, decimal=4
         )
         np.testing.assert_array_almost_equal(
-            tc_track.extent, u_coord.toggle_extent_bounds(bounds)
+            tc_track.extent, u_coord.toggle_extent_bounds(bounds), decimal=4
         )
 
     def test_generate_centroids(self):
@@ -718,13 +740,13 @@ def test_generate_centroids(self):
             storm_id=storms, provider=["usa", "bom"]
         )
         cen = tc_track.generate_centroids(10, 1)
-        cen_bounds = (157.585022, -19.200001, 257.585022, 10.799999)
+        cen_bounds = (157.6, -19.2, 257.6, 10.8)
         self.assertEqual(cen.size, 44)
         self.assertEqual(np.unique(cen.lat).size, 4)
         self.assertEqual(np.unique(cen.lon).size, 11)
         np.testing.assert_array_equal(np.diff(np.unique(cen.lat)), 10)
         np.testing.assert_array_equal(np.diff(np.unique(cen.lon)), 10)
-        np.testing.assert_array_almost_equal(cen.total_bounds, cen_bounds)
+        np.testing.assert_array_almost_equal(cen.total_bounds, cen_bounds, decimal=4)
 
     def test_interp_track_pass(self):
         """Interpolate track to min_time_step. Compare to MATLAB reference."""

doc/tutorial/climada_hazard_TropCyclone.ipynb

@@ -64,9 +64,9 @@
     "<a id='Part1.a'></a> \n",
     "### a) Load TC tracks from historical records\n",
     "\n",
-    "The best-track historical data from the International Best Track Archive for Climate Stewardship ([IBTrACS](https://www.ncdc.noaa.gov/ibtracs/)) can easily be loaded into CLIMADA to study the historical records of TC events. The constructor `from_ibtracs_netcdf()` generates the `Datasets` for tracks selected by [IBTrACS](https://www.ncdc.noaa.gov/ibtracs/) id, or by basin and year range. To achieve this, it downloads the first time the [IBTrACS data v4 in netcdf format](https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r00/access/netcdf/) and stores it in `~/climada/data/`. The tracks can be accessed later either using the attribute `data` or using `get_track()`, which allows to select tracks by its name or id. Use the method `append()` to extend the `data` list.\n",
+    "The best-track historical data from the International Best Track Archive for Climate Stewardship ([IBTrACS](https://www.ncdc.noaa.gov/ibtracs/)) can easily be loaded into CLIMADA to study the historical records of TC events. The constructor `from_ibtracs_netcdf()` generates the `Datasets` for tracks selected by [IBTrACS](https://www.ncdc.noaa.gov/ibtracs/) id, or by basin and year range. To achieve this, it downloads the first time the [IBTrACS data v4 in netcdf format](https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r01/access/netcdf/) and stores it in `~/climada/data/`. The tracks can be accessed later either using the attribute `data` or using `get_track()`, which allows to select tracks by its name or id. Use the method `append()` to extend the `data` list.\n",
     "\n",
-    "If you get an error downloading the IBTrACS data, try to manually access [https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r00/access/netcdf/](https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r00/access/netcdf/), click on the file `IBTrACS.ALL.v04r00.nc` and copy it to `~/climada/data/`.\n",
+    "If you get an error downloading the IBTrACS data, try to manually access [https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r01/access/netcdf/](https://www.ncei.noaa.gov/data/international-best-track-archive-for-climate-stewardship-ibtracs/v04r01/access/netcdf/), click on the file `IBTrACS.ALL.v04r01.nc` and copy it to `~/climada/data/`.\n",
     "\n",
     "To visualize the tracks use `plot()`.\n"
    ]