Merge pull request #98 from rubisco-sfa/update-gfed

FIX: GFED and other problems reading in multifile datsets

Author: nocollier

ilamb3/analysis/area.py

@@ -190,7 +190,6 @@ def __call__(
                 )
             ]
         )
-        print(df)
         return df, ref_.rename({varname: "extent"}), com_.rename({varname: "extent"})
 
     def plots(

ilamb3/cli/__init__.py

@@ -89,7 +89,7 @@ def run(
     config: Path,
     regions: str | None = None,
     region_sources: list[str] | None = None,
-    df_comparison: Path | None = None,
+    df_comparison: list[Path] | None = None,
     output_path: Path = Path("_build"),
     cache: bool = True,
     central_longitude: float = 0.0,
@@ -121,7 +121,8 @@ def run(
     if df_comparison is None:
         df_com = _dataframe_cmip()
     else:
-        df_com = pd.read_csv(df_comparison)
+        df_comparison = [Path(f) for f in str(df_comparison[0]).split(",")]
+        df_com = pd.concat([pd.read_csv(f) for f in df_comparison])
 
     # execute
     if HAS_MPI4PY:

ilamb3/configure/ilamb.yaml

@@ -53,12 +53,10 @@ Ecosystem and Carbon Cycle:
 
   Burnt Area:
 
-    GFED4.1S:
+    GFED-5-0:
       sources:
-        burntArea: burntArea/GFED4.1S/burntArea.nc
+        burntFractionAll: GFED-5-0/obs4MIPs_NASA-NOSR_GFED-5-0_mon_burntFractionAll_gr_v20251029.nc
       variable_cmap: YlOrBr
-      alternate_vars:
-      - burntFractionAll
 
   Gross Primary Productivity:
 
@@ -248,7 +246,7 @@ Hydrology Cycle:
         hfss: hfss/FLUXCOM/hfss.nc
       variable_cmap: Blues
 
-    WECANN:
+    WECANN-1-0:
       sources:
         hfss: WECANN-1-0/obs4MIPs_ColumbiaU_WECANN-1-0_mon_hfss_gn_v20250902.nc
       variable_cmap: Blues

ilamb3/dataset.py

@@ -729,10 +729,13 @@ def _get_interval(dset, dim, value, side):  # numpydoc ignore=GL08
         if "bounds" in coord.attrs:
             if coord.attrs["bounds"] in dset:
                 coord = dset[coord.attrs["bounds"]]
+                coord.load()
                 ind = ((coord[:, 0] <= value) & (coord[:, 1] >= value)).to_numpy()
                 ind = np.where(ind)[0]
-                assert len(ind) <= 2
-                assert len(ind) > 0
+                if len(ind) not in [1, 2]:
+                    raise ValueError(
+                        f"We could not find an appropriate interval for {value} in \n{coord}"
+                    )
                 if len(ind) == 2:
                     if side == "low":
                         return ind[1]

ilamb3/registry/ilamb.txt

@@ -6,7 +6,6 @@ biomass/Saatchi2011/biomass_0.5x0.5.nc sha1:af0dad922bc2d174293a22ab374b8662220f
 biomass/Thurner/biomass_0.5x0.5.nc sha1:d76bb837865b8bb4ed42ddd04a20454ed68db9bb
 biomass/USForest/biomass_0.5x0.5.nc sha1:cde74c443271f3e98d7d59e2c827e686a5abe95f
 biomass/XuSaatchi2021/XuSaatchi.nc sha1:244543e731dfd46f9842d6e40f4a2eeead122ee3
-burntArea/GFED4.1S/burntArea.nc sha1:cf9d73c6a8bfc594737c9ba6ca4df613df4a28ab
 cSoil/HWSD/soilc_0.5x0.5.nc sha1:5dd084a66a7e0d0b4ca6125db47af0eb0c119377
 cSoil/HWSD2/hwsd2_cSoil.nc sha1:9a6377e4c5ff457c08c194d2c376c46e003a4f84
 cSoil/Koven/fracpeat_0.5x0.5.nc sha1:fa7cab1379c66baf7c1977d3c4c676ee9678de49

ilamb3/registry/ilamb3.txt

@@ -5,6 +5,7 @@ WECANN-1-0/obs4MIPs_ColumbiaU_WECANN-1-0_mon_hfls_gn_v20250902.nc sha1:380cea38e
 Hoffman-1-0/obs4MIPs_UCI-ORNL_Hoffman-1-0_yr_fgco2_gm_v20250903.nc sha1:f06d909348c05fdea0a7af87b01bfd4cb440e075
 Hoffman-1-0/obs4MIPs_UCI-ORNL_Hoffman-1-0_yr_nbp_gm_v20250903.nc sha1:803453d7b3873732190b1a20b25c4b50d34642c9
 HWSD-2-0/obs4MIPs_IIASA-FAO_HWSD-2-0_fx_cSoil_gn_v20250903.nc sha1:9855622a525902f0e2625ee6293e8d7c2b18c24c
+GFED-5-0/obs4MIPs_NASA-NOSR_GFED-5-0_mon_burntFractionAll_gr_v20251029.nc sha1:0527f07ae72ea9520486f7f3d47f8ec26a7a017f
 RAPID-2023-1a/obs4MIPs_NOC_RAPID-2023-1a_mon_msftmz_gm_v20250902.nc sha1:0dfeea11872bdc453cf5d0a3b0cf1abf4d14369f
 WOA-23/obs4MIPs_NOAA-NCEI-OCL_WOA-23_monC_si_gn_v20250902.nc sha1:a51b67b0eb874ddd40ecb25cf9c087005186b74f
 WOA-23/obs4MIPs_NOAA-NCEI-OCL_WOA-23_monC_thetao_gn_v20250902.nc sha1:a5466f4b53b4d3a840b9a372be2cd4be8a0e5280

ilamb3/run.py

@@ -357,7 +357,8 @@ def _load_comparison_data(
         var: xr.open_mfdataset(
             sorted((df[df["variable_id"] == var]["path"]).to_list()),
             preprocess=fix_lndgrid_coords,
-            data_vars="all",
+            data_vars=None,
+            compat="no_conflicts",
         )
         for var in df["variable_id"].unique()
     }

ilamb3/transform/select.py

@@ -12,18 +12,18 @@
 
 class select_dim(ILAMBTransform):
     """
-    Select a depth if the dimension exists.
+    Select a value or slice in the dimension if the dimension exists.
 
     Parameters
     ----------
     dim : str
         The dimension.
     value : float
-        The value at which to select the nearest depth.
+        The value at which to select the nearest dimension.
     vmin : float
-        The minimum depth to slice.
+        The minimum value to slice.
     vmax : float
-        The maximum depth to slice.
+        The maximum value to slice.
 
     Note
     ----