Refactor zonal_mean_xy set

auxiliary_tools/cdat_regression_testing/654-zonal_mean_xy/654-zonal_mean_xy_run_script.py

@@ -0,0 +1,6 @@
+from auxiliary_tools.cdat_regression_testing.base_run_script import run_set
+
+SET_NAME = "zonal_mean_xy"
+SET_DIR = "654-zonal_mean_xy"
+
+run_set(SET_NAME, SET_DIR)

auxiliary_tools/cdat_regression_testing/template_cdat_regression_test_netcdf.ipynb

@@ -20,7 +20,7 @@
     "(dev and `main` branches).\n",
     "\n",
     "1. Make a copy of this notebook under `auxiliary_tools/cdat_regression_testing/<DIR_NAME>`.\n",
-    "2. Run `mamba create -n cdat_regression_test -y -c conda-forge \"python<3.12\" xarray dask pandas matplotlib-base ipykernel`\n",
+    "2. Run `mamba create -n cdat_regression_test -y -c conda-forge \"python<3.12\" xarray netcdf4 dask pandas matplotlib-base ipykernel`\n",
     "3. Run `mamba activate cdat_regression_test`\n",
     "4. Update `SET_DIR` and `SET_NAME` in the copy of your notebook.\n",
     "5. Run all cells IN ORDER.\n",

auxiliary_tools/cdat_regression_testing/template_run_script.py

@@ -12,11 +12,13 @@
      "meridional_mean_2d", "annual_cycle_zonal_mean", "enso_diags", "qbo",
      "area_mean_time_series", "diurnal_cycle", "streamflow", "arm_diags",
      "tc_analysis", "aerosol_aeronet", "aerosol_budget", "mp_partition",
-6. Run this script
-   - Make sure to run this command on NERSC perlmutter cpu:
-    `salloc --nodes 1 --qos interactive --time 01:00:00 --constraint cpu --account=e3sm
-    conda activate <NAME-OF-DEV-ENV>`
-   - python auxiliary_tools/cdat_regression_testing/<ISSUE-<SET_NAME>
+
+6. Run this script as a Python module
+   - `auxiliary_tools` is not included in `setup.py`, so `-m` is required
+     to run the script as a Python module
+   - Command: python -m auxiliary_tools.cdat_regression_testing.<ISSUE>-<SET_NAME>.<SCRIPT-NAME>
+   - Example: python -m auxiliary_tools.cdat_regression_testing.660_cosp_histogram.run_script
+
 7. Make a copy of the CDAT regression testing notebook in the same directory
    as this script and follow the instructions there to start testing.
 """

e3sm_diags/derivations/derivations.py

@@ -87,6 +87,7 @@
             ),
             (("pr",), lambda pr: qflxconvert_units(rename(pr))),
             (("PRECC", "PRECL"), lambda precc, precl: prect(precc, precl)),
+            (("sat_gauge_precip",), rename),
         ]
     ),
     "PRECST": OrderedDict(
@@ -767,7 +768,7 @@
                     "QFLX",
                 ),
                 lambda precc, precl, qflx: pminuse_convert_units(
-                    prect(precc, precl) - pminuse_convert_units(qflx)
+                    prect(precc, precl) - qflxconvert_units(qflx)
                 ),
             ),
             (

e3sm_diags/derivations/formulas.py

@@ -77,15 +77,16 @@ def qflx_convert_to_lhflx_approxi(var: xr.DataArray):
 
 
 def pminuse_convert_units(var: xr.DataArray):
-    if (
-        var.attrs["units"] == "kg/m2/s"
-        or var.attrs["units"] == "kg m-2 s-1"
-        or var.attrs["units"] == "kg/s/m^2"
-    ):
-        # need to find a solution for units not included in udunits
-        # var = convert_units( var, 'kg/m2/s' )
-        var = var * 3600.0 * 24  # convert to mm/day
-        var.attrs["units"] = "mm/day"
+    if hasattr(var, "units"):
+        if (
+            var.attrs["units"] == "kg/m2/s"
+            or var.attrs["units"] == "kg m-2 s-1"
+            or var.attrs["units"] == "kg/s/m^2"
+        ):
+            # need to find a solution for units not included in udunits
+            # var = convert_units( var, 'kg/m2/s' )
+            var = var * 3600.0 * 24  # convert to mm/day
+    var.attrs["units"] = "mm/day"
     var.attrs["long_name"] = "precip. flux - evap. flux"
     return var
 

e3sm_diags/driver/default_diags/zonal_mean_xy_model_vs_obs.cfg

@@ -3,7 +3,7 @@ sets = ["zonal_mean_xy"]
 case_id = "GPCP_v3.2"
 variables = ["PRECT"]
 ref_name = "GPCP_v3.2"
-reference_name = "GPCP v2.2"
+reference_name = "GPCP v3.2"
 seasons = ["ANN", "DJF", "MAM", "JJA", "SON"]
 regions = ["global"]
 

e3sm_diags/driver/utils/dataset_xr.py

@@ -294,7 +294,8 @@ def get_ref_climo_dataset(
         # TODO: This logic was carried over from legacy implementation. It
         # can probably be improved on by setting `ds_ref = None` and not
         # performing unnecessary operations on `ds_ref` for model-only runs,
-        # since it is the same as `ds_test``.
+        # since it is the same as `ds_test``. In addition, returning ds_test makes it difficult for trouble shooting
+
         if self.data_type == "ref":
             try:
                 ds_ref = self.get_climo_dataset(var_key, season)
@@ -304,6 +305,7 @@ def get_ref_climo_dataset(
                 self.model_only = True
 
                 logger.info("Cannot process reference data, analyzing test data only.")
+
         else:
             raise RuntimeError(
                 "`Dataset._get_ref_dataset` only works with "
@@ -396,6 +398,9 @@ def _get_climo_dataset(self, season: str) -> xr.Dataset:
             )
 
         ds = self._squeeze_time_dim(ds)
+        # slat and slon are lat lon pair for staggered FV grid included in remapped files
+        if "slat" in ds.dims:
+            ds = ds.drop_dims(["slat", "slon"])
 
         return ds
 

e3sm_diags/driver/utils/io.py

@@ -66,9 +66,7 @@ def _save_data_metrics_and_plots(
     logger.info(f"Metrics saved in {filepath}")
 
     # Set the viewer description to the "long_name" attr of the variable.
-    parameter.viewer_descr[var_key] = ds_test[var_key].attrs.get(
-        "long_name", "No long_name attr in test data"
-    )
+    parameter.viewer_descr[var_key] = ds_test[var_key].attrs.get("long_name", var_key)
 
     # Get the function arguments and pass to the set's plotting function.
     args = [