on-the-fly cmoriser for ACCESS native data

.zenodo.json

@@ -204,6 +204,11 @@
         {
             "affiliation": "DLR, Germany",
             "name": "Cammarano, Diego"
+        },
+        {
+            "affiliation": "ACCESS-NRI, Australia",
+            "name": "Yousong, Zeng",
+            "orcid": "0000-0002-8385-5367"
         }
     ],
     "description": "ESMValCore: A community tool for pre-processing data from Earth system models in CMIP and running analysis scripts.",

CITATION.cff

@@ -209,6 +209,11 @@ authors:
     family-names: Cammarano
     given-names: Diego
 
+    affiliation: "ACCESS-NRI, Australia"
+    family-names: Yousong
+    given-names: Zeng
+    orcid: "https://orcid.org/0000-0002-8385-5367"
+
 cff-version: 1.2.0
 date-released: 2024-05-08
 doi: "10.5281/zenodo.3387139"

doc/quickstart/find_data.rst

@@ -564,6 +564,64 @@ explained in :ref:`extra_facets`, and which content is :download:`available here
 </../esmvalcore/config/extra_facets/ipslcm-mappings.yml>`. These multi-variable
 files must also undergo some data selection.
 
+.. _read_access-esm:
+
+ACCESS-ESM
+^^^^^^^^^^
+
+ESMValTool can read native `ACCESS-ESM <https://research.csiro.au/access/about/esm1-5/>`__
+model output.
+
+.. warning::
+
+  This is the first version of ACCESS-ESM CMORizer for ESMValCore. Currently, 
+  Supported variables: ``pr``, ``ps``, ``psl``, ``rlds``, ``tas``, ``ta``, ``va``,
+  ``ua``, ``zg``, ``hus``, ``clt``, ``rsus``, ``rlus``.
+
+The default naming conventions for input directories and files for ACCESS output are
+
+* input directories: ``{institute}/{sub_dataset}/{exp}/{modeling_realm}/netCDF``
+* input files: ``{sub_dataset}.{special_attr}-*.nc``
+
+.. hint::
+
+  We only provide one default `input_dir` since this is how ACCESS-ESM native data was 
+  stored on NCI. Users can modify this path in the :ref:`config-developer` to match their local file structure.
+
+
+Thus, example dataset entries could look like this:
+
+.. code-block:: yaml
+
+  dataset:
+    - {project: ACCESS, mip: Amon, dataset:ACCESS_ESM1_5, sub_dataset: HI-CN-05, 
+      exp: history, modeling_realm: atm, special_attr: pa, start_year: 1986, end_year: 1986}
+
+
+Similar to any other fix, the ACCESS-ESM fix allows the use of :ref:`extra
+facets<extra_facets>`.
+By default, the file :download:`access-mappings.yml
+</../esmvalcore/config/extra_facets/access-mappings.yml>` is used for that
+purpose.
+For some variables, extra facets are necessary; otherwise ESMValCore cannot
+read them properly.
+Supported keys for extra facets are:
+
+==================== ====================================== =================================
+Key                  Description                            Default value if not specified
+==================== ====================================== =================================
+``raw_name``         Variable name of the variable in the   CMOR variable name of the
+                     raw input file                         corresponding variable
+``modeling_realm``   Realm attribute include `atm`, `ice`   No default (needs to be 
+                     and `oce`                              specified in extra facets or
+                                                            recipe if default DRS is used)
+```special_attr``    A special attribute in the filename    No default
+                     `ACCESS-ESM` raw data, it's related to 
+                     frquency of raw data
+``sub_dataset``      Part of the ACCESS-ESM raw dataset     No default
+                     root, need to specify if you want to
+                     use the cmoriser                                                                       
+==================== ====================================== =================================
 
 .. _data-retrieval:
 

esmvalcore/cmor/_fixes/access/access_esm1_5.py

@@ -0,0 +1,75 @@
+"""On-the-fly CMORizer for ACCESS-ESM."""
+import logging
+
+from iris.cube import CubeList
+
+from esmvalcore.cmor._fixes.native_datasets import NativeDatasetFix
+
+logger = logging.getLogger(__name__)
+
+
+class AllVars(NativeDatasetFix):
+    """Fixes for all variables."""
+
+    def fix_coord_system(self, cube):
+        """Delete coord_system to make CubeList able to merge."""
+        for dim in cube.dim_coords:
+            if dim.coord_system is not None:
+                dim.coord_system = None
+
+    def fix_height_value(self, cube):
+        """Fix height value to make it comparable to other dataset."""
+        if cube.coord('height').points[0] != 2:
+            cube.coord('height').points = [2]
+
+    def calculate_data_from_multivar(self, cubes):
+        """Get cube before calculate from multiple variables."""
+        rawname_list = self.extra_facets.get('raw_name',
+                                             self.vardef.short_name)
+        var = []
+        for rawname in rawname_list:
+            var.append(self.get_cube(cubes, rawname))
+        if self.vardef.short_name == 'rsus':
+            cube = var[0] - var[1]
+        if self.vardef.short_name == 'rlus':
+            cube = var[0] - var[1] + var[2] - var[3]
+        return cube
+
+    def fix_metadata(self, cubes):
+        """Fix metadata.
+
+        Parameters
+        ----------
+        cubes : iris.cube.CubeList
+            Input cubes.
+
+        Returns
+        -------
+        iris.cube.CubeList
+        """
+        if isinstance(
+                self.extra_facets.get('raw_name', self.vardef.short_name),
+                list):
+            cube = self.calculate_data_from_multivar(cubes)
+        else:
+            cube = self.get_cube(cubes)
+
+        # Fix coordinates
+        self.fix_scalar_coords(cube)
+        self.fix_var_metadata(cube)
+        self.fix_lon_metadata(cube)
+        self.fix_lat_metadata(cube)
+
+        # Fix coordinate 'height'
+        if 'height_0' in [var.var_name for var in cube.coords()]:
+            self.fix_height_metadata(cube)
+            if self.vardef.short_name == 'tas':
+                self.fix_height_value(cube)
+        # Fix coordinate 'pressure'
+        if 'pressure' in [var.var_name for var in cube.coords()]:
+            self.fix_plev_metadata(cube, coord='pressure')
+
+        # Fix coord system
+        self.fix_coord_system(cube)
+
+        return CubeList([cube])

esmvalcore/config-developer.yml

@@ -194,3 +194,14 @@ CESM:
   output_file: '{project}_{dataset}_{case}_{gcomp}_{scomp}_{type}_{mip}_{short_name}'
   cmor_type: 'CMIP6'
   cmor_default_table_prefix: 'CMIP6_'
+
+ACCESS:
+  cmor_strict: false
+  input_dir:
+    default:
+      - '{dataset}/{sub_dataset}/{exp}/{modeling_realm}/netCDF'
+  input_file:
+    default: '{sub_dataset}.{special_attr}-*.nc'
+  output_file: '{project}_{dataset}_{mip}_{exp}_{institute}_{sub_dataset}_{special_attr}_{short_name}'
+  cmor_type: 'CMIP6'
+  cmor_default_table_prefix: 'CMIP6_'

esmvalcore/config/extra_facets/access-mappings.yml

@@ -0,0 +1,68 @@
+# Extra facets for native ACCESS model output
+
+# A complete list of supported keys is given in the documentation (see
+# ESMValCore/doc/quickstart/find_data.rst).
+---
+
+ACCESS_ESM:
+
+  '*':
+
+    tas:
+      raw_name: fld_s03i236
+      modeling_realm: atm
+
+    pr:
+      raw_name: fld_s05i216
+      modeling_realm: atm
+
+    ps:
+      raw_name: fld_s00i409
+      modeling_realm: atm
+
+    clt:
+      raw_name: fld_s02i204
+      modeling_realm: atm
+
+    psl:
+      raw_name: fld_s16i222
+      modeling_realm: atm
+
+    hus:
+      raw_name: fld_s30i205
+      modeling_realm: atm
+
+    zg:
+      raw_name: fld_s30i207
+      modeling_realm: atm
+
+    va:
+      raw_name: fld_s30i202
+      modeling_realm: atm
+
+    ua:
+      raw_name: fld_s30i201
+      modeling_realm: atm
+
+    ta:
+      raw_name: fld_s30i204
+      modeling_realm: atm
+
+    rlus:
+      raw_name: 
+        - fld_s02i207 
+        - fld_s02i201 
+        - fld_s03i332 
+        - fld_s02i205
+      modeling_realm: atm
+
+    rlds:
+      raw_name: fld_s02i207
+      modeling_realm: atm
+
+    rsus:
+      raw_name: 
+        - fld_s01i235
+        - fld_s01i201 
+      modeling_realm: atm
+