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 +564,56 @@ 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, 
+  only two variables (`tas`, and `pr`) are fully supported.
+
+To read ACCESS-ESM native data, users need to add the directory file structure to 
+`config_developer.yml`.
+
+.. code-block:: yaml
+
+  ACCESS:
+    cmor_strict: false
+    input_dir:
+      default:
+        - '{institute}/{dataset_name}/{exp}/{modeling_realm}/netCDF'
+    input_file:
+      default: '{dataset}.{special_attr}-*.nc'
+    output_file: '{dataset}.{special_attr}'
+    cmor_type: 'CMIP6'
+    cmor_default_table_prefix: 'CMIP6_'
+
+.. 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 to match their local file structure.
+
+  Default `input_file` will automatically load all the native ACCESS-ESM data(1850-2014).
+  If you only need to evaluate a shorter period. Users can add a more specific time range
+  (`{dataset}.{special_attr}-YYYYMM*.nc` e.g.) to avoid the cost of loading unnecessary data.
+
+Thus, example dataset entries could look like this:
+
+.. code-block:: yaml
+
+  dataset:
+    - {project: ACCESS, institute: ACCESS-ESM1-5, mip: Amon, dataset:ACCESS_ESM, dataset_name: HI-CN-05, 
+      exp: history, modeling_realm: atm, special_attr: pa, start_year: 1986, end_year: 1986}
+
+
+`dataset` and `dataset_name` are not redundant, `dataset` is for ESMValCore to search for CMORizer,
+`dataset_name` is to search for data.
+
 
 .. _data-retrieval:
 

esmvalcore/cmor/_fixes/access/access_esm.py

@@ -0,0 +1,128 @@
+"""On-the-fly CMORizer for ACCESS-ESM.
+
+Note
+----
+This is the first version of ACCESS-ESM CMORizer in for ESMValCore
+Currently, only two variables (`tas`,`pr`) is fully supported.
+"""
+import logging
+import os
+
+from iris.cube import CubeList
+
+from esmvalcore.cmor._fixes.native_datasets import NativeDatasetFix
+
+logger = logging.getLogger(__name__)
+
+
+class Tas(NativeDatasetFix):
+    """Fix variable(tas) only."""
+
+    def __init__(self, vardef, extra_facets, session, frequency):
+        """Initialise some class variable Heritage from native_dataset."""
+        super().__init__(vardef, extra_facets, session, frequency)
+
+        self.cube = None
+
+        self.current_dir = os.path.dirname(__file__)
+
+    def fix_height_name(self):
+        """Fix variable name of coordinate 'height'."""
+        if self.cube.coord('height').var_name != 'height':
+            self.cube.coord('height').var_name = 'height'
+
+    def fix_long_name(self):
+        """Fix variable long_name."""
+        self.cube.long_name = 'Near-Surface Air Temperature'
+
+    def fix_var_name(self):
+        """Fix variable long_name."""
+        self.cube.var_name = 'tas'
+
+    def fix_coord_system(self):
+        """Delete coord_system to make CubeList able to merge."""
+        for dim in self.cube.dim_coords:
+            if dim.coord_system is not None:
+                self.cube.coord(dim.standard_name).coord_system = None
+
+    def fix_metadata(self, cubes):
+        """Fix metadata.
+
+        Fix name of coordinate(height), long name and variable name of
+        variable(tas).
+
+        Parameters
+        ----------
+        cubes : iris.cube.CubeList
+            Input cubes.
+
+        Returns
+        -------
+        iris.cube.CubeList
+        """
+        original_short_name = 'fld_s03i236'
+
+        self.cube = self.get_cube(cubes, var_name=original_short_name)
+
+        self.fix_height_name()
+
+        self.fix_long_name()
+
+        self.fix_var_name()
+
+        self.fix_coord_system()
+
+        return CubeList([self.cube])
+
+
+class Pr(NativeDatasetFix):
+    """Fix variable(pr) only."""
+
+    def __init__(self, vardef, extra_facets, session, frequency):
+        """Initialise some class variable Heritage from native_dataset."""
+        super().__init__(vardef, extra_facets, session, frequency)
+
+        self.cube = None
+
+        self.current_dir = os.path.dirname(__file__)
+
+    def fix_var_name(self):
+        """Fix variable long_name."""
+        self.cube.var_name = 'pr'
+
+    def fix_long_name(self):
+        """Fix variable long_name."""
+        self.cube.long_name = 'Precipitation'
+
+    def fix_coord_system(self):
+        """Delete coord_system to make CubeList able to merge."""
+        for dim in self.cube.dim_coords:
+            if dim.coord_system is not None:
+                self.cube.coord(dim.standard_name).coord_system = None
+
+    def fix_metadata(self, cubes):
+        """Fix metadata.
+
+        Fix name of coordinate(height), long name and variable name of
+        variable(tas).
+
+        Parameters
+        ----------
+        cubes : iris.cube.CubeList
+            Input cubes.
+
+        Returns
+        -------
+        iris.cube.CubeList
+        """
+        original_short_name = 'fld_s05i216'
+
+        self.cube = self.get_cube(cubes, var_name=original_short_name)
+
+        self.fix_var_name()
+
+        self.fix_long_name()
+
+        self.fix_coord_system()
+
+        return CubeList([self.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:
+      - '{institute}/{dataset_name}/{exp}/{modeling_realm}/netCDF'
+  input_file:
+    default: '{dataset}.{special_attr}-*.nc'
+  output_file: '{dataset}.{special_attr}'
+  cmor_type: 'CMIP6'
+  cmor_default_table_prefix: 'CMIP6_'