notebooks/scenario-metrics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Scenario metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "make: Entering directory '/home/martibosch/data-science/lausanne-greening-scenarios'\r\n",
      "make: Nothing to be done for 'scenarios'.\r\n",
      "make: Nothing to be done for 'scenario_metrics'.\r\n",
      "make: Leaving directory '/home/martibosch/data-science/lausanne-greening-scenarios'\r\n"
     ]
    }
   ],
   "source": [
    "!make -C .. scenarios scenario_metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import xarray as xr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "scenario_ds_filepath = '../data/processed/scenarios.nc'\n",
    "scenario_metrics_filepath = '../data/processed/scenario-metrics.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "metrics = [\n",
    "    'proportion_of_landscape', 'area_mn', 'shape_index_mn', 'edge_density'\n",
    "]\n",
    "metric_label_dict = {\n",
    "    'proportion_of_landscape': 'PLAND',\n",
    "    'area_mn': 'AREA_MN',\n",
    "    'shape_index_mn': 'SHAPE_MN',\n",
    "    'edge_density': 'ED'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "scenario_T_df = xr.open_dataset(scenario_ds_filepath)['T'].mean(\n",
    "    dim=['x', 'y']).to_dataframe().reset_index()\n",
    "scenario_metrics_df = pd.read_csv(scenario_metrics_filepath).rename(\n",
    "    columns=metric_label_dict)\n",
    "scenario_df = pd.merge(scenario_T_df, scenario_metrics_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 795.25x180 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# start: without line\n",
    "# g = sns.pairplot(scenario_df,\n",
    "#                  x_vars=metric_label_dict.values(),\n",
    "#                  y_vars=['T'],\n",
    "#                  hue='interaction')\n",
    "# g._legend.remove()\n",
    "# end\n",
    "# start: with line\n",
    "g = sns.PairGrid(scenario_df,\n",
    "                 x_vars=metric_label_dict.values(),\n",
    "                 y_vars=['T'],\n",
    "                 hue='interaction')\n",
    "g.map(plt.plot, linestyle='dashed', marker='x')\n",
    "# end\n",
    "g.axes.flat[0].set_ylabel('$\\\\overline{T} \\; [\\degree C]$')\n",
    "g.add_legend(title='Approach', bbox_to_anchor=(1.02, .5), frameon=True)\n",
    "g.fig.savefig('../reports/figures/scenario-metrics.pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us first list the simulated average temperature and metric values for the initial and final scenarios, which are deterministic and independent of the sampling approach (and hence independent of the scenario run too):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>T</th>\n",
       "      <th>AREA_MN</th>\n",
       "      <th>ED</th>\n",
       "      <th>SHAPE_MN</th>\n",
       "      <th>PLAND</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>change_prop</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>23.174244</td>\n",
       "      <td>0.137148</td>\n",
       "      <td>192.736342</td>\n",
       "      <td>1.206352</td>\n",
       "      <td>17.264191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>22.313562</td>\n",
       "      <td>1.517883</td>\n",
       "      <td>354.180004</td>\n",
       "      <td>1.246372</td>\n",
       "      <td>53.365610</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     T   AREA_MN          ED  SHAPE_MN      PLAND\n",
       "change_prop                                                      \n",
       "0.0          23.174244  0.137148  192.736342  1.206352  17.264191\n",
       "1.0          22.313562  1.517883  354.180004  1.246372  53.365610"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "endpoint_cond = (scenario_df['change_prop'] == 0) | (\n",
    "    scenario_df['change_prop'] == 1)\n",
    "scenario_df[endpoint_cond].drop(['interaction', 'scenario_run'],\n",
    "                                axis=1).groupby('change_prop').first()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us now list the simulated average temperature and metric values (averaged out of all the scenario runs) for each sampling approach and proportion of transformed pixels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>T</th>\n",
       "      <th>AREA_MN</th>\n",
       "      <th>ED</th>\n",
       "      <th>SHAPE_MN</th>\n",
       "      <th>PLAND</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>interaction</th>\n",
       "      <th>change_prop</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">cluster</th>\n",
       "      <th>0.125</th>\n",
       "      <td>23.093513</td>\n",
       "      <td>0.301594</td>\n",
       "      <td>197.523652</td>\n",
       "      <td>1.273277</td>\n",
       "      <td>21.776869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.250</th>\n",
       "      <td>22.984376</td>\n",
       "      <td>0.535728</td>\n",
       "      <td>239.398563</td>\n",
       "      <td>1.508770</td>\n",
       "      <td>26.289546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.375</th>\n",
       "      <td>22.887530</td>\n",
       "      <td>0.203913</td>\n",
       "      <td>310.247279</td>\n",
       "      <td>1.179456</td>\n",
       "      <td>30.802223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.500</th>\n",
       "      <td>22.764974</td>\n",
       "      <td>0.207450</td>\n",
       "      <td>410.298673</td>\n",
       "      <td>1.266777</td>\n",
       "      <td>35.314901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.625</th>\n",
       "      <td>22.623620</td>\n",
       "      <td>0.381483</td>\n",
       "      <td>464.811584</td>\n",
       "      <td>1.327290</td>\n",
       "      <td>39.827578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.750</th>\n",
       "      <td>22.488111</td>\n",
       "      <td>0.731016</td>\n",
       "      <td>473.705808</td>\n",
       "      <td>1.294763</td>\n",
       "      <td>44.340256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.875</th>\n",
       "      <td>22.378273</td>\n",
       "      <td>1.125851</td>\n",
       "      <td>436.434237</td>\n",
       "      <td>1.275849</td>\n",
       "      <td>48.852933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">random</th>\n",
       "      <th>0.125</th>\n",
       "      <td>23.082969</td>\n",
       "      <td>0.073215</td>\n",
       "      <td>330.766023</td>\n",
       "      <td>1.175691</td>\n",
       "      <td>21.776869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.250</th>\n",
       "      <td>22.963175</td>\n",
       "      <td>0.089278</td>\n",
       "      <td>435.186904</td>\n",
       "      <td>1.278716</td>\n",
       "      <td>26.289546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.375</th>\n",
       "      <td>22.826046</td>\n",
       "      <td>0.148264</td>\n",
       "      <td>506.051892</td>\n",
       "      <td>1.398372</td>\n",
       "      <td>30.802223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.500</th>\n",
       "      <td>22.687901</td>\n",
       "      <td>0.289986</td>\n",
       "      <td>542.719359</td>\n",
       "      <td>1.397413</td>\n",
       "      <td>35.314901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.625</th>\n",
       "      <td>22.558785</td>\n",
       "      <td>0.537328</td>\n",
       "      <td>546.217723</td>\n",
       "      <td>1.332756</td>\n",
       "      <td>39.827578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.750</th>\n",
       "      <td>22.447719</td>\n",
       "      <td>0.850377</td>\n",
       "      <td>515.837512</td>\n",
       "      <td>1.296232</td>\n",
       "      <td>44.340256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.875</th>\n",
       "      <td>22.364208</td>\n",
       "      <td>1.176964</td>\n",
       "      <td>452.048122</td>\n",
       "      <td>1.283405</td>\n",
       "      <td>48.852933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"7\" valign=\"top\">scatter</th>\n",
       "      <th>0.125</th>\n",
       "      <td>23.090456</td>\n",
       "      <td>0.056511</td>\n",
       "      <td>346.034464</td>\n",
       "      <td>1.129920</td>\n",
       "      <td>21.776869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.250</th>\n",
       "      <td>22.961758</td>\n",
       "      <td>0.066709</td>\n",
       "      <td>453.900181</td>\n",
       "      <td>1.226749</td>\n",
       "      <td>26.289546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.375</th>\n",
       "      <td>22.815506</td>\n",
       "      <td>0.098572</td>\n",
       "      <td>515.756687</td>\n",
       "      <td>1.314237</td>\n",
       "      <td>30.802223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.500</th>\n",
       "      <td>22.676811</td>\n",
       "      <td>0.145654</td>\n",
       "      <td>532.233781</td>\n",
       "      <td>1.324834</td>\n",
       "      <td>35.314901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.625</th>\n",
       "      <td>22.566680</td>\n",
       "      <td>0.197523</td>\n",
       "      <td>503.104816</td>\n",
       "      <td>1.280580</td>\n",
       "      <td>39.827578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.750</th>\n",
       "      <td>22.456726</td>\n",
       "      <td>0.559311</td>\n",
       "      <td>481.061317</td>\n",
       "      <td>1.248731</td>\n",
       "      <td>44.340256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.875</th>\n",
       "      <td>22.351171</td>\n",
       "      <td>1.259266</td>\n",
       "      <td>451.289479</td>\n",
       "      <td>1.272145</td>\n",
       "      <td>48.852933</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 T   AREA_MN          ED  SHAPE_MN      PLAND\n",
       "interaction change_prop                                                      \n",
       "cluster     0.125        23.093513  0.301594  197.523652  1.273277  21.776869\n",
       "            0.250        22.984376  0.535728  239.398563  1.508770  26.289546\n",
       "            0.375        22.887530  0.203913  310.247279  1.179456  30.802223\n",
       "            0.500        22.764974  0.207450  410.298673  1.266777  35.314901\n",
       "            0.625        22.623620  0.381483  464.811584  1.327290  39.827578\n",
       "            0.750        22.488111  0.731016  473.705808  1.294763  44.340256\n",
       "            0.875        22.378273  1.125851  436.434237  1.275849  48.852933\n",
       "random      0.125        23.082969  0.073215  330.766023  1.175691  21.776869\n",
       "            0.250        22.963175  0.089278  435.186904  1.278716  26.289546\n",
       "            0.375        22.826046  0.148264  506.051892  1.398372  30.802223\n",
       "            0.500        22.687901  0.289986  542.719359  1.397413  35.314901\n",
       "            0.625        22.558785  0.537328  546.217723  1.332756  39.827578\n",
       "            0.750        22.447719  0.850377  515.837512  1.296232  44.340256\n",
       "            0.875        22.364208  1.176964  452.048122  1.283405  48.852933\n",
       "scatter     0.125        23.090456  0.056511  346.034464  1.129920  21.776869\n",
       "            0.250        22.961758  0.066709  453.900181  1.226749  26.289546\n",
       "            0.375        22.815506  0.098572  515.756687  1.314237  30.802223\n",
       "            0.500        22.676811  0.145654  532.233781  1.324834  35.314901\n",
       "            0.625        22.566680  0.197523  503.104816  1.280580  39.827578\n",
       "            0.750        22.456726  0.559311  481.061317  1.248731  44.340256\n",
       "            0.875        22.351171  1.259266  451.289479  1.272145  48.852933"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scenario_df[~endpoint_cond].drop('scenario_run', axis=1).groupby(\n",
    "    ['interaction', 'change_prop']).mean()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (lausanne-greening-scenarios)",
   "language": "python",
   "name": "lausanne-greening-scenarios"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.8"
  },
  "name": "scenario-metrics.ipynb"
 },
 "nbformat": 4,
 "nbformat_minor": 4
}