Merge pull request #49 from JGCRI/acs-gridded-enriched

Acs gridded enriched
JGCRI · Apr 7, 2022 · c44b4a9 · c44b4a9
2 parents a570463 + c1cc59e
commit c44b4a9
Show file tree

Hide file tree

Showing 3 changed files with 26 additions and 3 deletions.
diff --git a/modules/tas_psl_stitching_paper1.py b/modules/tas_psl_stitching_paper1.py
@@ -10,6 +10,7 @@
 OUTPUT_DIR = pkg_resources.resource_filename('stitches', 'data')
 
 # Stitch one realization per target ensemble, match in specified tolerance.
+filter_future = True
 Nmatches = 1
 tolerance = 0.075
 # #####################################################
@@ -103,7 +104,7 @@ def get_orig_GSAT_data(target_df):
 del(i2)
 
 # #####################################################
-model_list = ["CAMS-CSM1-0", "MIROC6"]
+model_list = [ "CAMS-CSM1-0", "MIROC6"]
 
 # model_list = ["ACCESS-CM2", "ACCESS-ESM1-5", "AWI-CM-1-1-MR", "CAMS-CSM1-0", "CanESM5", "CAS-ESM2-0",
 #               "CESM2", "CESM2-WACCM", "FGOALS-g3", "FIO-ESM-2-0", "GISS-E2-1-G", "GISS-E2-1-H",
@@ -165,7 +166,7 @@ def get_orig_GSAT_data(target_df):
                               reproducible=True)
 
     # Save the recipe
-    rp.to_csv(OUTPUT_DIR + "/tas_psl_pr/" + model_to_use + "_ssp245_rp.csv")
+    rp.to_csv(OUTPUT_DIR + "/tas_psl_pr/" + model_to_use + "_ssp245_rp.csv",index=False)
 
     # Stitch save the stitched GSATs for quality assurance:
     gsat_245 = stitches.gmat_stitching(rp)
@@ -227,7 +228,7 @@ def get_orig_GSAT_data(target_df):
                               reproducible=True)
 
     # Save the recipe
-    rp.to_csv(OUTPUT_DIR + "/tas_psl_pr/" + model_to_use + "_ssp370_rp.csv")
+    rp.to_csv(OUTPUT_DIR + "/tas_psl_pr/" + model_to_use + "_ssp370_rp.csv", index=False)
 
     # Stitch save the stitched GSATs for quality assurance:
     gsat_370 = stitches.gmat_stitching(rp)
@@ -286,3 +287,20 @@ def get_orig_GSAT_data(target_df):
                                       model_to_use+ '.csv'), index=False)
 
 # end for loop over ESMs
+
+
+# Filter generated files to future if needed
+if filter_future:
+    from pathlib import Path
+    # read in each stitched realization, subset to 2015
+    entries = Path((OUTPUT_DIR + '/tas_psl_pr/stitched/'))
+    for entry in entries.iterdir():
+        print(entry.name)
+        ds = xr.open_dataset((OUTPUT_DIR + '/tas_psl_pr/stitched/') + entry.name)
+        ds = ds.sel(time = slice('2015-01-31', '2100-12-31')).copy()
+        ds.to_netcdf((OUTPUT_DIR + '/tas_psl_pr/stitched_future/') + entry.name)
+        del(ds)
+
+
+
+
diff --git a/stitches/fx_recepie.py b/stitches/fx_recepie.py
@@ -764,6 +764,7 @@ def make_recipe(target_data, archive_data, N_matches, res="mon", tol=0.1, non_ta
     else:
         out = recipe.copy()
 
+    out = out.sort_values(by=['stitching_id', 'target_start_yr']).reset_index(drop=True).copy()
     return out
 
 
diff --git a/stitches/fx_stitch.py b/stitches/fx_stitch.py
@@ -227,6 +227,8 @@ def gridded_stitching(out_dir, rp):
                             'archive_variable', 'archive_model', 'archive_ensemble', 'stitching_id',
                             'archive_start_yr', 'archive_end_yr'})
 
+    rp = rp.sort_values(by=['stitching_id', 'target_start_yr']).reset_index(drop=True).copy()
+
     # Determine which variables will be downloaded.
     variables = find_var_cols(rp)
     if not (len(variables) >= 1):
@@ -331,6 +333,8 @@ def gmat_stitching(rp):
                             'archive_variable', 'archive_model', 'archive_ensemble', 'stitching_id',
                             'archive_start_yr', 'archive_end_yr', 'tas_file'})
 
+    rp = rp.sort_values(by=['stitching_id', 'target_start_yr']).reset_index(drop=True).copy()
+
     # One the assumptions of this function is that it only works with tas, so
     # we can safely add tas as the variable column.
     rp['variable'] = 'tas'