add year vs region

notebooks/wp5/glacier_distribution.ipynb

@@ -214,6 +214,40 @@
     "    title=\"Glacier distribution around the year 2000 according to the RGI v6.0\",\n",
     ")"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "17",
+   "metadata": {},
+   "source": [
+    "# Plot year VS region"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "18",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def weighted_average(df, field_name, weights_name):\n",
+    "    df = df[df[field_name].notnull() & df[weights_name].notnull()]\n",
+    "    weights = df[weights_name]\n",
+    "    return (df[field_name] * weights).sum() / weights.sum()\n",
+    "\n",
+    "\n",
+    "gdf[\"region\"] = gdf[\"RGIID\"].str[6:8].astype(int)\n",
+    "grouped = gdf[[\"year\", \"AREA\", \"region\"]].groupby(\"region\")\n",
+    "arithmetic_year = grouped[\"year\"].mean()\n",
+    "weighted_year = grouped.apply(weighted_average, \"year\", \"AREA\", include_groups=False)\n",
+    "\n",
+    "ax = arithmetic_year.plot(label=\"arithmetic\")\n",
+    "ax = weighted_year.where(weighted_year != 0).plot(label=\"weighted\")\n",
+    "ax.set_xlabel(\"RGI region number\")\n",
+    "ax.set_ylabel(\"Year of digitalization\")\n",
+    "ax.grid()\n",
+    "ax.legend()"
+   ]
   }
  ],
  "metadata": {