{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Etude de la concentration de CO2 dans l'atmosphère depuis 1958" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Importation des librairies nécessaires à l'analyse:" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import pandas as pd\n", "import numpy as np" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Importation des données\n", "\n", "Les données sont accessible sur le site de [l'institut Scripps](https://scrippsco2.ucsd.edu/data/atmospheric_co2/primary_mlo_co2_record.html). Ils s'agit d'un suivi des teneurs en CO2 mesurées au Mauna Loa Observatory à Hawaii depuis 1958. Elles sont téléchargée en date du 02/06/2021. Si le fichier de données n'a plus de version locale, il sera téléchargé." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "data_url = 'https://scrippsco2.ucsd.edu/assets/data/atmospheric/stations/in_situ_co2/monthly/monthly_in_situ_co2_mlo.csv'\n", "\n", "data_file = \"C02-atmosphere.csv\"\n", "import os\n", "import urllib.request\n", "if not os.path.exists(data_file):\n", " urllib.request.urlretrieve(data_url, data_file)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Les 54 premières lignes du fichiers sont une description des données et quelques indications, elle ne seront pas prise en compte lors de la création du dataFrame de l'analyse. \n", "La structure des données, comme indiquée dans le fichier CSV est la suivante: \n", "\n", "| Nom de colonne | Libellé de colonne |\n", "|----------------|-----------------------------------------------------------------------------------------------------------------------------------|\n", "| Year | Année aucours de laquelle la mesure a été faite |\n", "| Month | Mois au cours duquel la mesure a été faite\n", " |\n", "| Date | Date de la mesure au format Excel\n", " |\n", "| Date | Date de la mesure au format ISO\n", " |\n", "| CO2 (ppm) | Taux de CO2 en micro mole par mole (ppm)\n", " |\n", "| CO2 adjusted (ppm)| Taux de CO2 auquel on a retiré les variations saisonnières\n", " |\n", "| CO2 smoothed(ppm) | Taux de CO2 ajusté\n", " |\n", "| CO2 smoothed and adjusted (ppm) | Taux de CO2 auquel on a retiré les variations saisonnières et ajusté\n", " |\n", "| CO2 completed (ppm) | Identique à la colonne 5, les valeurs manquantes sont prises dans la colonne 7\n", " |\n", "| CO2 adjusted completed (ppm) | Identique à la colonne 6, les valeurs manquantes sont prises dans la colonne 8\n", " | \n", " \n", " " ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
YrMnDateDateCO2seasonallyfitseasonallyCO2seasonally
0adjustedadjusted fitfilledadjusted filled
1Excel[ppm][ppm][ppm][ppm][ppm][ppm]
2195801212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
3195802212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
4195803212591958.2027315.70314.44316.19314.91315.70314.44
\n", "
" ], "text/plain": [ " Yr Mn Date Date CO2 seasonally fit \\\n", "0 adjusted \n", "1 Excel [ppm] [ppm] [ppm] \n", "2 1958 01 21200 1958.0411 -99.99 -99.99 -99.99 \n", "3 1958 02 21231 1958.1260 -99.99 -99.99 -99.99 \n", "4 1958 03 21259 1958.2027 315.70 314.44 316.19 \n", "\n", " seasonally CO2 seasonally \n", "0 adjusted fit filled adjusted filled \n", "1 [ppm] [ppm] [ppm] \n", "2 -99.99 -99.99 -99.99 \n", "3 -99.99 -99.99 -99.99 \n", "4 314.91 315.70 314.44 " ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "raw_data = pd.read_csv(data_file, skiprows=54)\n", "raw_data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Les lignes 0 et 1 vont gêner l'analyse et ne contiennent que des indications sur les données. Nous les retirons. Les titres de colonnes contiennent des espaces qui gêne leur appel. Nous renommons donc également les colonnes:" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yearmonthDate1Date2CO2CO2 overallC02_3CO2_4C02_5CO2_6
2195801212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
3195802212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
4195803212591958.2027315.70314.44316.19314.91315.70314.44
5195804212901958.2877317.45315.16317.30314.99317.45315.16
6195805213201958.3699317.51314.70317.87315.07317.51314.70
\n", "
" ], "text/plain": [ " year month Date1 Date2 CO2 CO2 overall C02_3 \\\n", "2 1958 01 21200 1958.0411 -99.99 -99.99 -99.99 \n", "3 1958 02 21231 1958.1260 -99.99 -99.99 -99.99 \n", "4 1958 03 21259 1958.2027 315.70 314.44 316.19 \n", "5 1958 04 21290 1958.2877 317.45 315.16 317.30 \n", "6 1958 05 21320 1958.3699 317.51 314.70 317.87 \n", "\n", " CO2_4 C02_5 CO2_6 \n", "2 -99.99 -99.99 -99.99 \n", "3 -99.99 -99.99 -99.99 \n", "4 314.91 315.70 314.44 \n", "5 314.99 317.45 315.16 \n", "6 315.07 317.51 314.70 " ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = raw_data.drop(labels=[0,1], axis=0).copy()\n", "\n", "col_list = data.columns\n", "data.rename(columns={col_list[0]: 'year', col_list[1]: 'month', col_list[2]: 'Date1',\n", " col_list[3]: 'Date2', col_list[4]: 'CO2', col_list[5]: 'CO2 overall',\n", " col_list[6]: 'C02_3', col_list[7]: 'CO2_4', col_list[8]: 'C02_5',\n", " col_list[9]: 'CO2_6'}, inplace=True)\n", "data.head()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On converti les colonnes 'year' et 'month' en période que l'on défini ensuite comme index" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yearmonthDate1Date2CO2CO2 overallC02_3CO2_4C02_5CO2_6
period
1958-01195801212001958.0411-99.99-99.99-99.99-99.99-99.99-99.99
1958-02195802212311958.1260-99.99-99.99-99.99-99.99-99.99-99.99
1958-03195803212591958.2027315.70314.44316.19314.91315.70314.44
1958-04195804212901958.2877317.45315.16317.30314.99317.45315.16
1958-05195805213201958.3699317.51314.70317.87315.07317.51314.70
\n", "
" ], "text/plain": [ " year month Date1 Date2 CO2 CO2 overall C02_3 \\\n", "period \n", "1958-01 1958 01 21200 1958.0411 -99.99 -99.99 -99.99 \n", "1958-02 1958 02 21231 1958.1260 -99.99 -99.99 -99.99 \n", "1958-03 1958 03 21259 1958.2027 315.70 314.44 316.19 \n", "1958-04 1958 04 21290 1958.2877 317.45 315.16 317.30 \n", "1958-05 1958 05 21320 1958.3699 317.51 314.70 317.87 \n", "\n", " CO2_4 C02_5 CO2_6 \n", "period \n", "1958-01 -99.99 -99.99 -99.99 \n", "1958-02 -99.99 -99.99 -99.99 \n", "1958-03 314.91 315.70 314.44 \n", "1958-04 314.99 317.45 315.16 \n", "1958-05 315.07 317.51 314.70 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def convert_month(year,month):\n", " return pd.Period(year = int(year), month = int(month), freq='M')\n", "\n", "data['period'] = [convert_month(year,month) for year,month in zip(data['year'],data['month'])]\n", "data = data.set_index('period')\n", "data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On vérifie qu'il n'y a pas de trou dans les périodes:" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "periods = data.index\n", "for p1, p2 in zip(periods[:-1], periods[1:]):\n", " delta = p2.to_timestamp() - p1.end_time\n", " if delta > pd.Timedelta('1s'):\n", " print(p1, p2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Toutes les périodes sont bien renseignées. Quand il n'y a pas de données pour la période, la valeur -99.99 est entrée. Nous enlevons pour le moment ces valeurs. La colonne `'index'` est créée avant pour tenir compte de l'espacement irrégulier des périodes. Il faut convertir les valeurs de CO2 en données numériques:" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yearmonthDate1Date2CO2CO2 overallC02_3CO2_4C02_5CO2_6index
period
1958-03195803212591958.2027315.70314.44316.19314.91315.70314.443.0
1958-04195804212901958.2877317.45315.16317.30314.99317.45315.164.0
1958-05195805213201958.3699317.51314.70317.87315.07317.51314.705.0
1958-07195807213811958.5370315.86315.19315.86315.22315.86315.197.0
1958-08195808214121958.6219314.93316.19313.99315.29314.93316.198.0
\n", "
" ], "text/plain": [ " year month Date1 Date2 CO2 CO2 overall C02_3 \\\n", "period \n", "1958-03 1958 03 21259 1958.2027 315.70 314.44 316.19 \n", "1958-04 1958 04 21290 1958.2877 317.45 315.16 317.30 \n", "1958-05 1958 05 21320 1958.3699 317.51 314.70 317.87 \n", "1958-07 1958 07 21381 1958.5370 315.86 315.19 315.86 \n", "1958-08 1958 08 21412 1958.6219 314.93 316.19 313.99 \n", "\n", " CO2_4 C02_5 CO2_6 index \n", "period \n", "1958-03 314.91 315.70 314.44 3.0 \n", "1958-04 314.99 317.45 315.16 4.0 \n", "1958-05 315.07 317.51 314.70 5.0 \n", "1958-07 315.22 315.86 315.19 7.0 \n", "1958-08 315.29 314.93 316.19 8.0 " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data_valuesonly = data.copy()\n", "data_valuesonly['CO2'] = pd.to_numeric(data_valuesonly['CO2'])\n", "data_valuesonly['index'] = np.linspace(1,len(data_valuesonly['CO2']), len(data_valuesonly['CO2']))\n", "\n", "periods_novalue = []\n", "for i in data_valuesonly.index:\n", " if data_valuesonly['CO2'][i] == -99.99:\n", " periods_novalue.append(i)\n", "data_valuesonly = data_valuesonly.drop(periods_novalue)\n", "data_valuesonly.head()" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEKCAYAAAAcgp5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xl8XGd1+P/PmRlJo30b7ZItyZZ3O3biOHsMWYghKUkpFBNS8iVtQwtlLZTmS6EFvmkpLS0UQiFsv7RQQmhYQiAJWZ04ju3Y8RIvkqzF1r5Ls0ia0TLP749752rseJFjaz/v18sv3bmz5LmRffTo3POcR4wxKKWUmr9cMz0ApZRSU0sDvVJKzXMa6JVSap7TQK+UUvOcBnqllJrnNNArpdQ8p4FeKaXmOQ30Sik1z2mgV0qpec4z0wMA8Pl8pry8fKaHoZRSc8revXt7jDF553rdrAj05eXl7NmzZ6aHoZRSc4qInJjM6zR1o5RS89ykA72IuEVkn4g8bj/+FxGpFpGDIvJLEcmKe+19IlInIjUicstUDFwppdTknM+M/uPA0bjHTwNrjDHrgFrgPgARWQVsBVYDW4Bvi4j74gxXKaXU+ZpUoBeRUuBW4Puxc8aY3xtjxuyHO4FS+/h24GFjTMQY0wjUAZsu3pCVUkqdj8nO6L8O/A0QPcPz9wBP2MclQHPccy32OaWUUjPgnIFeRG4Duowxe8/w/OeAMeAnsVOnedkbdjcRkXtFZI+I7Onu7j6PISullDofk5nRXwO8U0SOAw8DN4jIjwFE5G7gNuD9ZmKrqhagLO79pUDbqR9qjHnQGLPRGLMxL++cZaBKKaXepHMGemPMfcaYUmNMOdZN1ueMMXeJyBbgs8A7jTFDcW95DNgqIkkiUgFUAbunYOxKKTWn/XB7I0+83j7l/50LqaP/FpAOPC0i+0XkOwDGmMPAI8AR4EngI8aY8QseqVJKzSPGGB54vo5nq7um/L91XitjjTEvAC/Yx0vP8rr7gfsvZGBKKTWfGGPY2dDHlZU5iAidgQi9gyOsKc6Y8v+2roxVSqlp8PO9Lbzvezt57IB1y3L38T4A1pVlne1tF4UGeqWUmgZH2gIAHGj2A/B8dRe+tETWl2qgV0qpeeFE7+BJX6s7gqwtycTlOl1F+sWlgV4ppaZBc/+w/XWI8aihvjtEVUH6tPy3NdArpdRFNjQyxnVffY7nqjsB60ZsS79Vhd7SP0xjzyAjY1Gq8tOmZTwa6JVS6iKr7gjS3DfMZ35+EIDXmvoJj0ZZXpDO0Mg4L9f1ALC6OHNaxqOBXimlLrJYHj4QHgXgp7ubSfd6+PBblwDw1OEOEtzCUp3RK6XU3NTQbQV6j8sKsS8d6+aty/NZXmjl5HfU97I0P51Ez/SEYA30Sil1kcUC/fDoOL2hCJ2BCMsK0ijNTnFes6Jwem7EggZ6pZS66Oq6Qs7xq8f7ASjLSSEtyUN6ksd5PF000Cul1AWq7w7hH7by8XtP9FPTGeSty62uvLsbrRWwscDudlt186VZydM2Pg30Sil1AaJRw41f28bN/7YNmLjR+ulblgOw54Qd6O20zdoSq9Km3Jc6bWM8r6ZmSimlTlbXbaVpuoIRImPjHGgeYG1JJkvyrIqagy1+kjwufGmJAHzzfRs41Brg8vLsaRujzuiVUuo8/Hp/K4/ubXEe728acI47/RFa+odZnJuKN8FNTqoV3EuzkxGxUjZZKYlcW+VzHk8HndErpdR5+PjD+wF416UliAj7mvud55r6hmj3D1OWbeXfy7KT6RscOanaZibojF4ppSapNxRxjht7rBLKfU0DlOVYgX1XYy9RM3HjdXGulYcvzZ6+G6+no4FeKaUm6XjvxK6prQPD9A2OUNsZ5B1rigB45qi1W9QqezOR8lwr4M/0jF5TN0opdQYHmgfwD49y/TKrVLLdP+w81+4Ps+d4P1EDd2wo4Wd7mjnaHsDtEqryrcVQ925eQmqSh3ddWjoj44/RQK+UUqdhjOH2B14G4MAX3kZmSgLtA2Hn+Q5/mNrOIJW+VFYWZVCUmczA0Ch5aUlOa4O0JA8f2rxkRsYfT1M3Sil1GqHImHO8o97qNrm/ZYCslARyUxNp94dp7Bmkwq6HL8nyAlCQkTT9gz0HDfRKKXUa/YOjznFHIEyHP8xThzp496WlFGV5aRsY5njvoLPwKXbjNS/dOyPjPRsN9EopdRp9QyPOcU8owrbaLsaihq2byijMSOZAywDh0agT6HPtBVFJCbMvrM6+ESml1CzQPxgX6IMjHO8dIsEtVPjSKMr0MjBkzfgr7Jn8zSsLWF6Qzl/fvGxGxns2ejNWKaVOo3XAqrBJ93roCUUIRkYpy0nB7RIKMyfSMxV5VqCvKkjnqU9ePyNjPRed0SulFHCkLcBHf7qPyNg4ANtquynJSmZ9WRY9oQiH2wIstfvXFMUF+qKM2ZeTP5UGeqWUArY++Aq/OdDGkbYAAEfbA1y6OJu8tCSqO4Kc6B1io92ILH5G73JNX8+aN0sDvVJKAYGwVU55rCtEZGyctoFhKnJT8KUnERmLAlZ6BmCR3eLg9vXFMzPY86Q5eqXUgmeMweMSxqKGWnv2HjVWz/ieuP42+elWjXxpdgq/+9h1zh6ws53O6JVSC05z3xC3/PuL1HYGAegdHGEsagCo6Qyyo85aILVxcQ75cXXxBXH5+FXFGbjnQNoGziPQi4hbRPaJyOP24xwReVpEjtlfs+Nee5+I1IlIjYjcMhUDV0qpN+tHLx+npjPIA8/XAdDab1fYJHk41hni9dYA+elJLMpNYWl+mvO+nJTEGRnvhTqfGf3HgaNxj/8WeNYYUwU8az9GRFYBW4HVwBbg2yLivjjDVUqpC3ei12ox3OG3etfEZvbXVvnoCUVoHRhyWg3HB/q5cOP1dCYV6EWkFLgV+H7c6duBh+zjh4A74s4/bIyJGGMagTpg08UZrlJKnb/67pBTNgnQZgf4zoD19VCrtd3f+rIsxqKG6o4gxfbm3d4EN9/Yup5nPrV5+gd+kUx2Rv914G+AaNy5AmNMO4D9Nd8+XwI0x72uxT6nlFLTLhQZ48avbeNvH33dORdrN9wRCDM6HuUX+1q5cWU+efbN1oGhUUqyJjYLuX19yUkz+7nmnIFeRG4Duowxeyf5maf73cac5nPvFZE9IrKnu7t7kh+tlFLnp8neLOSX+1oBK/APDI2Sk5pIeDTK3hP9BMNj3LK60NnjFSa6Uc4Hk5nRXwO8U0SOAw8DN4jIj4FOESkCsL922a9vAcri3l8KtJ36ocaYB40xG40xG/Py8i7gEpRS6sya+oZOetzYbeXnr1qSC8CzRzsBWFuSiS9tosVwcdbMbv93MZ0z0Btj7jPGlBpjyrFusj5njLkLeAy4237Z3cCv7ePHgK0ikiQiFUAVsPuij1wppSYhtrcrwMhYlFcarNLJq+1Av7OhD49LWJSTQlncln/zKdBfyIKprwCPiMifAk3AewCMMYdF5BHgCDAGfMQYM37mj1FKqalzpD3gHLf0D/E/u5qozEvlqkor0L/e6qc8NwWP20VmysTcN7ahyHxwXoHeGPMC8IJ93AvceIbX3Q/cf4FjU0qp87aroZf8DC8VvlTCo+PsaujFl5ZETyhCXVeIpr4h/uqGqpNm7LFSSoDiTC8etwtvwvypCteVsUqpeaM3FOG9D+5k64OvALCzoZeuYISP3bgUgF2NfUQNVPpS8Sa4ybVvvi7OnQj0z336LTz9qdnZbvjN0kCvlJo39p7oB6AzEMEYw7HOEABb1hQiAjvqewGotHvIpyRZs/bFORNpGm+CmyTP/JnNgwZ6pdQ8ElsAZR1HONYVxJeWRH66l9zUJI7a+fpY/t0tctLj+Uq7Vyql5o3OwESnyYbuEMe6QlTlT2wW0hOKkJ+eRLo3AYBP37Kcl2p72Lx8fpd464xeKTVnHWr1s6O+x3kcP6Ov7w5R1xlyVrTGNguJn73ftq6Yf373OhLc8zsUzu+rU0rNa7d9czt3fm+X08fmWFeIy8uz8biEfU0DBCNjVBXYgd5uMVyZN3dbGbxZGuiVUnNCVyDMttqJdinh0YnlOXtP9OMfHuVIW4BLF2VTkOHlJbunfGxGn5Jo3WBdVZwxjaOeHTRHr5SaE+76wS5qO0Mc/Ie3keFN4GCL33nucGuAA81+Rsaj3LquiD0n+p0KnFig/8DV5eSkJnLnpkUzMv6ZpDN6pdScUGuXSh5oHgCsGnmABLdQ0xmkpiNAcaaXdaVZTj4+yeMiz+5fU5KVzIc2L5kzu0JdTBrolVKzXnyaprYzhDGGX+9v5fLybFYUZtAdjNDQM8gSe/ZekWvdcPWlJSGy8AL7qTTQK6VmveO9E43J2gaGOdQaoL57kHddWoovLZGeUIT6rhBL7ButsQVR49E3dEhfkDTQK6VmvZqOoHPcNjDMgRYrfXNdlQ9fWhKH2wIMjow7Af6qJbkkelzcvqF4RsY72+jNWKXUrNMVDPOD7Y188qZleBPcvHain5RENxsWZdE2MExD9yDJCW6KM5PxpU/0kI/N6Isyk6n58hZN29h0Rq+UmnW+8kQ1393WwK/sXaEOtwVYU5zJopwUWgeGqe8OUeFLxWX3kY+JzegBDfJxNNArpWad2k4rVRMrkTzeO0SFL5XizGR6QiMcaQ84N15XFKY774stilIn00CvlJpR0ajhxztPEAiPAtYN1FjXyeb+IUKRMXpCERb7UijJtnrIdwcjVNqtDFYWZVDhS+Wty/N0Fn8GmqNXSs2o11v9/N2vDvH7I5381z2bONE7SGQsCkBz37Azu6/0pZGVkuC8Lzaj9ya4ee6vN2uQPwud0SulZlTbwDAAL9stC2IVNm9Znke7f5j9TVaFzdrSTEridoVaovn4SdNAr5SaUS39VqA3xqp5P9TmRwTeujyfqIFttd1keD0UZ3qdFa8w/3vIX0wa6JVS02Y8avjJrhM09w0552KLoaIGguFRfvlaK9cs8TldJ1+u62FxbioiQoLbxcqiDLasLiQlUTPPk6X/p5RS0+bxg2187peHuKoyl5/eeyX+4VF+sqvJeX5/8wBt/jB/dl0lZdlW2eRY1LAobk/X3370WjRTc350Rq+Umjb1XVY1zZ4TfUSjhv+xg/wfXGKtYH2uuguAqoI0ijK9TkBfHFcr73KJ5uTPkwZ6pdSU+d6LE4ueANr81g5Qo+OGjkCYl+t6WJqfxhffuRqA52OBPj8dj9uFyw7o8Yui1PnT1I1Sakr0DY5w/++OAnDHhhIA2v3DzvMN3YPUdAbZvCyP7JQEUhPdHO8dIt3roSDDamvg9bgYHBnXQH+BdEavlJoSr7dObAwSCI9ijKG6Pchli7MBa0/X7mCEyjzrRmuZHcyr8tOc1MzPPnQV1y/LY21p5vRfwDyigV4pNSVi9fEATb1DNPQM0js4wpbVhQDssdsbxBqRlWbHAv1ES4M1JZn81z2bSPdOLJRS508DvVJqSsQH+taBYV5t7APghpX5JLiFPcetx7GFT/l2uqYsJxl1cWmOXil1Ufzv3hbqu0N8+m3LcbuEo+0B0r0eguExWvqHOdzmx5eWSKUvleyURNr9YdwuYVGOFeg/dfMyBoZGuHWd9pC/2DTQK6UuWCgyxqd/fgCATRU5bCjLYlttNx+4qpyf7m6itX+YV4/3sXFxDiJCbloSXcEIZdnJJHqsxIIvLYlvv/+ymbyMeeucqRsR8YrIbhE5ICKHReSL9vn1IrJTRPaLyB4R2RT3nvtEpE5EakTklqm8AKXUzGuPS9PUd4V45mgXo+OG29cXU5qdzNH2AM19w6xflAVYN1wBKu38vJpak8nRR4AbjDGXAOuBLSJyJfBV4IvGmPXAF+zHiMgqYCuwGtgCfFtE3FMxeKXUzNh7op+Hdhx3+tO02/XxAPXdg9R2Bkn0uFhTbDUie6WhF4Bye9PuS+2AH3usptY5UzfG+k6G7IcJ9h9j/8mwz2cCbfbx7cDDxpgI0CgidcAm4JWLOG6l1Az6o//cAcCNK/MpzU5x6uNzUhNpGxgmyeNicU4KLpc4PeQByn1WZc0fX15GRnICm5flTf/gF6BJ5ejtGfleYCnwgDFml4h8AnhKRP4V6zeDq+2XlwA7497eYp879TPvBe4FWLRo0Zu+AKXU9IpGjXN8qNVPaXYKz1V3kZOayLrSTDr8YYyBxfZsPVY2CRMz+JRED++6tHR6B76ATaq80hgzbqdoSoFNIrIG+Evgk8aYMuCTwA/sl5+uCYV5wwljHjTGbDTGbMzL05/qSs0VfUMjzvGxzhDGGF6p7+WW1YWUZCXTGQhzom+QcrsRWayHvEusTULU9DuvOnpjzADwAlbu/W7gF/ZTP8dKz4A1gy+Le1spE2kdpdQcY4zhaHvAycd3BSLOc23+MF3BCIHwGCuL0slP99I/NEp4NMpiu198bEHUnVfob+4zZTJVN3kikmUfJwM3AdVYwXuz/bIbgGP28WPAVhFJEpEKoArYfbEHrpSaHk8e6uDt33iJ/3i2DoCmvkHnubaBYY60BQBYVpDu9KgBnBn9quIMfv/J6/ny7WumcdQq3mRy9EXAQ3ae3gU8Yox5XEQGgG+IiAcIY+fbjTGHReQR4AgwBnzEGDM+NcNXSk21p490AlZrYYADLX48LuHaKh+t/cPsPt6HxyVcUprF8MjEP/XFORMVNcsK0lEzZzJVNweBDac5vx047eoGY8z9wP0XPDql1IyrsTfnju0K9WJtN2tLM6nwpfJqYx/V7QGqCtJJTnSTlz4xoy/O8p7289T00143SqkzMsbQ1GsF+Jb+YVoHhjncFuDWtUWUZCUzODLOobaAk6YpyJgI7h63hpfZQlsgKKXOqKV/mGBkjEpfKg09g+yo6wGsrpJ9g1b1TXcw4vSL96Ul8t6NZfzRZVo6OZvoj1yllKMnFOGKf3yGbzxj1Vb8al8rInDXlYsB2G4H+mUF6RRnTSyEKrUDvYjwz+9ex6aKnGkeuTobDfRKKcdzR7voDET492dqAdjV2MeKwgxns5Dtx3rwpSWRk5p4Ug5ed4Ca3TTQK7WAhUfHGY9b6bqvecA5DkXG2NfUz8bF2U7uvXdwhGUFVl28L3XixqsG+tlNA71SC5QxhhWff5JPPbLfOVfXFXSOnzzUweDIOBvLs8lPTyLRvrkaK5V0uYQKX6zNgW4WMptpoFdqgWqzO07+er+1cD0UGeNoe9BpIRy78bq6OBOXS3C7rO4m8TXxj3zoKg598RYStMJmVtPvjlIL1OG4zbuHRsb49f5WQpExPrR5CQAH7edjs/XL7Rusa0oynPflpSeRlqTFe7OdfoeUWqCer+l2jtsGwjx7tIvy3BRuWpkPQF1XCF9aotOI7OvvXU9L/xBrSzJnZLzqzdNAr9QCEB4d567v70IEfvxnV5DgcvH0kU7y0pPoDkZoHRimrivEJWVZZHgT8LiEsahxOk+C1Ws+JzVxBq9CvVmaulFqHvr2C3VOjxqwdoTac6KfV4/3U9MRpKYzSE8owgfs+vjjPYO09A9R4UvF5RKnyqZEb7LOCxrolZpnAuFRvvpkDX/+X3sYjIwB1gYhMc19w1R3WB0nb1xZgNslvFLfS9RApV1Fc0WllY/PS0tCzX0a6JWaZ2K9aWBiJeuhtgDpXitT29I/xJG2AAluoaogjcIMLy/bryu3A/3b1xQBkJGcMJ1DV1NEA71Sc1w0augKTGzOfbx3ol/86y3WTP5Qq5+rKnPJTE6guX+Il471cNnibBLcLkqykgnaM/8Ke6u/m1bm8/0PbHQqcNTcpoFeqTnuhy83sukfn6Wmw1rstLuxj+QENxW+VI60B9hR30NjzyBXLcmlLCeZuq4QNZ1BNlXkAhPthHNTE8lMsWbwIsJNqwq0dHKe0ECv1Bz36GutAPxqv/V1R30vV1bmUJ6bQmcgzI66XtwuYevliyjNSmFnQx/GwFJ7YVTshmuZtjGYtzTQKzWHGWNo6bdy8rUdQcKj4zR0h1hbmuWUTh5pD7A0L43kRPdJrQpiN15LsqwAn6n5+HlLfy9Tag5r94cJhq38ek1nkJqOIFEDq4rSGY9G6R0coaYjyIZFWcDJs/bKPCvQv3VFHretK+ITN1VN/wWoaaEzeqXmkJGxKP+7t4XR8SiAk5ffVJ5Duz/MYXuj7pVFGfjSkhiPGloHhqnMs9I08TP6lERrnleUmcy37ryUpfm6r+t8pYFeqTnkG8/W8umfH+C3B9uBif1cNy/PYzxq2FbbRWqim7LslJP2b42laZbYAX+j3V9eLQyaulFqDjnQbJVLNvZYJZSHWv0UZXpZWWTNxp+r7mJdaRYul+CLW+wUaydc7kvldx+7znmsFgad0Ss1S0XGxvmTH+xi+7Ee51yvvU/rkfYAxhh2NvRyRUWOs63f6Lhxgn78jL4ibyKwryrOIDnRPR2XoGYJDfRKzVKv1Pfy0rEe7vrBLudc28AwAEfbA3QHI/SERlhflkVR5kTufVWR1V0yPtBneLWiZiHT1I1Ss9QzRyeako1HDV3BMP7hUbJSEmjpH3ZuvFbkpZHhnfinHJvRpyd5yE5J4D0by6Z34GrW0Rm9UrOAMYYv/eYIvznQ5pzbVjvRL76lf8i5AXvPNRUA7GzsBaA8NwURcV67ssjaGEREeO3zN3Pf21dM+fjV7KYzeqVmgVeP9/PDlxvhZbhtXRH+4VGa+4a5cUU+z1Z3cbx3iO11PSzJS2VjuVUxs7uxD49LnJ7xj/3VNSR53M5GIcBJPwDUwqUzeqVmgfrukHPcFYxwxE7LvHWFtdtT28Awuxv7uHqJz+kVv69pgNLsZDz2fq3rSrNYXqi18OqNNNArNQu09g87x/VdISf/Hgv0z1V3MTQyzlVLcsmPu8m6OFfLJNW5nTPQi4hXRHaLyAEROSwiX4x77qMiUmOf/2rc+ftEpM5+7papGrxSc9V/7zzB915scB63DQzjsrMs9T2DHG7zU5jhpSQrmczkBF6pt/LxywvTSUvykGynZ8pztRGZOrfJ5OgjwA3GmJCIJADbReQJIBm4HVhnjImISD6AiKwCtgKrgWLgGRFZZowZn5pLUGpu8Q+P8vlfHQLg7qvLSfS4ONwW4MrKXF5p6KU7GOFwW4DVxdZN1YKMJGo7Q4hYLQxEBI9bYFRn9GpyzjmjN5ZYAjHB/mOAvwS+YoyJ2K/rsl9zO/CwMSZijGkE6oBNF33kSs1Rh9v8Jx33DY5Q0xnkmqU+slMSaekfor47FBforZx8cWYySR5rJh9rZBa/EEqpM5lUjl5E3CKyH+gCnjbG7AKWAdeJyC4R2SYil9svLwGa497eYp9TakH65rPHWPn5JxmzG5EdbQ86z9V3D7K7sQ+AKypyyE1NZEedtX/rqmJr4VN+uhXolxWkOe8rzvQ671HqXCYV6I0x48aY9UApsElE1mClfbKBK4HPAI+IVct1unouc+oJEblXRPaIyJ7u7u7TvEWp2a+6w2pFEPNaUz93PPAy3cGIc+5rT9cyPDrOq8f7rfe0B8hKScDjEhq6Q+xs6MWb4GJdaRa+tCQ67G0BYzP63LREAFbY9fEAD997FY/+5dVOB0qlzua8qm6MMQPAC8AWrJn6L+zUzm4gCvjs8/FL8UqBtlM+CmPMg8aYjcaYjXl5eW9y+ErNnO3Hetjy9Zd4ZM/EL7D//EQ1+5sH+O+dJwCrrXBMdUfA/hpkbUkmpdnJNPdbZZOXLc4m0eOi0J6pJ3lcTkvhm1cVcM3SXO7ctMj5rEW5KVymHSjVJE2m6iZPRLLs42TgJqAa+BVwg31+GZAI9ACPAVtFJElEKoAqYPfUDF+pmfNSnfWb6EF7A25jDLV22+DmPmvXp2NdE2majkCYrkCYI+0BLim1+tM09w1R1xViTYmVpom1E85OSXQWO11ensNP/uxK3epPvWmT+b2vCHhIRNxYPxgeMcY8LiKJwA9F5BAwAtxtrN9hD4vII8ARYAz4iFbcqPmovsuqUWi1G4219A/TPzQKQKedfoktfALo8Id56VgP41HDreuKaPeHefS1FgAq7OqZJfY+rn12l0qlLoZzBnpjzEFgw2nOjwB3neE99wP3X/DolJpF+gdHSPd6nJWoDXZP+PYBK6jHZval2clOoD/aHiQ5wc2q4gza/WEaewZxibUBSHGW1/nsRXY9/HVVPtKSPHzgqsXTdl1q/tOVsUpNwvDIOBu+/DRffvwIAGPjUZp6rfRM7ObpwdYBEt0uNi/LoysQcfrFryxKpyQrmQ5/mMbeQcpyUkj0uJwe8gDl9ow+3ZvA/i/czGduWT7NV6jmMw30Sk3CwZYBAB56xbrJ2tw/zFjUUOlLxT88Snh0nIPNflYWpVOWk0IwMsbBFj9H2gPcsaGEokwvHYEwjd2DTlAvypyY0RdmTBx73C5tRqYuKg30Sp3Gsc6gk34BnFp3sG66NvZY+fkrl+QCVguDQ61+1pZmUpBh9aJ50W4zfNnibAozvYyMRTnSHnC28SuJm9G7XBrY1dTRQK/UKVr6h7j531/kfQ/udM69ENcbvm9whOoOq5rmmiU+AHY19hGMjLGuNIsCe4HTS8d6nHx8/Ow9FugrfKncc00F//NnV0z5NamFTVdbKHWK1+2bqg09g3QFwyS53exr6mdtSSavt/pp6BnkQPMA5bkpzmrVZ45Yu0GtLckkwb5Zu/t4HxW+VLwJbqeNAVgbdIOVovnCH6yazktTC5TO6NWC91pTP08e6nAet8S1DK7pCLLnRB9RA3deYS1Yqm4P8NKxHq6szKXAnqlvr+vB7RKW5KU5i54Aquxyyfg9XSu0EZmaZhro1YL3rm/v4C9+vJcOv5WTb+kfcp6r6QhyoMWPS+CmlQUAPHW4k6GRcW5dV0R6koeURDeRsSiL7GqatKT4/VuttgXxG3XHl1UqNR000KsFrSc00ZPmtaZ+jDHsauzj8vJsUhLdtA2EOdLmZ0leGnnpSWR4PexssHrDL8lLQ0Scipklp+kkeUWl1XTM7RLet6mMf3n3OqcOX6npon/j1IIWW90K0NAd4tmjXVR3BLljQwkFGV46A2FqOoPOFn3FWcmMRY3Vl8YO8LHZemXeRHfJm1Zd9xnHAAAgAElEQVTmk+H1cHn5RHfJf3rXOt6zMb4NlFLTQwO9WlCa+4b42atNzuNj8YG+Z5CX63vwJrh478YyCjKSqO8O0dw3zAo70Mfy74tzUpySyFQ7VRPrUwPwnbsuY8/f3ezcmFVqJmnVjVpQPvGz/ew90c8lZVmsKMzgNwfaKMtJJj/dS2v/MN3BCFX56Xjc1ox9Z4NVP7+swAr0sTLJ8rig/sFryslKTuBGO4cPaHpGzSoa6NWCctzuT7P9WA9V+ensax7g7qsW0xWM8FpTP6NjhquXWougCuKqZ1YU2r3hU600TXHcc9dV5XFdlbbaVrOXTjvUvBUeHedX+1qJRq2NQcajhkDY6i55oneI5r4hRsaiVOWnU5yVTHPfMB2BMFX51uw9tvAJcHrDX70kl7KcZO68QpuOqblDZ/Rq3nrg+Tq++VwdHrdw27pi2gaGGR23gn6T3QcerNbAkfGJDUJite/xm4DE8vFXL/Xx0t/cMJ2XodQF0xm9mrcO2Ctcd9l59hN2t8mslASa+6wNuAGW5qdRElfbXmWvdl1qB/wPbV4ybWNWaipooFfzQkv/EOV/+1unFQFAq73wqdHOyx+wO1C+fU0hLf3D1HaGyEtPIjM54aSWwaXZVm/4ZQXpHPyHt/Gpm5dN12UoNSU00Kt54df7rW2JH3ihDrA6TMZaGZzoswL9tppuqvLTWFuSxch4lB31Pc4ip1hwX1mUgTuuk2SGN2HarkGpqaI5ejUn7ajvYWleGvn2oqXYatVgeAyAzkCEyFiUdK+HtoEwJ3oH2X28j89uWcEie+/Vdn+YG1fmA5CW5GH3/72R7NTEGbgapaaWzujVnFPTEeTO7+3ir366D7Bm7/ubrLRMS/+Qs7MTwB3rSxiPGp46bDUtu6Iyxwn0YLUxiMnP8OoCJzUv6d9qNee8Ut8DWJuBGGPoCkYIRsao9KUSHo3SOzjCi7Xd5KQm8o61RQA8eagDEVhZmHFSU7H4QK/UfKWBXs164dFxImPjzuPWgYk2wt3BCA3dVg7+uiprE5DmviFePNbNtUt9ziYfrzUNUJadQnKi+6RVq7HKGqXmMw30atb7w2/v4B3feMl5HB/o67sHqekIAPCWFVa+/ekjnfSERti8LI/89CSSPNZf8/jukj/98yv5442lJ+38pNR8pTdj1awWCI9ytN0K5M19Q5TlpFDbGaIqP41jXSGa+4c40OInPz2JyxZnA/CEvYnIpoocXC6hJDuZhu7Bk9I0Vy3J5Sp7v1el5jud0atZJRo1jMWtUq2x92YFqO8O0eEPU9cV4vb1xQB0BcK8Ut/LxvJsMrwJZHg9NPYM4k1wOZtvJye4AWsFrFILkQZ6Nat8+ucHuPKfnsUYq1VBW1ya5njPIC/XWTdib1hRQIbXw67GPjoCYa63m4rF6uErfGlO24LNy2LPTSyKUmoh0dSNmjVGx6P8Yl8rYK1mrcxLo23A2t7PJXC8d4jA8Ci5qYmsKEwnP8PLjnqrjHKFvWVfaXYyR9oDJ+XjP3XzMtaXZXHtUt80X5FSs4PO6NWM2Xuin4/9dB+jdqqmun0iTbPProtv6hskKyWBVcUZNNobg1y91IfLJeSnJzFud6aMVdcU2Auo4m+yetwu3ra6EJGJFa9KLSQa6NWMed/3dvLYgTZnVr6vud95rs5uOLbneD+XlGZR4UvjUKufzkCE1cXW7D3f3sLPl5ZIZrLVquB9mxaR7vXwluX503kpSs1qGujVtInNvgEiY+OMjFkz+b3Hre6S+5sGyEtPYml+GnVdIXpDEY51hbiiMofiLC+9gyMATgOy2Oy9Im63p1XFGRz8+7dxjaZplHKcM9CLiFdEdovIARE5LCJfPOX5T4uIERFf3Ln7RKRORGpE5JapGLiaW15v8bP8757guWqru2Rtx8RerW1+Kw+/r3mA9WVZlGQl0+EPs7vR+gFwRUUuPntnJ8BpKZyZYs3ic+OeAzRFo9QpJjOjjwA3GGMuAdYDW0TkSgARKQNuBpzdlkVkFbAVWA1sAb4tIu6LPXA1t/xyXytjUcMPtx8H4PVWq1e8Ly2JtoFhBoZGaOwZZMOiLAozvHQEwuxq7MOb4GJtSSY5cc3GYpU1a0syAbjzikXTezFKzTHnDPTGEpt+Jdh/Yr+D/zvwN3GPAW4HHjbGRIwxjUAdsOniDVnNRS/UdAHgH7a28jvU5ifD6+GKihza/WFnk5D1pVkUZHrpCUXY1djHpYuySfS4yE2bCPSx3Px1VXnU/L8tXL9M92tV6mwmlaMXEbeI7Ae6gKeNMbtE5J1AqzHmwCkvLwGa4x632OdO/cx7RWSPiOzp7u5+k8NXs1VDd4jwqNWfpjsYocHe/CPWvuBQq581JZmUZCfTOjDMvqZ+RGBtaSYFGUkYA0fbA6wsit14tdI1i3JSTkrNJHn0l0WlzmVSgd4YM26MWQ+UAptEZB3wOeALp3n56RKk5g0njHnQGLPRGLMxL09nZPNJQ3eIG762jQ/+6FXAah0McOmiLPoGRwiER6luD7K2JJOiTC8jY1G2H+uhIjeVdG8ChRkTpZGxpmMri9J54M5LefQvr57+C1JqjjuvqhtjzADwAlZ6pgI4ICLHsX4AvCYihVgz+LK4t5UCbRdjsGpuiPWaecXuCR/b6WlThdVb5oWabkbGo6wuyXQqaPac6HeCesFpAr2IcOu6IvLST77xqpQ6t8lU3eSJSJZ9nAzcBOwzxuQbY8qNMeVYwf1SY0wH8BiwVUSSRKQCqAJ2T9kVqBn3y30t7LB7xAPsbx5wjoPhUV5p6MUl1qYfgLMJyNqSTIozJ9oSVNpNxwrjFjst1X7xSl2wybRAKAIesitnXMAjxpjHz/RiY8xhEXkEOAKMAR8xxoyf6fVqbtvV0Msnf2bdpqn+8ha8CW5eb/GT6HExMhalqW+IJ15v57Z1xSwrSAfg94c7SEvysDgnhf6hEeezYm0LclISSU5wk5OaqFv7KXURnDPQG2MOAhvO8ZryUx7fD9x/QSNTs9LwyDgGQ0qi9Vdnz4mJ1awneofITkmgIxDmXRtK+MW+Vg40++kfGmVdaSYF6Um4XcLouOHSRRm4XHJS2WSsu6TLJez9/E1ERqMopS6croxV5+WPv/sKm//lBae7ZHVcG+G6rhC77EVOt11ibeH3vF1WuSQ/DY/b5ZRGrrFr4EWE1ESrcqYqro1wSqJHZ/NKXSTavVJNmn9o1FnoVNsZYnlhOtXtAa6szGFnQx/N/UPsa+qnKNPL5mX5ZKUksK3GKp1d4rNn63ZpZGyxE8CjH74aj0tI9yZM8xUptTDojF5N2hF7pyeA2s4g4dFxGnoGubw8h/QkD+0DwxxtD3Lp4mzcLqEww8vIeJREj4sSuxf8B68p59JFWbw1runYisIMluanT/v1KLVQaKBXZ7T3RD8P7TjuPK7rOjlN8/jBdsajhk0VORRleWnoGaS5f8hJweTHmo7lpuK2NwH5s+sq+cWHr3H61Cilpp6mbtRpRaOGP/rPHQBcs9TndJRMS/KQluShdWCYlv5hfGlJXLvUR1FmMttqrTRNrLqmwM7HL8lPPf1/RCk1LXRGrwDY2dDLLf/+Ih12J8mmviHnuViNfF13iCX5aRRkJNEZCHOsK8jywjREhOKsidr3WNuC3DQr0Me3EVZKTT8N9AtUdUfAqZwB+M8X6qnpDPLfO48DcNC+6QpWPj4aNdR0hFial0ZeupeuQIRjnSFn9l5kL3xyu4RFOVZ3yTs2FHPzqgLesbZomq5KKXU6GugXgPDo+ElB/bEDbWz5+kv85mC7c67e3tGpqc9qV/BqYx8piW7WlWZS1xViR30vPaEI1y/zkZ+RRE1nkOHRcSfQx1oZuEWcfPyKwgy+94GNrC6eqLBRSk0/DfTz3PDIOCs+/yRfeaLaOfc7O8C/Ym/hNxgZc/rRNNspm92NfVy2OJvFuam0DYQ50GK1NXjrinxKsyfaFiwrsG68blycDcB7NpZO8RUppc6XBvp57qAdoL/7YoNzrtaunnm91XouFsR9aYm09A/TPzhCTWeQKytzKc700uEPU9cVojDDS4Y3wcnBA05ZZLkvlV3/90Y+d+vKabkupdTkaaCf5+IbjMVaBB+3e8PXdASJjI3z9JFOvAku/nhjmbPhB8CGRVlWG+HxKLsb+5xOkuviFjvFNuUGq+tkrDWCUmr20H+V81x8oK/pCOIfHiVqYOvlZTz8ajMt/cO81mTt1bq80JqdP33E2td1dVEmwfAYYG0YcvOqAsCqpvnrm5cxbt6wzYBSahbSGf088+v9reyNazS2r2mADYuyAGjqG+RImx+XwNvtSpgTvYMcbQtwSVmWk3v//eEOSrKSyUxJOKmN8JK4XjQfvbGKT9y0bDouSSl1gTTQzyP13SE+/vB+Pvgjq/1/u3+YjkCYd6wpwiXWBiB13SEW56ZSkWvVtj97tIuR8SjrS7MoszfdDkbGWFVs5eHj6+O1N7xSc5MG+jmsvjvEvqaJ2fuRNqsXTSA8RjRq2N9kpW02lmdTlJlMc98Qh9sCLM1PoyDTWsz0pL0b1PpFWSft3rTKvuEa30Z4ab4GeqXmIg30c9jWB3fyh9/eQU8oAkBnIOw8d7x3kP3NAyS6XawqzqAsJ5nna7o50TvEdVU+kjxufGlJ9A6OkJ+eRGGG96RNt2MzehHhz6+rYHlBOr40bRus1FykgX6OuPe/9nDP//cqo+PWZhzGGLqDVoB/1a6SidXCg1Uyua9pgFXFGSR53JRmp+AfHiXR7eLta6z8fCwtc0lZlhPkP3DVYutcaZbzWZ+7dRVPfuK6k34QKKXmDg30s9BgZIwf7zzBeNSqavEPjfL7I508V93lpFrqukLO66s7ggyNjPHzPc0sL0jH4xKqO4K83upnfZkVsGP595VF6U6KJtUuhYy9BuBLt6/hwBfedtK+rYAGeaXmMA30s9C/PFXD3/3qEM8etcocD7VN9J05aveE/9+9LXhcQkqim4aeQXY19jE4Ms5Hb1xKSXYy22q6GR4ddzb42Lw8D4ANi7Kdz7pjQzG+tCTeYj8Xoy2ElZpftI5+FordVH2loZe3rS7kYIsV6LNTEjjRO4Qxhl/vb+Mty/MJhEfp9IfZUddDosfFTSsL+Nmrzbx0zOo4GVvFur4si8c/ei3lcZ0k33v5It57+aJpvjql1HTTGf0sMDI2sQm2McZpURDbhu/11gEW5aRwSVkWjT2DNPUN0REIs3l5HkWZXtoDw+yo7+WyRdl4E9yU2d0jE9xyUqXMmpJM0pL0Z7tSC40G+hlW3x1i/Zd+zwPP1wHQ7g8zMDSKLy2Jhp5BRsaivN7qZ21pJuW5qRzvHWS3ffN1U3kOhRleWvqHqekIOgujYm2CCzK8JHr0W6zUQqdRYBpFxsZZ+w9P8cPtjc65X7zWwtDION97yWo6dthO29ywwsqb13YGae4bZm1JJuW5KQyNjPPEoQ4ykxOoyk+jMNOLMTAWNU4Lg1jL4KJTbqgqpRYmDfRTqDMQZmBoxHn8Ym0PwfAYX3r8iHPuQLOVfw+GxwiPjvNaUz8icF2VFeifr+4CrEZisfz6c9VdbFycjcvegDsmFuivXerjtnVFfGPrhqm9QKXUnKAJ2ykyNDLGFf/4LGlJHvZ/4WY8bheH7F2bvAnWz9do1HC4zU9KopuhkXGO9w7yyKvNvHV5PuWxFgV2oF9dkkmvvTAKYGN5DgAFcbP2Sp+Vj89JTeRbd1469ReplJoTdEY/RfYct1oThCJj1Nm7NzXY7YHDo1H8Q6Psbeqnf2iUd19mbdbxamMfvYMj3LSywKlj3988QIUvlczkBErtWniATRVWmWR80zHNxyulTkcjwxQ5FregqaF7kLHxKK/FdZVs6hvitwfbSfS4uOtKazWqM3svzsCXlkiWXc++xq6FT/S4WFWUQYbX45wryEjiy7evZttn3jIdl6WUmoM00J9Dc98Qf/zdV07qI7OttputD75CKGL1ajfG8I+/O8q//b7G2Zu1uj1AcoIbgPquEN99sYHWgWE+dmMVYPWieeJQO5uX5VHpS8UlsMPe2q8yLxURwZdmrWCN3+jjsb+6ht2fu4kkj/XZIsKfXFXO4tyJ+nillIp3zkAvIl4R2S0iB0TksIh80T7/LyJSLSIHReSXIpIV9577RKRORGpE5JapvICp9oPtjexu7ON/djU5577yRDU7G/p4dG8LYJVEPvhiA//xXB0v1/VijGFHfS/XL/PhS0ukzT/MzoZelhWkce/1lYC1uUdnIMKW1YV43C7y072MjEXJSU0k3WvN5C+1yyVjN1kBPG4XXvsHiFJKTcZkZvQR4AZjzCXAemCLiFwJPA2sMcasA2qB+wBEZBWwFVgNbAG+LSJzIjKNRw0/3N5IV9zsvd1vNQprtPPrg5Ex6u2ce+xrTWfQef0Th9rpDERoHRjmyspcijKTaRsIc6wz5CxYyk1NdHZxWltqzdZjOflYDTzA3//Bav7fHWu4dqlvqi5ZKbUAnDPQG0ss4Zxg/zHGmN8bY8bs8zuBUvv4duBhY0zEGNMI1AGbLvK4L4r/ePYYL9f1OI9/f7iDLz1+hI8/vN85FwvwrQNWwH+hpttZydrUNwTAy8d6cAlU5afR2DPoVNesK82kOMtLXVeIjkCYJfbGHWU5KQyPjuNxiVNdE6t5j994OzXJw11XLsbl0oZiSqk3b1I5ehFxi8h+oAt42hiz65SX3AM8YR+XAM1xz7XY52bUo3tbeGTPxLAGI2P829O1vP/7E5cSuxl6qM2PMYZo1HCi1wrmrXYL4KcOd5CTmsgtqwto6rP6zjx2oI23rSqkMi+VnlCEQ21+RKygXZSZ7PyQqLTr4GOz9nJfqlMpE5vRX1I6kY9XSqmLYVJ19MaYcWC9nYf/pYisMcYcAhCRzwFjwE/sl59u+vmGXaRF5F7gXoBFi6a+sdZf//wAALesLiQzOcHpAglWysbtEg7YG2kHw2Oc6B0iagyRsSiFGV46g2GGRsZ4vrqLd6wtIis1geeru2kdGKYrGOGqJbkc6wqyu7GPQ60BluSlkZLoOWkrvtiCp7IcqySyMq7B2GduWc6Vlblv6CSplFIX6ryqbowxA8ALWLl3RORu4Dbg/SZWbmLN4Mvi3lYKtJ3msx40xmw0xmzMy5va4Ba/0CgW4PfZ2+yBVQETDI9S1x3i6iW5gLWJxzN2m+B3X1aKMbD9WA/ByBhXL81lUU4KI+NRfnuwHYANi7LwpSXRPzTK9rpuLrPbARfF1bnH0jQrCq30zKaKHOe5lEQPt6wudKpplFLqYplM1U1erKJGRJKBm4BqEdkCfBZ4pzFmKO4tjwFbRSRJRCqAKmD3xR/6mXUFrLa9MbWdEzXtsRuou+zGYAA1HUEONPsxBm5eVWB9RjDMD7Y3sqk8xwn+z9mpnRWFGU765ZE9zaQneVhVlOH0mAmPRnn72kJgou8MQHKiFcRvXVvE05+8nj+9tuLiXrhSSp3GZGb0RcDzInIQeBUrR/848C0gHXhaRPaLyHcAjDGHgUeAI8CTwEfs1M+0+dQjB7jz+7vYUW8F++O9g85zjd2DRKOGPSf6eIcdjJv6hnjqcAfeBBe3rrW22TvY4qczEOHWdUWUZFvB+pmjXSS4hQpfqjNTr+8e5PKKHDxuF2uKJ/LrsfbAa0vemHN3uYSqgnTdtUkpNS3OmaM3xhwE3tAdyxiz9CzvuR+4/8KGNjn/9LujvFDTzaMfvpq0JA/h0XG227P5I20Brl7io6YjSJLHRWGml/ZAmOqOIANDo9ywooDtx3po7R9me10P1yzxkZ/hJSXRzbPVE+WPRZnJiEBPKMKKwnQS7c+KubLSSsFUFUz0fo/9IEj0uPjNX11LcqKuTVNKzYw5H32++2IDNZ1BZ4Psw3Hb7rX0DxONGp481MF1VT6KMr10+sNsq7U29Li+ykdJdgoHW/009gw6jcIW56bS3DeMS2BlYQaJHhcF6VZgX2WXP8Zv4HFFhZXaSXC7+PiNVdy6tgh3XEnk2tJMluZPLHpSSqnpNKcD/ej4xM5MOxus9gEPPF+PxyXkpibS3DfE4bYAHYEwb19TRGGGl45AmG21XawsyiA/w0tJlteptrnCnpnHNsuuzEtz8uopSdbXFUUTAfv6ZXksL0hndfFE7fsnb17GA+/XzpFKqdljTrcp7vBPrGDd2dBLZGycl+t6eO/lZXQFIzT1DvFsdSci8JbledR2BWkdGKYzEOb/XF0OQIl9szQl0e3k0zeUZfHT3U3O7B2smviG7kFWFU3k3B/64OWaZ1dKzXpzekbfE4rgTXBx2eJsDrUF2NnQR2QsynVVeZRmJ9PcP8Rz1V2sL8siNy2JwgxrN6bRceN0fyy0c+krCtNJcFv/O2Jb8q2Km6l/7T2X8J27LuUquwIH0CCvlJoT5nSg37Aom6Nf2sInbqpiPGr43ovWdnwby7Mpy7a23TvY4ufGFfnAyVvrxWrZF+daZZJ/uGFi8W5VQTr/8b4NvG/TxEIub4KbLWtOzr0rpdRcMKdTN2DNqpcVWHnz7XU9lGYn40tLOqnj46WLrcVLBXHb7lXmWYuXblldyG8/di2ri08ug3znJcVTPXSllJoWc3pGH5OXlkSinXapsuvXLylzuiZTYbcaiNW2A06axu2SNwR5pZSaT+b8jB6sBUi5aYm0+8PO7D4tyUOCWxgdN05pZLo3gV9++GrNrSulFpR5EegBPrtlBdUdQe65ttw5t/2zN9DUN3RSm98Ndg8apZRaKOZNoL9jwxs7IRdkeE/Kyyul1EI0L3L0SimlzkwDvVJKzXMa6JVSap7TQK+UUvOcBnqllJrnNNArpdQ8p4FeKaXmOQ30Sik1z4kxZqbHgIh0Aycu4kcuApou4uedKhPwn/NVb56O/+x0/Oc2169Bx392sfEvNsbknevFsyLQX2wi0j2Zi7+Az3/QGHPvFH6+jv/sn6/jP/d/Y05fg47/nJ9/XuOfr6mbgSn+/N9M8efr+M9Ox39uc/0adPxnd17jn6+Bfkp/LTbGTPVfEh3/Wej4J2WuX4OO/yzOd/zzNdA/ONMDuEA6/pk118cPc/8adPwX0bzM0SullJowX2f0SimlbHMi0IvID0WkS0QOxZ27REReEZHXReQ3IpJhny8XkWER2W//+U7ce94rIgdF5LCIfHU2jt9+bp393GH7ee9cGb+IvD/u//1+EYmKyPqZHP+buIYEEXnIPn9URO6Le89c+B4kisiP7PMHROQts2D8ZSLyvP3/87CIfNw+nyMiT4vIMftrdtx77hOROhGpEZFbZvIaznf8IpJrvz4kIt865bOm/3tgjJn1f4DrgUuBQ3HnXgU228f3AF+2j8vjXxf3+lysutY8+/FDwI2zcPwe4CBwSdy43XNl/Ke8by3QMNP//9/E9+BO4GH7OAU4bv+9mhPfA+AjwI/s43xgL9akbibHXwRcah+nA7XAKuCrwN/a5/8W+Gf7eBVwAEgCKoD6mfx38CbGnwpcC/wF8K24z5mR8c+JGb0x5kWg75TTy4EX7eOngT86x8dUArXGmG778TOTeM9FcZ7jfxtw0BhzwH5vrzFmnLkz/njvA35qH8/Y+OG8r8EAqSLiAZKBESDA3PkerAKetd/XhVXqt5GZHX+7MeY1+zgIHAVKgNuxgh321zvs49uxfthGjDGNQB2waaau4XzHb4wZNMZsB8KnfNSMjH9OBPozOAS80z5+D1AW91yFiOwTkW0icp19rg5YYad2PFjfkPj3TLczjX8ZYETkKRF5TUT+xj4/V8Yf771MBPrZNn448zX8LzAItGPNvv7VGNPH7LuGM43/AHC7iHhEpAK4zH5uVoxfRMqBDcAuoMAY0w5WMMX6DQSsINoc97YW+9yMX8Mkx38mMzL+uRzo7wE+IiJ7sX6VGrHPtwOLjDEbgE8B/yMiGcaYfuAvgZ8BL2H9Oj427aOecKbxe7B+5Xu//fUPReTGOTR+AETkCmDIGHMIYBaOH858DZuAcaAYK23w1yJSOQuv4Uzj/yFWYNwDfB3YAYzNhvGLSBrwKPAJY0zgbC89zTkz09dwHuM/rZka/5zdHNwYU42V5kBElgG32ucjQMQ+3isi9Viz5D3GWmTwG/s992L9Y54RZxo/1j/QbcaYHvu532HlZp+dI+OP2crEbD72nlkzfns8Z7qGO4EnjTGjQJeIvIyV+miYTddwln8DY8AnY68TkR3AMfu5GRu/iCRgBcmfGGN+YZ/uFJEiY0y7iBQBXfb5Fk6e6ZYCbTBz13Ce4z+jmRj/nJ3Ri0i+/dUF/B3wHftxnoi47eNKoApoOOU92cCHge9P/8gtZxo/8BSwTkRS7F/tNgNHTnnPbB5/7Nx7gIfP8J4ZH/8p4zn1GpqAG8SSClwJVJ/ynhm/hrP8G0ixx42I3Iw1m5/Rv0MiIsAPgKPGmH+Le+ox4G77+G7g13Hnt4pIkp1+qgJ225817dfwJsZ/ts+a/u/BVN/tvRh/sGaG7cAo1k/6PwU+jnXnuxb4ChOLv/4IOIyVp3wN+INTPueI/WfrbBy//fq77Gs4BHx1Do7/LcDOM3zOtI//TfwdSgN+bn8PjgCfmelrOM/xlwM1WDcMn8HqcDjT478W6yb3QWC//ecdWFUoz2L9xvEskBP3ns9hVdvUAG+fyWt4k+M/jnUDPWR/z1bN1Ph1ZaxSSs1zczZ1o5RSanI00Cul1DyngV4ppeY5DfRKKTXPaaBXSql5TgO9UpMgIn8hIh84j9eXS1ynSaVm0pxdGavUdBERjzHmO+d+pVKzkwZ6tSDYjaiexGpEtQFrkdEHgJXAv2EtkuoB/o+xlrO/gNUj5hrgMRFJB0LGmH8Vq7/+d7BaGNcD9xhj+kXkMqw+M0PA9um7OqXOTlM3aiFZDjxojFmH1Xb4I8A3gXcbY2JB+v6412cZYzYbY752yuf8FwG6YEcAAAEaSURBVPBZ+3NeB/7ePv8j4GPGmKum8iKUOl86o1cLSbMx5mX7+MfA/wXWAE9brUxwY7UZiPnZqR8gIplYPwC22aceAn5+mvP/Dbz94l+CUudPA71aSE7t9xEEDp9lBj54Hp8tp/l8pWYFTd2ohWSRiMSC+vuAnUBe7JxYe8WuPtsHGGP8QH/chjZ/gtVWegDwi8i19vn3X/zhK/Xm6IxeLSRHgbtF5LtY3Qa/idUW+j/s1IsHa6OOw+f4nLuB74hIClYL7A/a5z8I/FBEhuzPVWpW0O6VakGwq24eN8asmeGhKDXtNHWjlFLznM7olVJqntMZvVJKzXMa6JVSap7TQK+UUvOcBnqllJrnNNArpdQ8p4FeKaXmuf8fKPLql5X4fegAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data_valuesonly['CO2'].plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## La contribution périodique \n", "On remarque une évolution périodique, due à des variations saisonnières, superposées à une variation continue et plus lente. On veut maintenant caractériser l'évolution périodique" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data_valuesonly['CO2']['2015-01':'2019-01'].plot()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## La contribution lente \n", "On veut extraire la contribution lente et l'extrapoler à 2025. Une première approche est une évolution linéaire à partir de 1958." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data_valuesonly['CO2'].plot()" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Generalized Linear Model Regression Results
Dep. Variable: CO2 No. Observations: 753
Model: GLM Df Residuals: 751
Model Family: Gaussian Df Model: 1
Link Function: identity Scale: 20.521
Method: IRLS Log-Likelihood: -2205.0
Date: Fri, 11 Jun 2021 Deviance: 15412.
Time: 15:22:50 Pearson chi2: 1.54e+04
No. Iterations: 3 Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err z P>|z| [0.025 0.975]
Intercept 305.3562 0.334 913.159 0.000 304.701 306.012
index 0.1326 0.001 174.904 0.000 0.131 0.134
" ], "text/plain": [ "\n", "\"\"\"\n", " Generalized Linear Model Regression Results \n", "==============================================================================\n", "Dep. Variable: CO2 No. Observations: 753\n", "Model: GLM Df Residuals: 751\n", "Model Family: Gaussian Df Model: 1\n", "Link Function: identity Scale: 20.521\n", "Method: IRLS Log-Likelihood: -2205.0\n", "Date: Fri, 11 Jun 2021 Deviance: 15412.\n", "Time: 15:22:50 Pearson chi2: 1.54e+04\n", "No. Iterations: 3 Covariance Type: nonrobust\n", "==============================================================================\n", " coef std err z P>|z| [0.025 0.975]\n", "------------------------------------------------------------------------------\n", "Intercept 305.3562 0.334 913.159 0.000 304.701 306.012\n", "index 0.1326 0.001 174.904 0.000 0.131 0.134\n", "==============================================================================\n", "\"\"\"" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import statsmodels.api as sm\n", "\n", "data_valuesonly[\"Intercept\"]=1\n", "logmodel=sm.GLM(data_valuesonly['CO2'], data_valuesonly[['Intercept','index']]).fit()\n", "logmodel.summary()" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data_pred = pd.DataFrame({'index': np.linspace(1,len(data_valuesonly['CO2']), len(data_valuesonly['CO2'])),\n", " 'Intercept': 1})\n", "data_pred['CO2'] = logmodel.predict(data_pred[['Intercept','index']])\n", "data_pred['period'] = data_valuesonly.index\n", "data_pred.plot(x=\"period\",y=\"CO2\",kind='line',color='r')\n", "plt.scatter(x=data_valuesonly.index,y = data_valuesonly[\"CO2\"])\n", "plt.grid(True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Avec cette prédiction, la teneur en CO2 dans l'atmosphère en avril 2025 serait de $412\\ ppm$. On reste donc en dessous des dernières valeurs atteintes. Pour être plus réalistes, on veut maintenant estimer la teneur en CO2 avec une approximation linéaire à partir de l'an 2000 (index 505 pour le mois de Janvier 2000) au vue de la croissance plus rapide sur les dernières années:" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Generalized Linear Model Regression Results
Dep. Variable: CO2 No. Observations: 248
Model: GLM Df Residuals: 246
Model Family: Gaussian Df Model: 1
Link Function: identity Scale: 5.4151
Method: IRLS Log-Likelihood: -560.35
Date: Fri, 11 Jun 2021 Deviance: 1332.1
Time: 15:34:11 Pearson chi2: 1.33e+03
No. Iterations: 3 Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err z P>|z| [0.025 0.975]
Intercept 272.7958 1.322 206.343 0.000 270.205 275.387
index 0.1866 0.002 90.406 0.000 0.183 0.191
" ], "text/plain": [ "\n", "\"\"\"\n", " Generalized Linear Model Regression Results \n", "==============================================================================\n", "Dep. Variable: CO2 No. Observations: 248\n", "Model: GLM Df Residuals: 246\n", "Model Family: Gaussian Df Model: 1\n", "Link Function: identity Scale: 5.4151\n", "Method: IRLS Log-Likelihood: -560.35\n", "Date: Fri, 11 Jun 2021 Deviance: 1332.1\n", "Time: 15:34:11 Pearson chi2: 1.33e+03\n", "No. Iterations: 3 Covariance Type: nonrobust\n", "==============================================================================\n", " coef std err z P>|z| [0.025 0.975]\n", "------------------------------------------------------------------------------\n", "Intercept 272.7958 1.322 206.343 0.000 270.205 275.387\n", "index 0.1866 0.002 90.406 0.000 0.183 0.191\n", "==============================================================================\n", "\"\"\"" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "logmodel=sm.GLM(data_valuesonly['CO2'][505:], data_valuesonly[['Intercept','index']][505:]).fit()\n", "logmodel.summary()" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data_pred = pd.DataFrame({'index': data_valuesonly['index'][505:],\n", " 'Intercept': 1})\n", "data_pred['CO2'] = logmodel.predict(data_pred[['Intercept','index']])\n", "data_pred['period'] = data_valuesonly.index[505:]\n", "data_pred.plot(x=\"period\",y=\"CO2\",kind='line',color='r')\n", "plt.scatter(x=data_valuesonly.index,y = data_valuesonly[\"CO2\"])\n", "plt.grid(True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cette fois, la teneur en CO2 en avril 2025 est estimée à $422\\ ppm$. " ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }