geoschem · yantosca · Mar 8, 2024 · Feb 12, 2024 · Feb 12, 2024 · Feb 12, 2024
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -17,6 +17,8 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Environment files `docs/environment_files/gcpy_requirements.txt` and `docs/environment_files/read_the_docs_requirements.txt`
 - New benchmark script `gcpy/benchmark/modules/benchmark_models_vs_sondes.py`
 - Added fixed level budget diagnostic to budget operations table
+- Function `rename_speciesconc_to_speciesconcvv` in `gcpy/benchmark/modules/benchmark_utils.py`
+- Function `copy_file_to_dir` in `gcpy/util.py`.  This is a wrapper for `shutil.copyfile`.
 
 ### Changed
 - Bump pip from 23.2.1 to 23.3 (dependabot suggested this)
@@ -33,15 +35,21 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Refactored `get_vert_grid.py` (in `gcpy/grid.py`) to accept the `p_sfc` argument; Also never-nested the if-block logic.
 - `benchmark_slurm.sh` script now saves output to a log file with the same base name as the YAML config file
 - `benchmark_models_vs_obs.py` now reads the observational data paths and metadata from  `1yr_fullchem_benchmark.yml`
+- Moved benchmark code from `gcpy/` to `gcpy/benchmark/modules/`
+- Benchmark functions now call `rename_speciesconc_to_speciesconcvv`
+- Create radionuclide, STE flux, and mass conservation tables for Ref and Dev versions in TransportTracers benchmarks
+- Use new function `copy_file_to_dir` to copy the benchmark script and configuration file to the benchmark results folders
 
 ### Fixed
 - CS inquiry functions in `gcpy/cstools.py` now work properly for `xr.Dataset` and `xr.DataArray` objects
 - Prevent an import error by using `seaborn-v0_8-darkgrid` in`gcpy/benchmark/modules/benchmark_models_vs_obs.py`
 - `gcpy/file_regrid.py` now creates GCHP files with `DELP_DRY` instead of `DELPDRY`
 - Fixed bugs in GCHP vs GCHP sections of 1-yr benchmark run scripts to allow comparison of GCHP runs with different grid resolutions
-- Fixed silent bug in transport tracer benchmark GCC vs GCHP mass tables preventing them from being generated
+- Fixed silent bug in transport tracer benchmark GCC vs GCHP mass
+  tables preventi ng them from being generated
 - Import error in `gcpy/examples/diagnostics/compare_diags.py`
 - Added missing `n_cores` to `gcpy/examples/diagnostics/compare_diags.yml`
+- Add `docs/requirements.txt` symbolic link to `docs/environment_files/read_the_docs_requirements.txt` for RTD builds 
 
 ### Removed
 - Example script `gcpy/examples/plotting/mda8_o3_timeseries.py`

diff --git a/docs/requirements.txt b/docs/requirements.txt
@@ -0,0 +1 @@
+environment_files/read_the_docs_requirements.txt
diff --git a/gcpy/__init__.py b/gcpy/__init__.py
@@ -6,22 +6,16 @@
 from .examples import *
 
 from .append_grid_corners import *
-from .benchmark_funcs import *
-from .budget_ox import *
-from .budget_tt import *
 from .constants import *
 from .cstools import *
 from .date_time import *
 from .file_regrid import *
 from .grid import *
 from .grid_stretching_transforms import *
-from .mean_oh_from_logs import *
-from .oh_metrics import *
 from .plot import *
 from .raveller_1D import *
 from .regrid import *
 from .regrid_restart_file import *
-from .ste_flux import *
 from .units import *
 from .util import *
 from ._version import *
diff --git a/gcpy/benchmark/benchmark_slurm.sh b/gcpy/benchmark/benchmark_slurm.sh
@@ -34,7 +34,7 @@ config="1yr_fullchem_benchmark.yml"
 #config="1yr_tt_benchmark.yml"
 
 # Call the run_benchmark script to make the plots
-python -m gcpy.benchmark.run_benchmark "${config}" > "${config/.yml/.log}: 2>&1
+python -m gcpy.benchmark.run_benchmark "${config}" > "${config/.yml/.log}" 2>&1
 
 # Turn off python environment
 mamba deactivate

diff --git a/gcpy/aod_species.yml → gcpy/benchmark/modules/aod_species.yml b/gcpy/aod_species.yml → gcpy/benchmark/modules/aod_species.yml
diff --git a/gcpy/benchmark_categories.yml → ...enchmark/modules/benchmark_categories.yml b/gcpy/benchmark_categories.yml → ...enchmark/modules/benchmark_categories.yml
diff --git a/gcpy/benchmark/modules/benchmark_drydep.py b/gcpy/benchmark/modules/benchmark_drydep.py
@@ -6,7 +6,9 @@
 import numpy as np
 from gcpy import util
 from gcpy.plot.compare_single_level import compare_single_level
-import gcpy.benchmark.modules.benchmark_utils as bmk_util
+from gcpy.benchmark.modules.benchmark_utils import \
+    get_lumped_species_definitions, make_output_dir, \
+    pdf_filename, print_sigdiffs, read_ref_and_dev
 
 # Suppress numpy divide by zero warnings to prevent output spam
 np.seterr(divide="ignore", invalid="ignore")
@@ -90,23 +92,23 @@ def make_benchmark_drydep_plots(
     """
 
     # Create directory for plots (if it doesn't exist)
-    dst = bmk_util.make_output_dir(
+    dst = make_output_dir(
         dst,
         collection,
         subdst,
         overwrite=overwrite,
     )
 
     # Read data
-    refdata, devdata = bmk_util.read_ref_and_dev(
+    refdata, devdata = read_ref_and_dev(
         ref,
         dev,
         time_mean=time_mean
     )
 
     # Get common variables between Ref and Dev
     if varlist is None:
-        varlist = bmk_util.get_common_varnames(
+        varlist = get_common_varnames(
             refdata,
             devdata,
             prefix="DryDepVel_",
@@ -115,7 +117,7 @@ def make_benchmark_drydep_plots(
 
     # Create surface plots
     sigdiff_list = []
-    pdfname = bmk_util.pdf_filename(
+    pdfname = pdf_filename(
         dst,
         collection,
         subdst,
@@ -145,7 +147,7 @@ def make_benchmark_drydep_plots(
     )
 
     # Write significant differences to file (if there are any)
-    bmk_util.print_sigdiffs(
+    print_sigdiffs(
         sigdiff_files,
         sigdiff_list,
         sigdiff_type="sfc",

diff --git a/gcpy/benchmark_funcs.py → gcpy/benchmark/modules/benchmark_funcs.py b/gcpy/benchmark_funcs.py → gcpy/benchmark/modules/benchmark_funcs.py
@@ -20,15 +20,14 @@
 from gcpy.constants import COL_WIDTH, MW_AIR_g, skip_these_vars, TABLE_WIDTH
 from gcpy.plot.compare_single_level import compare_single_level
 from gcpy.plot.compare_zonal_mean import compare_zonal_mean
+from gcpy.benchmark.modules.benchmark_utils import \
+    AOD_SPC, EMISSION_SPC, EMISSION_INV, add_lumped_species_to_dataset, \
+    archive_lumped_species_definitions, get_species_categories, \
+    archive_species_categories, rename_speciesconc_to_speciesconcvv
 
 # Suppress numpy divide by zero warnings to prevent output spam
 np.seterr(divide="ignore", invalid="ignore")
 
-# YAML files
-AOD_SPC = "aod_species.yml"
-EMISSION_SPC = "emission_species.yml"
-EMISSION_INV = "emission_inventories.yml"
-
 
 def create_total_emissions_table(
         refdata,
@@ -1031,14 +1030,8 @@ def make_benchmark_conc_plots(
 
     # Rename SpeciesConc_ to SpeciesConcVV_ for consistency with new
     # naming introduced in GEOS-Chem 14.1.0
-    for v in refds.data_vars.keys():
-        if v.startswith('SpeciesConc_'):
-            spc = v.replace('SpeciesConc_', '')
-            refds = refds.rename({v: 'SpeciesConcVV_' + spc})
-    for v in devds.data_vars.keys():
-        if v.startswith('SpeciesConc_'):
-            spc = v.replace('SpeciesConc_', '')
-            devds = devds.rename({v: 'SpeciesConcVV_' + spc})
+    refds = rename_speciesconc_to_speciesconcvv(refds)
+    devds = rename_speciesconc_to_speciesconcvv(devds)
 
     # -----------------------------------------------------------------
     # Kludge, rename wrong variable name
@@ -1229,22 +1222,22 @@ def make_benchmark_conc_plots(
         print("\nComputing lumped species for full chemistry benchmark")
 
         print("-->Adding lumped species to ref dataset")
-        refds = util.add_lumped_species_to_dataset(refds)
+        refds = add_lumped_species_to_dataset(refds)
 
         print("-->Adding lumped species to dev dataset")
-        devds = util.add_lumped_species_to_dataset(devds)
+        devds = add_lumped_species_to_dataset(devds)
 
         if diff_of_diffs:
             print("-->Adding lumped species to dev datasets")
-            second_refds = util.add_lumped_species_to_dataset(second_refds)
-            second_devds = util.add_lumped_species_to_dataset(second_devds)
+            second_refds = add_lumped_species_to_dataset(second_refds)
+            second_devds = add_lumped_species_to_dataset(second_devds)
 
-        util.archive_lumped_species_definitions(dst)
+        archive_lumped_species_definitions(dst)
         print("Lumped species computation complete.\n")
 
     # Get the list of species categories
-    catdict = util.get_species_categories(benchmark_type)
-    util.archive_species_categories(dst)
+    catdict = get_species_categories(benchmark_type)
+    archive_species_categories(dst)
 
     # Make sure that Ref and Dev datasets have the same variables.
     # Variables that are in Ref but not in Dev will be added to Dev
@@ -1618,7 +1611,7 @@ def make_benchmark_emis_plots(
             Set this flag to True to separate plots into PDF files
             according to the benchmark species categories (e.g. Oxidants,
             Aerosols, Nitrogen, etc.)  These categories are specified
-            in the YAML file benchmark_species.yml.
+            in the YAML file benchmark_categories.yml.
             Default value: False
         plot_by_hco_cat: bool
             Set this flag to True to separate plots into PDF files
@@ -1898,7 +1891,7 @@ def createfile_hco_cat(c):
     # ==================================================================
     if plot_by_spc_cat:
 
-        catdict = util.get_species_categories(benchmark_type)
+        catdict = get_species_categories(benchmark_type)
         # in case any emissions are skipped (for use in nested pdf bookmarks)
         warninglist = ([])
         # for checking if emissions species not defined in benchmark category
@@ -4037,10 +4030,11 @@ def make_benchmark_aerosol_tables(
     mw["Air"] = MW_AIR_g
 
     # Get the list of relevant AOD diagnostics from a YAML file
+    ifile= AOD_SPC
     aod = util.read_config_file(
         os.path.join(
             os.path.dirname(__file__),
-            "aod_species.yml"
+            ifile,
         ),
         quiet=True
     )
@@ -4085,10 +4079,7 @@ def make_benchmark_aerosol_tables(
 
     # Rename SpeciesConc_ to SpeciesConcVV_ for consistency with new
     # naming introduced in GEOS-Chem 14.1.0
-    for v in ds_spc.data_vars.keys():
-        if v.startswith('SpeciesConc_'):
-            spc = v.replace('SpeciesConc_', '')
-            ds_spc = ds_spc.rename({v: 'SpeciesConcVV_' + spc})
+    ds_spc = rename_speciesconc_to_speciesconcvv(ds_spc)
 
     # Get troposphere mask
     tropmask = get_troposphere_mask(ds_met)
@@ -4941,15 +4932,6 @@ def make_benchmark_mass_conservation_table(
             Path names of restart files.
         runstr: str
             Name to put in the filename and header of the output file
-        refstr: str
-            A string to describe ref (e.g. version number)
-        dev: str
-            Path name of "Dev" (aka "Development") data set file.
-            The "Dev" data set will be compared against the "Ref" data set.
-        devmet: list of str
-            Path name of dev meteorology data set.
-        devstr: str
-            A string to describe dev (e.g. version number)
 
     Keyword Args (optional):
         dst: str
@@ -5074,7 +5056,7 @@ def make_benchmark_mass_conservation_table(
     # Print masses to file
     # ==================================================================
     # Create file
-    outfilename = os.path.join(dst, "Passive_mass.txt")
+    outfilename = os.path.join(dst, f"Passive_mass.{runstr}.txt")
 
     with open(outfilename, 'w') as f:
         titlestr = '  Global Mass of Passive Tracer in ' + runstr + '  '

diff --git a/gcpy/benchmark/modules/benchmark_models_vs_obs.py b/gcpy/benchmark/modules/benchmark_models_vs_obs.py
@@ -24,7 +24,8 @@
 from gcpy.util import verify_variable_type, dataset_reader, make_directory
 from gcpy.cstools import extract_grid
 from gcpy.grid import get_nearest_model_data
-from gcpy.benchmark.modules.benchmark_utils import get_geoschem_level_metadata
+from gcpy.benchmark.modules.benchmark_utils import \
+    get_geoschem_level_metadata, rename_speciesconc_to_speciesconcvv
 
 
 def read_nas(
@@ -199,38 +200,11 @@ def read_model_data(
         multi_files=True,
         verbose=verbose,
     )
-
-    # Set temporary variable name for use below
-    varname_tmp = varname
-
-    # First try reading the data as-is
-    try:
-        dataset = reader(
-            filepaths,
-            drop_variables=skip_these_vars,
-            data_vars=[varname_tmp]
-        ).load()
-
-    # If we encounter a ValueError, it may be because the data is
-    # older # and may have e.g. SpeciesConc fields instead of
-    # SpeciesConcVV fields.  Reset the varname_tmp and try again.
-    except ValueError:
-        varname_tmp = varname_tmp.replace("VV", "")
-        dataset = reader(
-            filepaths,
-            drop_variables=skip_these_vars,
-            data_vars=[varname_tmp]
-        ).load()
-
-        # Rename to the original name to avid confusion with data
-        # from GEOS-Chem versions prior to 14.1.0
-        with xr.set_options(keep_attrs=True):
-            dataset = dataset.rename({varname_tmp: varname})
-
-    # If we fail again, then throw an error!
-    except [FileNotFoundError, OSError, IOError] as exc:
-        msg = f"get_model_data: Could not read Ref data for {varname}!"
-        raise exc(msg) from exc
+
+    # Read data and rename SpeciesConc_ to SpeciesConcVV_, if necessary
+    # (needed for backwards compatibility with older versions.)
+    dataset = reader(filepaths,drop_variables=skip_these_vars).load()
+    dataset = rename_speciesconc_to_speciesconcvv(dataset)
 
     # Create a DataArray object and convert to ppbv (if necessary)
     with xr.set_options(keep_attrs=True):

diff --git a/gcpy/benchmark/modules/benchmark_models_vs_sondes.py b/gcpy/benchmark/modules/benchmark_models_vs_sondes.py
@@ -12,7 +12,8 @@
 from gcpy.cstools import extract_grid
 from gcpy.grid import get_nearest_model_data, get_vert_grid
 from gcpy.util import make_directory, verify_variable_type
-from gcpy.benchmark.modules.benchmark_utils import read_ref_and_dev
+from gcpy.benchmark.modules.benchmark_utils import \
+    read_ref_and_dev, rename_speciesconc_to_speciesconcvv
 
 # Tell matplotlib not to look for an X-window
 os.environ["QT_QPA_PLATFORM"] = "offscreen"
@@ -42,6 +43,11 @@ def get_ref_and_dev_model_data(
         multi_file=True
     )
 
+    # Rename variables starting with "SpeciesConc_" to "SpeciesConcVV_",
+    # for backwards compatibility with GC versions prior to 14.1.0.
+    ref_data = rename_speciesconc_to_speciesconcvv(ref_data)
+    dev_data = rename_speciesconc_to_speciesconcvv(dev_data)
+
     # Return the species of interest and convert to ppbv
     return ref_data[varname]*1.0e9, dev_data[varname]*1.0e9
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		environment_files/read_the_docs_requirements.txt