adding provenance

esmvaltool/diag_scripts/eccc_extremes/simple_timeseries.py

@@ -40,8 +40,13 @@ def obtain_reference(data_group: list):
     maxs = list()
 
     for dataset_f in ref_fnames.keys():
+        dataset_n = ref_fnames[dataset_f][0]['dataset']
+        if len(group_metadata(data_group, 'dataset')[dataset_n])>1:
+            key = ref_fnames[dataset_f][0]['alias']
+        else: 
+            key = dataset_n
         ref_cb = iris.load_cube(dataset_f)
-        reference_dic[dataset_f] = ref_cb
+        reference_dic[key] = ref_cb
         # make key dataset if there's only one filename of this dataset otherwise alias
         mins.append(ref_cb.collapsed('time', iris.analysis.MIN).data)
         maxs.append(ref_cb.collapsed('time', iris.analysis.MAX).data)
@@ -55,7 +60,7 @@ def obtain_reference(data_group: list):
 def create_provenance(caption: str):
     '''Creates provenance dictionary'''
 
-    provenance_dic = {'authors': ['malinina24'], 
+    provenance_dic = {'authors': ['malinina_elizaveta'], 
                       'caption': caption,
                       'references': ['malinina24']}
 
@@ -93,41 +98,43 @@ def plot_timeseries(data_dic: dict, reference_dic: dict, cfg: dict,
         model_var_data = list()
         weights = list()
 
-        for i in range(0,len(data_dic[dataset]['var_data'])):
-            var_cb = data_dic[dataset]['var_data'][i]
+        for i in range(0,len(data_dic[dataset])):
+            var_cb = data_dic[dataset][i]
             model_var_data.append(var_cb.data)
             weights.append(var_cb.attributes['ensemble_weight']*
-                            var_cb[i].attributes['reverse_dtsts_n'])
+                            var_cb.attributes['reverse_dtsts_n'])
         weights = np.array(weights)
         model_var_data = np.array(model_var_data)
         mean_var_arr = np.average(model_var_data, axis=0)
         min_var_arr = np.min(model_var_data, axis=0)
         max_var_arr = np.max(model_var_data, axis=0)
 
-        years = [var_cb.coord('time').cell(i).year for i in range(len(var_cb.coord('time')))]
+        years = [var_cb.coord('time').cell(i).point.year for i in range(var_cb.coord('time').shape[0])]
 
         color_st = eplot.get_dataset_style(dataset, cfg.get('color_style'))
         ax_ts.plot(years, mean_var_arr, c=color_st['color'], zorder=3, label=dataset)
-        if len(data_dic[dataset]['var_data'])>1:
+        if len(data_dic[dataset])>1:
             ax_ts.fill_between(years, min_var_arr, max_var_arr, color=color_st['color'], lw=0, alpha=0.25)
 
         for ref_data in reference_dic.keys():
             ref_cb = reference_dic.get(ref_data)
             ref_color_st = eplot.get_dataset_style(ref_data, cfg.get('color_style'))
-            ref_years = [ref_cb.coord('time').cell(i).year for i in range(len(ref_cb.coord('time')))]
+            ref_years = [ref_cb.coord('time').cell(i).point.year for i in range(ref_cb.coord('time').shape[0])]
             ax_ts.plot(ref_years,ref_cb.data, label=ref_data, c=ref_color_st['color'], zorder=3)
 
         ax_ts.axhline(color='grey', zorder=1)
         ax_ts.set_ylim(b_min, b_max)
         ax_ts.legend(loc=0, ncols=4, fancybox=False, frameon=False)
 
         variable = cfg['var_label'] if cfg.get('var_label') else var_cb.var_name
+        exp_variable = variable.replace('_', ' ')
         units = cfg['units'] if cfg.get('units') else var_cb.units
         region = cfg['region'] if cfg.get('units') else 'region'
 
-        ax_ts.set_ylabel(f'{variable.replace('_', ' ')} , {units}')
+        ax_ts.set_ylabel(f'{exp_variable}, {units}')
+        ax_ts.set_xlabel('year')
 
-        default_caption = f'Timeseries of {variable} in {region}'
+        default_caption = f'Timeseries of {exp_variable} in {region}'
 
         caption = cfg['figure_caption'] if cfg.get('figure_caption') else default_caption
 
@@ -160,7 +167,11 @@ def main(cfg):
     maxs.append(ref_max)
     mins.append(ref_min)
 
-    datasets = group_metadata(groups, 'dataset')
+    remaining_metadata = []
+    for k in groups.keys():
+        remaining_metadata.extend(groups[k])
+
+    datasets = group_metadata(remaining_metadata, 'dataset')
     ens_var_cubelist = iris.cube.CubeList()
     for dataset in datasets.keys():
         filepaths = list(group_metadata(datasets[dataset], 'filename').keys())
@@ -175,8 +186,8 @@ def main(cfg):
             mod_var_cubelist.append(mod_cb)
             mins.append(mod_cb.collapsed('time', iris.analysis.MIN).data)
             maxs.append(mod_cb.collapsed('time', iris.analysis.MAX).data)
-        data_dic[dataset] = {'var_data': mod_var_cubelist}
-        data_dic['Multi-Model-Mean'] = {'var_data' : ens_var_cubelist}
+        data_dic[dataset] = mod_var_cubelist
+    data_dic['Multi-Model'] = ens_var_cubelist
 
     plot_timeseries(data_dic, reference_dic, cfg, 
                     min_val=np.asarray(mins).min(), 

esmvaltool/diag_scripts/eccc_extremes/variability_evaluation.py

@@ -1,18 +1,13 @@
-import csv
 import iris
-import climextremes as cex
 import iris.cube
-import pandas as pd
 import logging
 import numpy as np
 import matplotlib.pyplot as plt
 import os
-from scipy.stats import genextreme as gev
 
 # import internal esmvaltool modules here
-from esmvaltool.diag_scripts.shared import run_diagnostic, select_metadata, group_metadata, save_figure
+from esmvaltool.diag_scripts.shared import run_diagnostic, group_metadata, save_figure
 import esmvaltool.diag_scripts.shared.plot as eplot
-from esmvaltool.diag_scripts.ocean import diagnostic_tools as diagtools
 # # This part sends debug statements to stdout
 logger = logging.getLogger(os.path.basename(__file__))
 # logging.getLogger().addHandler(logging.StreamHandler(sys.stdout))
@@ -38,17 +33,21 @@ def obtain_reference(data_group: list):
     ref_fnames = group_metadata(data_group, 'filename')
 
     for dataset_f in ref_fnames.keys():
+        dataset_n = ref_fnames[dataset_f][0]['dataset']
+        if len(group_metadata(data_group, 'dataset')[dataset_n])>1:
+            key = ref_fnames[dataset_f][0]['alias']
+        else: 
+            key = dataset_n
         ref_cb = iris.load_cube(dataset_f)
-        reference_dic[dataset_f] = ref_cb.data.std()
-        # make key dataset if there's only one filename of this dataset otherwise alias
+        reference_dic[key] = ref_cb.data.std()
 
     return reference_dic
 
 
 def create_provenance(caption: str):
     '''Creates provenance dictionary'''
 
-    provenance_dic = {'authors': ['malinina24'], 
+    provenance_dic = {'authors': ['malinina_elizaveta'], 
                       'caption': caption,
                       'references': ['malinina24']}
 
@@ -109,9 +108,6 @@ def plot_stdevs(data_dic, reference_dic, cfg):
     mpl_st_file = eplot.get_path_to_mpl_style(cfg.get('mpl_style'))
     plt.style.use(mpl_st_file)
 
-    col_mod = (25/255, 14/255, 79/255)
-    col_obs = 'indianred'
-
     fig_stds, ax_stds = plt.subplots(nrows=1, ncols=1)
 
     fig_stds.set_size_inches(8., 9.)
@@ -124,35 +120,43 @@ def plot_stdevs(data_dic, reference_dic, cfg):
         color_st = eplot.get_dataset_style(model, cfg.get('color_style'))
         if model != 'Multi-Model':
             for i in range(0, len(data_dic[model])):
-                ax_stds.scatter(data_dic[model][i], nm+1, marker='o',
-                                    c=color_st['color'], zorder=2, label='individual member')
+                single_dot = ax_stds.scatter(data_dic[model][i], nm+1, marker='o',
+                                    c=color_st['color'], zorder=2, 
+                                    label='individual member')
             y_labs[nm+1] = model
         else:
-            ax_stds.scatter(data_dic[model], 0, c=color_st['color'], s=70, marker='s', zorder=3)
+            square = ax_stds.scatter(data_dic[model], 0, c=color_st['color'], s=70, 
+                                     marker='s', zorder=3, label='full ensemble')
             y_labs[0] = 'ALL'
 
+    legend_handles = [square, single_dot]
+
     for ref_dataset in reference_dic.keys():
         ref_color_st = eplot.get_dataset_style(ref_dataset, cfg.get('color_style'))
-        ax_stds.axvline(reference_dic[ref_dataset], c=ref_color_st['color'], zorder=2, 
+        ref_line = ax_stds.axvline(reference_dic[ref_dataset], c=ref_color_st['color'], zorder=2, 
                                                         label=ref_dataset)
+        legend_handles.append(ref_line)
     ax_stds.set_ylim(len(data_dic.keys()) -0.8, -0.2)
 
     ax_stds.set_yticks(y_ticks, labels=y_labs)
     ax_stds.grid(which='both', c='silver', zorder=1)
 
-    ax_stds.set_xlabel('StD of '+cfg['var_label'] + ' anomalies, C')
-    ax_stds.text(0.02, 0.97, cfg.get('litera')+' '+cfg['region'], fontsize='xx-large', transform=ax_stds.transAxes)
-
     variable = cfg.get('var_label')
+    exp_variable = variable.replace('_', ' ')
     units = cfg.get('units')
     region = cfg['region'] if cfg.get('units') else 'region'
 
+    ax_stds.set_xlabel(f'StD of {exp_variable}, {units}')
+
     default_caption = f'{variable} varibility in {region}'
 
     caption = cfg['figure_caption'] if cfg.get('figure_caption') else default_caption
     fig_stds.suptitle(caption)
 
     prov_dic = create_provenance(caption)
+
+    plt.legend(handles=legend_handles, bbox_to_anchor =(0.5,-0.1), 
+               loc='lower center', ncols=4, fancybox=False, frameon=False)
 
     plt.tight_layout()
 
@@ -171,9 +175,13 @@ def main(cfg):
 
     reference_dic = obtain_reference(groups.pop('reference'))
 
+    remaining_metadata = []
+    for k in groups.keys():
+        remaining_metadata.extend(groups[k])
+
     data_dic = {}
 
-    datasets = group_metadata(groups, 'dataset')
+    datasets = group_metadata(remaining_metadata, 'dataset')
     ens_var_cubelist = iris.cube.CubeList()
     for dataset in datasets.keys():
         filepaths = list(group_metadata(datasets[dataset], 'filename').keys())

esmvaltool/recipes/eccc_extremes/recipe_region_tnn_analytics.yml

@@ -27,19 +27,20 @@ preprocessors:
           - Alberta
     area_statistics:
       operator: mean
-    sesonal_statistics:
-      seasons: JASONDJFMAMJ
+    seasonal_statistics:
+      seasons: [SONDJFMAMJ, JA]
+      operator: min
+    annual_statistics:
       operator: min
     anomalies:
       period: full
       reference:
-      reference:
-        start_year: 1980
-        start_month: 7
+        start_year: 1981
+        start_month: 1
         start_day: 1
         end_year: 2010
-        end_month: 6
-        end_day: 30
+        end_month: 12
+        end_day: 31
 
   preproc_era:
     custom_order: True  
@@ -52,19 +53,20 @@ preprocessors:
           - Alberta
     area_statistics:
       operator: mean 
-    sesonal_statistics:
-      seasons: JASONDJFMAMJ
+    seasonal_statistics:
+      seasons: [SONDJFMAMJ, JA]
+      operator: min
+    annual_statistics:
       operator: min
     anomalies:
       period: full
       reference:
-        start_year: 1980
-        start_month: 7
+        start_year: 1981
+        start_month: 1
         start_day: 1
         end_year: 2010
-        end_month: 6
-        end_day: 30
-
+        end_month: 12
+        end_day: 31
 
 datasets_cmip: &datasets_cmip
   - {dataset: ACCESS-CM2, institute: CSIRO-ARCCSS, ensemble: r(1:10)i1p1f1, grid: gn}
@@ -121,16 +123,16 @@ diagnostics:
       txnx_all:
         short_name: tasmin
         exp: [historical, ssp245]
-        timerange: '19410701/20230630'
+        timerange: '19410901/20230630'
         mip: day
         project: CMIP6
-        preprocessor: preproc_txnx_flat
+        preprocessor: preproc_model
         additional_datasets: *datasets_cmip
       reference: 
         short_name: tasmin
-        timerange: '19410701/20230630'
+        timerange: '19410901/20230630'
         mip: day
-        preprocessor: preproc_era_txnx 
+        preprocessor: preproc_era
         additional_datasets: *dataset_era
     scripts:
       timeseries: