Add Fig 8 metrics to sea ice package

pcmdi_metrics/sea_ice/README.md

@@ -70,7 +70,7 @@ A [demo parameter file](https://github.com/PCMDI/pcmdi_metrics/blob/405_sic_ao/p
 
 ## Postprocessing
 
-A script is provided to create a multi-model bar chart using results from multiple runs of the sea ice driver. This script can be found in `./scripts/sea_ice_figures.py`. 
+Two postprocessing scripts are provided in `./scripts/sea_ice_figures.py`. The script `sea_ice_figures.py` creates a multi-model bar chart for all sectors using results from a model ensemble. The script `sea_ice_total_errors.py` plots the total errors for the Arctic and Antarctic for a model ensemble.
 
 Example command:
 ```

pcmdi_metrics/sea_ice/lib/sea_ice_lib.py

@@ -255,6 +255,35 @@ def process_by_region(ds, ds_var, ds_area, pole):
     return clims, means
 
 
+def get_area(data, ds_area):
+    xvar = find_lon(data)
+    coord_i, coord_j = get_xy_coords(data, xvar)
+    total_area = (data * ds_area).sum((coord_i, coord_j), skipna=True)
+    if isinstance(total_area.data, dask.array.core.Array):
+        ta_mean = total_area.data.compute().item()
+    else:
+        ta_mean = total_area.data.item()
+    return ta_mean
+
+
+def get_ocean_area_for_regions(ds, ds_var, area_val, pole):
+    # ds should have land/sea mask applied
+    regions_list = ["arctic", "antarctic", "ca", "na", "np", "sa", "sp", "io"]
+    areas = {}
+    # Only want spatial slice
+    if "time" in ds:
+        ds = ds.isel({"time": 0})
+    xvar = find_lon(ds)
+    yvar = find_lat(ds)
+    for region in regions_list:
+        data = choose_region(region, ds, ds_var, xvar, yvar, pole)
+        tmp = get_area(data, area_val)
+        areas[region] = tmp
+        print(tmp)
+        del data
+    return areas
+
+
 def find_lon(ds):
     for key in ds.coords:
         if key in ["lon", "longitude"]:

pcmdi_metrics/sea_ice/scripts/sea_ice_total_errors.py

@@ -0,0 +1,186 @@
+#!/usr/bin/env python
+import argparse
+import glob
+import json
+import os
+
+import matplotlib.pyplot as plt
+import numpy as np
+
+# ----------------
+# Load Metrics
+# ----------------
+parser = argparse.ArgumentParser(
+    prog="sea_ice_figures.py", description="Create figure for sea ice metrics"
+)
+parser.add_argument(
+    "--filelist",
+    dest="filelist",
+    default="sea_ice_metrics.json",
+    type=str,
+    help="Filename of sea ice metrics to glob. Permitted to use '*'",
+)
+parser.add_argument(
+    "--output_path",
+    dest="output_path",
+    default=".",
+    type=str,
+    help="The directory at which to write figure file",
+)
+args = parser.parse_args()
+
+filelist = args.filelist
+metrics_output_path = args.output_path
+
+model_list = []
+print(filelist)
+metrics = {"RESULTS": {}}
+for metrics_file in glob.glob(filelist):
+    with open(metrics_file) as mf:
+        results = json.load(mf)
+    for item in results["DIMENSIONS"]["model"]:
+        model_list.append(item)
+    metrics["RESULTS"].update(results["RESULTS"])
+
+model_list.sort()
+tmp = model_list[0]
+reference_data_set = list(metrics["RESULTS"][tmp]["arctic"]["model_mean"].keys())[0]
+
+# ----------------
+# Make figure
+# ----------------
+sector_list = ["Arctic", "Antarctic"]
+sector_short = ["arctic", "antarctic"]
+fig7, ax7 = plt.subplots(2, 1, figsize=(5, 4))
+mlabels = model_list
+ind = np.arange(len(mlabels))  # the x locations for the groups
+width = 0.7
+n = len(ind)
+for inds, sector in enumerate(sector_list):
+    # Assemble data
+    mse_clim = []
+    mse_ext = []
+    reg_clim = []
+    reg_ext = []
+    rgn = sector_short[inds]
+    for nmod, model in enumerate(model_list):
+        mse_clim.append(
+            float(
+                metrics["RESULTS"][model][rgn]["model_mean"][reference_data_set][
+                    "monthly_clim"
+                ]["mse"]
+            )
+        )
+        mse_ext.append(
+            float(
+                metrics["RESULTS"][model][rgn]["model_mean"][reference_data_set][
+                    "total_extent"
+                ]["mse"]
+            )
+        )
+        reg_clim.append(
+            float(
+                metrics["RESULTS"][model][rgn]["model_mean"][reference_data_set][
+                    "monthly_clim"
+                ]["sector_mse"]
+            )
+        )
+        reg_ext.append(
+            float(
+                metrics["RESULTS"][model][rgn]["model_mean"][reference_data_set][
+                    "total_extent"
+                ]["sector_mse"]
+            )
+        )
+
+    # plot bars
+    ax7[inds].bar(
+        ind,
+        mse_ext,
+        width,
+        color="r",
+        edgecolor="k",
+        linewidth=0.1,
+        label="Ann. Mean",
+        bottom=np.zeros(np.shape(mse_ext)),
+    )
+    ax7[inds].bar(
+        ind,
+        mse_clim,
+        width,
+        color="b",
+        edgecolor="k",
+        linewidth=0.1,
+        label="Ann. Cycle",
+        bottom=mse_ext,
+    )
+    bottom = [mse_ext[x] + mse_clim[x] for x in range(0, len(mse_ext))]
+    ax7[inds].bar(
+        ind,
+        reg_ext,
+        width,
+        color="y",
+        edgecolor="k",
+        linewidth=0.1,
+        label="Ann. Mean Reg.",
+        bottom=bottom,
+    )
+    bottom = [mse_ext[x] + mse_clim[x] + reg_ext[x] for x in range(0, len(mse_ext))]
+    ax7[inds].bar(
+        ind,
+        reg_clim,
+        width,
+        color="g",
+        edgecolor="k",
+        linewidth=0.1,
+        label="Ann. Cycle Reg.",
+        bottom=bottom,
+    )
+
+    # X axis label
+    if inds == len(sector_list) - 1:
+        ax7[inds].set_xticks(ind, mlabels, rotation=90, size=4, weight="bold")
+    else:
+        ax7[inds].set_xticks(ind, labels="")
+    ax7[inds].set_xlim(-1, len(mse_ext))
+
+    # Y axis
+    tmp = [
+        mse_ext[x] + mse_clim[x] + reg_ext[x] + reg_clim[x]
+        for x in range(0, len(mse_ext))
+    ]
+    datamax = np.nanmax(np.array(tmp))
+    ymax = (datamax) * 1.05
+    ax7[inds].set_ylim(0.0, ymax)
+    ticks = range(0, round(ymax), 10)
+    labels = [str(round(x, 0)) for x in ticks]
+    ax7[inds].set_yticks(ticks, labels, fontsize=5)
+
+    # subplot frame styling
+    ax7[inds].tick_params(color=[0.3, 0.3, 0.3])
+    for spine in ax7[inds].spines.values():
+        spine.set_edgecolor([0.3, 0.3, 0.3])
+        spine.set_linewidth(0.5)
+    # labels etc
+    ax7[inds].set_ylabel("10${^1}{^2}$km${^4}$", size=6, weight="bold")
+    ax7[inds].grid(True, linestyle=":", linewidth=0.5)
+    ax7[inds].annotate(
+        sector,
+        (0.35, 0.85),
+        xycoords="axes fraction",
+        size=6,
+        weight="bold",
+        bbox=dict(facecolor="white", edgecolor="white", pad=1),
+    )
+
+# Add legend, save figure
+leg = ax7[0].legend(loc="upper right", fontsize=5, edgecolor=[0.3, 0.3, 0.3])
+leg.get_frame().set_linewidth(0.5)  # legend styling
+t = plt.suptitle(
+    "Mean Square Error relative to " + reference_data_set, fontsize=8, y=0.93
+)
+plt.tight_layout()
+figfile = os.path.join(metrics_output_path, "total_MSE_bar_chart.png")
+plt.savefig(figfile, dpi=600)
+print("Figure written to ", figfile)
+print("Done")