{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Concentration de CO2 dans l'atmosphère depuis 1958"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import os.path\n",
    "import xlrd\n",
    "import math"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On utilise les données du *Scripps CO2 Program* disponibles lors du 24 novembre 2023. Ces données sont stockées dans le fichier `monthly_in_situ_co2_mlo.csv`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_url = \"monthly_in_situ_co2_mlo.csv\" if os.path.isfile(\"monthly_in_situ_co2_mlo.csv\") else \"https://scrippsco2.ucsd.edu/assets/data/atmospheric/stations/in_situ_co2/monthly/monthly_in_situ_co2_mlo.csv\"\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Les lignes 1 à 57 incluses ne sont que des commentaires à propos des méthodes de mesures, des auteurs, et des données elles-mêmes. Dans ces commentaires, on apprend qu'il y a 11 colonnes.\n",
    "\n",
    "<table>\n",
    "    <tr>\n",
    "        <th>Yr</th>\n",
    "        <td>L'année au format YYYY.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>Mn</th>\n",
    "        <td>Le mois au format MM.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>Date</th>\n",
    "        <td>La date au format Excel.</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>Date.1</th>\n",
    "        <td>La date en fraction d'année (par exemple, 1958.5000 correspond au millieu de l'année 1958)</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>CO2</th>\n",
    "        <td>CO2 en parties par million (ppm)</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>seasonally</th>\n",
    "        <td><b>CO2</b> après un ajustement saisonnier (ce qui permet d'enlever certaines oscillations)</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>fit</th>\n",
    "        <td><b>CO2</b> après un lissage (ce qui permet de boucher les trous dans les données)</td>\n",
    "    <tr>\n",
    "        <th>seasonally.1</th>\n",
    "        <td><b>seasonally</b> après un lissage</td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>CO2.1</th>\n",
    "        <td><b>CO2</b> où les trous ont été bouchés par les valeurs de <b>fit</b></td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>seasonally.2</th>\n",
    "        <td><b>seasonally</b> où les trous ont été bouchés par les valeurs de <b>seasonally.1</b></td>\n",
    "    </tr>\n",
    "    <tr>\n",
    "        <th>Sta</th>\n",
    "        <td>Identifiant de la station faisant les mesures</td>\n",
    "    </tr>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Puisque les lignes 1 à 57 ne sont que des commentaires, on les ignore. On met aussi l'option `skipinitialspace` à `True` pour enlever tous les caractères blancs en trop."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Yr</th>\n",
       "      <th>Mn</th>\n",
       "      <th>Date</th>\n",
       "      <th>Date.1</th>\n",
       "      <th>CO2</th>\n",
       "      <th>seasonally</th>\n",
       "      <th>fit</th>\n",
       "      <th>seasonally.1</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "      <th>Sta</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>adjusted</td>\n",
       "      <td>NaN</td>\n",
       "      <td>adjusted fit</td>\n",
       "      <td>filled</td>\n",
       "      <td>adjusted filled</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Excel</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>[ppm]</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>21200</td>\n",
       "      <td>1958.0411</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21231</td>\n",
       "      <td>1958.1260</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>21259</td>\n",
       "      <td>1958.2027</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "      <td>316.19</td>\n",
       "      <td>314.91</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>21290</td>\n",
       "      <td>1958.2877</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "      <td>317.30</td>\n",
       "      <td>314.99</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>21320</td>\n",
       "      <td>1958.3699</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "      <td>317.89</td>\n",
       "      <td>315.06</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>21351</td>\n",
       "      <td>1958.4548</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>21381</td>\n",
       "      <td>1958.5370</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "      <td>315.85</td>\n",
       "      <td>315.22</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>21412</td>\n",
       "      <td>1958.6219</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "      <td>313.97</td>\n",
       "      <td>315.29</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Yr   Mn   Date     Date.1     CO2 seasonally     fit  seasonally.1  \\\n",
       "0     NaN  NaN    NaN        NaN     NaN   adjusted     NaN  adjusted fit   \n",
       "1     NaN  NaN  Excel        NaN   [ppm]     [ppm]    [ppm]         [ppm]   \n",
       "2  1958.0  1.0  21200  1958.0411  -99.99     -99.99  -99.99        -99.99   \n",
       "3  1958.0  2.0  21231  1958.1260  -99.99     -99.99  -99.99        -99.99   \n",
       "4  1958.0  3.0  21259  1958.2027  315.71     314.44  316.19        314.91   \n",
       "5  1958.0  4.0  21290  1958.2877  317.45     315.16  317.30        314.99   \n",
       "6  1958.0  5.0  21320  1958.3699  317.51     314.69  317.89        315.06   \n",
       "7  1958.0  6.0  21351  1958.4548  -99.99     -99.99  317.27        315.14   \n",
       "8  1958.0  7.0  21381  1958.5370  315.87     315.20  315.85        315.22   \n",
       "9  1958.0  8.0  21412  1958.6219  314.93     316.22  313.97        315.29   \n",
       "\n",
       "    CO2.1     seasonally.2  Sta  \n",
       "0  filled  adjusted filled  NaN  \n",
       "1   [ppm]            [ppm]  NaN  \n",
       "2  -99.99           -99.99  MLO  \n",
       "3  -99.99           -99.99  MLO  \n",
       "4  315.71           314.44  MLO  \n",
       "5  317.45           315.16  MLO  \n",
       "6  317.51           314.69  MLO  \n",
       "7  317.27           315.14  MLO  \n",
       "8  315.87           315.20  MLO  \n",
       "9  314.93           316.22  MLO  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data = pd.read_csv(data_url, skiprows=57, skipinitialspace=True)\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On voit que les lignes 0 et 1 de notre tableau ne sont que des informations à propos des formats et des unités de mesure. Nous allons donc les retirer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Yr</th>\n",
       "      <th>Mn</th>\n",
       "      <th>Date</th>\n",
       "      <th>Date.1</th>\n",
       "      <th>CO2</th>\n",
       "      <th>seasonally</th>\n",
       "      <th>fit</th>\n",
       "      <th>seasonally.1</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "      <th>Sta</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>21200</td>\n",
       "      <td>1958.0411</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21231</td>\n",
       "      <td>1958.1260</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>21259</td>\n",
       "      <td>1958.2027</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "      <td>316.19</td>\n",
       "      <td>314.91</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>21290</td>\n",
       "      <td>1958.2877</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "      <td>317.30</td>\n",
       "      <td>314.99</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>21320</td>\n",
       "      <td>1958.3699</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "      <td>317.89</td>\n",
       "      <td>315.06</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>21351</td>\n",
       "      <td>1958.4548</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>21381</td>\n",
       "      <td>1958.5370</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "      <td>315.85</td>\n",
       "      <td>315.22</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>21412</td>\n",
       "      <td>1958.6219</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "      <td>313.97</td>\n",
       "      <td>315.29</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>21443</td>\n",
       "      <td>1958.7068</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "      <td>312.43</td>\n",
       "      <td>315.35</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1958.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>21473</td>\n",
       "      <td>1958.7890</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "      <td>MLO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Yr    Mn   Date     Date.1     CO2 seasonally     fit seasonally.1  \\\n",
       "2   1958.0   1.0  21200  1958.0411  -99.99     -99.99  -99.99       -99.99   \n",
       "3   1958.0   2.0  21231  1958.1260  -99.99     -99.99  -99.99       -99.99   \n",
       "4   1958.0   3.0  21259  1958.2027  315.71     314.44  316.19       314.91   \n",
       "5   1958.0   4.0  21290  1958.2877  317.45     315.16  317.30       314.99   \n",
       "6   1958.0   5.0  21320  1958.3699  317.51     314.69  317.89       315.06   \n",
       "7   1958.0   6.0  21351  1958.4548  -99.99     -99.99  317.27       315.14   \n",
       "8   1958.0   7.0  21381  1958.5370  315.87     315.20  315.85       315.22   \n",
       "9   1958.0   8.0  21412  1958.6219  314.93     316.22  313.97       315.29   \n",
       "10  1958.0   9.0  21443  1958.7068  313.21     316.11  312.43       315.35   \n",
       "11  1958.0  10.0  21473  1958.7890  -99.99     -99.99  312.42       315.41   \n",
       "\n",
       "     CO2.1 seasonally.2  Sta  \n",
       "2   -99.99       -99.99  MLO  \n",
       "3   -99.99       -99.99  MLO  \n",
       "4   315.71       314.44  MLO  \n",
       "5   317.45       315.16  MLO  \n",
       "6   317.51       314.69  MLO  \n",
       "7   317.27       315.14  MLO  \n",
       "8   315.87       315.20  MLO  \n",
       "9   314.93       316.22  MLO  \n",
       "10  313.21       316.11  MLO  \n",
       "11  312.42       315.41  MLO  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data.drop(range(2), inplace=True)\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hideCode": true
   },
   "source": [
    "Pour avoir un maximum de données, nous allons travailler sur les colonnes **CO2.1** et **seasonally.2**, qui ont le moins de valeurs manquantes.  Nous allons donc enlever toutes les colonnes nous étant inutiles. Pour les dates, nous allons garder uniquement la colonne **Date**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Date</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21200</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21231</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21259</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>21290</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>21320</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>21351</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>21381</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>21412</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>21443</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>21473</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Date   CO2.1 seasonally.2\n",
       "2   21200  -99.99       -99.99\n",
       "3   21231  -99.99       -99.99\n",
       "4   21259  315.71       314.44\n",
       "5   21290  317.45       315.16\n",
       "6   21320  317.51       314.69\n",
       "7   21351  317.27       315.14\n",
       "8   21381  315.87       315.20\n",
       "9   21412  314.93       316.22\n",
       "10  21443  313.21       316.11\n",
       "11  21473  312.42       315.41"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data = raw_data.drop([\"Yr\", \"Mn\", \"Date.1\", \"CO2\", \"seasonally\", \"fit\", \"seasonally.1\", \"Sta\"], axis=\"columns\")\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Un autre problème, qui n'est pas directement visible, est que les données sont actuellement des chaînes de caractères. Nous allons donc convertir **Date** en entier, tandis que nous convertirons **CO2.1** et **seasonally.2** en nombres à virgules."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Date</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21200</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21231</td>\n",
       "      <td>-99.99</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21259</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>21290</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>21320</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>21351</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>21381</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>21412</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>21443</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>21473</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Date   CO2.1  seasonally.2\n",
       "2   21200  -99.99        -99.99\n",
       "3   21231  -99.99        -99.99\n",
       "4   21259  315.71        314.44\n",
       "5   21290  317.45        315.16\n",
       "6   21320  317.51        314.69\n",
       "7   21351  317.27        315.14\n",
       "8   21381  315.87        315.20\n",
       "9   21412  314.93        316.22\n",
       "10  21443  313.21        316.11\n",
       "11  21473  312.42        315.41"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data[\"Date\"] = raw_data[\"Date\"].astype(int)\n",
    "raw_data[\"CO2.1\"] = raw_data[\"CO2.1\"].astype(float)\n",
    "raw_data[\"seasonally.2\"] = raw_data[\"seasonally.2\"].astype(float)\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous pouvons constater que certaines lignes contiennent la valeur -99.99. Les commentaires au début du fichier nous indiquent que cela correspond à des valeurs manquantes. Nous allons donc enlever toutes les lignes contenant ces valeurs."
   ]
  },
  {
   "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>Date</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21259</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>21290</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>21320</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>21351</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>21381</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>21412</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>21443</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>21473</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>21504</td>\n",
       "      <td>313.33</td>\n",
       "      <td>315.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>21534</td>\n",
       "      <td>314.67</td>\n",
       "      <td>315.44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Date   CO2.1  seasonally.2\n",
       "4   21259  315.71        314.44\n",
       "5   21290  317.45        315.16\n",
       "6   21320  317.51        314.69\n",
       "7   21351  317.27        315.14\n",
       "8   21381  315.87        315.20\n",
       "9   21412  314.93        316.22\n",
       "10  21443  313.21        316.11\n",
       "11  21473  312.42        315.41\n",
       "12  21504  313.33        315.21\n",
       "13  21534  314.67        315.44"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data = raw_data.drop(raw_data.index[raw_data['CO2.1'] == -99.99])\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ensuite, nous voulons convertir les dates en objets Period, qui permettent d'avoir une meilleure visibilité pour les graphes."
   ]
  },
  {
   "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>Date</th>\n",
       "      <th>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1958-03</td>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1958-04</td>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1958-05</td>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1958-06</td>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1958-07</td>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1958-08</td>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1958-09</td>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1958-10</td>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1958-11</td>\n",
       "      <td>313.33</td>\n",
       "      <td>315.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1958-12</td>\n",
       "      <td>314.67</td>\n",
       "      <td>315.44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Date   CO2.1  seasonally.2\n",
       "4  1958-03  315.71        314.44\n",
       "5  1958-04  317.45        315.16\n",
       "6  1958-05  317.51        314.69\n",
       "7  1958-06  317.27        315.14\n",
       "8  1958-07  315.87        315.20\n",
       "9  1958-08  314.93        316.22\n",
       "10 1958-09  313.21        316.11\n",
       "11 1958-10  312.42        315.41\n",
       "12 1958-11  313.33        315.21\n",
       "13 1958-12  314.67        315.44"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dates = []\n",
    "for i, date in enumerate(raw_data[\"Date\"]):\n",
    "    year, month, _, _, _, _ = xlrd.xldate.xldate_as_tuple(date, 0)\n",
    "    dates.append(pd.Period(year=year, month=month, freq=\"M\"))\n",
    "\n",
    "raw_data[\"Date\"] = dates\n",
    "\n",
    "raw_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Puisque nous voulons créer un graphe avec la date en abscisse, on définit la date comme l'index du tableau."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>CO2.1</th>\n",
       "      <th>seasonally.2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1958-03</th>\n",
       "      <td>315.71</td>\n",
       "      <td>314.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-04</th>\n",
       "      <td>317.45</td>\n",
       "      <td>315.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-05</th>\n",
       "      <td>317.51</td>\n",
       "      <td>314.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-06</th>\n",
       "      <td>317.27</td>\n",
       "      <td>315.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-07</th>\n",
       "      <td>315.87</td>\n",
       "      <td>315.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-08</th>\n",
       "      <td>314.93</td>\n",
       "      <td>316.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-09</th>\n",
       "      <td>313.21</td>\n",
       "      <td>316.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-10</th>\n",
       "      <td>312.42</td>\n",
       "      <td>315.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-11</th>\n",
       "      <td>313.33</td>\n",
       "      <td>315.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1958-12</th>\n",
       "      <td>314.67</td>\n",
       "      <td>315.44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          CO2.1  seasonally.2\n",
       "Date                         \n",
       "1958-03  315.71        314.44\n",
       "1958-04  317.45        315.16\n",
       "1958-05  317.51        314.69\n",
       "1958-06  317.27        315.14\n",
       "1958-07  315.87        315.20\n",
       "1958-08  314.93        316.22\n",
       "1958-09  313.21        316.11\n",
       "1958-10  312.42        315.41\n",
       "1958-11  313.33        315.21\n",
       "1958-12  314.67        315.44"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted_data = raw_data.set_index('Date')\n",
    "\n",
    "sorted_data[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Maintenant, on peut faire un graphe représentant le taux de CO2 dans l'atmosphère."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa572362898>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sorted_data[\"CO2.1\"].plot()\n",
    "sorted_data[\"seasonally.2\"].plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous pouvons ains observer deux phénomènes. Le premier est que la quantité de CO2 dans l'atmosphère augmente avec le temps. Le deuxième est qu'il y a des oscillations récurrentes. Pour étudier ces oscillations, nous allons faire une soustraction entre la première courbe et la deuxième."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa5701f25f8>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEKCAYAAAAGvn7fAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsvX2sdtlVH7b2ee59x1UTKaoYqVUIGirSVAkiSTtFqZIqLXUlKEg0HyQkVdWQCjeqqFI1aesKKA0kRAmUKKQQMwGCIIRAwlfiD+wB2RiDDR4b24xt/IE/sPHHjG1sz9jM3Ps8Z/eP86y9f2ut317n3Hnve4c7fbY0eueee/Y+e6+992+t9Vtr71tqrXIqp3Iqp3Iqz50yPdsdOJVTOZVTOZXrLSdgP5VTOZVTeY6VE7Cfyqmcyqk8x8oJ2E/lVE7lVJ5j5QTsp3Iqp3Iqz7FyAvZTOZVTOZXnWDkB+6mcyqmcynOsnID9VE7lVE7lOVZOwH4qp3Iqp/IcK2fPxkc/53M+pz7wwAPPxqdP5VRO5VRubXnDG97wsVrr/WvvPSvA/sADD8gjjzzybHz6VE7lVE7l1pZSyvu3vHeiYk7lVE7lVJ5j5QTsp3Iqp3Iqz7FyAvZTOZVTOZXnWDkB+6mcyqmcynOsnID9VE7lVE7lOVZOwH4qp3Iqp/IcKydgP5VTOZVTeY6VE7CfyqmcyqlcYznMVR779FPm2Zs+8El59Lc+dWN9OAH7qZzKqZwKKZeHWX75PR83z171jsfkpb/24fbzU5cH+cJverl59m0vf4d88bf+nDz+xNPt2X/zXb8oX/GPXnPvO30sJ2A/lVM5lVtbPnuxl8eesNbxw2/7qLz/459pP//WJ39Hvvqh18qnPnvZnv3Y6z8gf/elb28/11rlgRe+RP7OS97Wnn3by98hf/Gh18mvfbBb2n/ln75e/qcffmP7+fEnnpYnn97Lt0JbD7/tIyIi8snPXlzDCJ9ZOQH7qZzKqdzTsj/McpireXaxn83Pn/jMhfzqb/62efbPXvd+eR1YzO9+7El54IUvkbd/+NPt2Z/7x6+VL/47P2fqfe0PPiLP/46fbz9/1yvfLa97zyfk37zlQ+3Z//7jb5HvefV72s+Xh6V/3/ua97Zn+p2Pf6Zb3r7U47CmUvqz47/w6MbLCdhP5VROZVP5+JNPyyc+Y63Q9zz+pAHtD3/qd+RfveGD5p0v+PqXyZd/5y+0n9/10SfkP/iGl8lL3tLpi6960S/Jn/nuXzL1vuGnHpWvfuh17eefeXR5/1+/uQM0gjwWBWqRxRoXseDry35eFA17oyT15mPb5pW6Xu9elxOwn8qpPIdLrVU+8/TePHvvxz5jALHWKn/3ZW+Xdz/2ZHv202/6LXnghS+RJ57q9MV//Ld/Vv6jb3m4/fy+j31GvuT/+Xn5joff0Z79t9/7y/I3/+Wb5Un3zV//yBPt/3/tGERUykJE5Dce/4yslW4dZ+/U8OyI2Wm9y70CdH9pbgphXK8BO3n2bJYTsJ/KqdyCUmuVt37IZlW8+QOflFe89SPm2Ve96JdMIO9HX/8B+SPf9HJ578c6cP4X3/4q+bJ/2C3oD3/qKfmen3+PfM0P/Ep79l2vfLeIiHzok5a/xvKRY+bHr7z3E/3Zp55q/R2PZfmXWbTznNQ7/luoXb0UT/mIDKxqVy6Jxd76ueF7hVAxrC83VU7Afiqnck1lf5jl02Dhiog8+fRent4fzM9/48febAJ5/9dPPyp/8Otfauo98MKXyLe8uAfyfuRXPiBf/p2vkVe/8/H27Cu/6xflBT/0hvbzYa7y+vf9tgnuvfwI/O95vFvjvigAzUB7zw18h9UoQCuIZqDWATqWy3kmT/33xn1CCqb3Sb83rnh5mEPbW753kdTbk77cVDkB+6mciiu/c3EI3O3L3/oRk/r28Seflgf/9sMmN/mv/+ib5Iv+71eYel/4TS+XP/+PX9t+/sHXvk9+/I0flH/8878Bz95PAen7IJCn1jpme/ii4IRFW834ZQZglVAMse34TgO1BNi7BR1bzxRCVk/7yxRD62cyGAVhpqQyGbR6pJ/7REnd63LXwF5KeV4p5VdKKW8upby1lPK3rqNjp3IqVy0/8+hHTHDv009dyoshE0JkybT4F7/ym+3njz/5tDzwwpeYgN///CO/Kl/2D39BPnvReeL/8YfeIH8RAnmveffH5GNPXpjMCgwGYvk1AP8tPDErPdNiXJFZj3NmHh/LgVAVmzI7WEbI8VlKQySWsCo4Vj8bir5+uSfATvrpS5Md+V42lG7pRxlkyu1el+uw2J8WkS+ptf5REfljIvKlpZQ/cQ3tnsr/D8pjTzxlgnYiIo/+1qcMpfGG939C/sNvfJkB7W/8qUflBT/Y/wrXx598Wv7aP3uD/DWgJv7mj71Zvu6f/6qhIb7hpx6VF/7Er7Wf33e0gH/4l/sfplHLnFnRWjr/Gktqdc7r1iPjmesWnvgIahZot1idsZ40gB7XY+Cv1irzHnq9cZ9UdrR+IgO1jhmYtqyYBO0YFSMNoMdjuWiyi7+71Rx7XYrunPPjf89+WPhU7mn54G9/1vz8vo99Rl7564+ZZ9/4U4/Kw2/7aPv5Ne/6mDzwwpfI2z7UaY4/8a0/Z3KOa63yFf/oNfJXvr8H8r77lb8hT13O8ob39zznH3rd++UV0PbTR1D7zU/0fun/P02sOC0zszr1dylPPM6YyIBgCzVC6YQNVucloQWyQKUWBacdoSFSgGYUzvHflGNPxqJK5iKhlRhXrpSIz48XyTl2Bf1OqVwtXsDq+TafjXItHHspZVdKeZOIPCYiD9daf/k62j2VZ15qrW2TaHn3Y08YQN4fZvnWl77dWMLf/5r3mpQ2EZEv/85fMCfyXv3Ox+VP/b1XGurhP//2V8nX/MDrTb0fet375WvBqtZA3uvf17Mo/J5RcHrjb36yj+X47wZWwADtFjBsFjQ+axxpZnkf+0TBaQNPTH7XeGIaAFy3vDnHvp6y1753RYDuud8RDFV2NP0woWK0Xk6pJPWIDLRPLA1Rx0eDoMd/M8+NWfo6PpXPs2G5Xwuw11oPtdY/JiKfKyJfXEr5Qv9OKeUFpZRHSimPPP7447GRU6Flf5hDwOz17/uEPALg+NTlQb76odeao8//4OF3yhd8/cvkqcuekfH873i1/Km/98r288++/aPy0KvfI9/8b97ann3zi98WDqG89UOfln/yC+81P4uIvOWDn5SrlC1BLGqpbaAhDlnwK8vsOP7LeOItljdrWoE9BzWiEDaAWp4RwqgY++9V62Uy6AHH2E/9HQPFbB1k9bL5VCOGUjH6DvmdPtPvMhorA+ZMmWbK5l6Xa82KqbV+UkReJSJfSn73UK31wVrrg/fff/91fvZZL/vDbFLaRBbwRYC82M+Bqnj4bR+Vf/nIB9rPT+8P8sALXyI/+voe3Pv7L3+H/Olve5V86JO/05591YteK3/+RT3T4m0f/rS87j2fkG/46Ufbsx983cIZI7D7cqGbiCzcPF0tIZj1HYIkGS+tRUGNWU5XpS8ySy30yXxv3ZXuyoYB9Aa+N7H0udV57Geaskcs78aVJ+C0jzzxFhmwwKGWLoPxWNhK0PezDJ8sm4ZRMRlAe/C1npvtE65p712x4Okhmc97Xa4jK+b+UsrvO/7/vyUizxeRX7/bdp+N8sRTl/KBT1ju+MVv+ZC866P91Nxvfvyz8vzv+Hlz8dBf+J7Xyh/6hp9pP9da5ate9Fr5y/+kZ1F8+yveIV/zA683hzm+9gcfkf/tX72l/fzJY27zt7/ine3Za971MRGRYEVj6WCBz+y/eb24UTZxqwnIMFDLrFUtF8R63JLZ0axOWNFb8omZN9A39NUUgpbMUsvphIxfHmvFDjIRoLdw5T0jBGQ+a5/GMmD91NKCoHuyDuYxPeSpESzZWm6WN1t3yVgC+BKDYk8ydVoWDpF59xBmU/8my3VY7P+eiLyylPIWEXm9LBz7i6+h3c3lU79zabTp4088Lb/07o+Zd77vNe+VV76jW8zvfuwJeeCFL5E3wsVDf/a7f0n+s7//SlPv6/75r8p/9Q9e3X7+/l98r7z7sScNv4x8sEhfAHiM+j3HI9PZjW/sCHO3UobVqCXcNnTiSmdHpnNQG9fzIMPrJcDOrMct/LIeCU/4Xlb0dzz4FS219k6S2aFjp6CW0gljGiJzkgJPjGNx77BySSiV1qc5sXYP43WgbTKAphkorl62flhQO/d2jmMgv+seAplPZ+njvHjPgtJRt5mKqbW+pdb6x2utX1Rr/cJa6zdfR8dG5X/9sTfJt728OwTv+ugT8kf/1ivkx4DS+Avf81r5y99r47ff8uK3ydf80x7ce9U7Fp4fAfpdj41P52nZEsTKeOIM1A4JyGTWMbWEN1mrMuyTD7xurpdYalc6yZdKmNU7hLZVLodEuW3bmOOxMBlkVucmjj0Dp4xOSJQio9u08Fxsq9xYnxh9oeXQLOhYj3llvh5dP8d/OVeuynQs84xjZzLw9XA+PdizPdvege9mVy1cZ7l1J09/4o2/Jd/1yn5q750fXcBYgVpE2r0YWbralgt+Mp54SiqyRamHQPJc2mg5bVkHmaWfZ2hI+J6WHlTKuM5YTxfz04fI7WdWrrf0mQwyq/NiH9vuNMQ6qFnqR8c+Bmh2qKe3mVidLUtlzC9Tnri9k4GT9onwvQRgfc44pxO2WN5jw4DthUsyV+13CR2l30uVG10j1nPDfe2BncYZCO8fFUKv59cPKpubOrR064Ddl2yj5DTE8m/G9z7TtDMGatu+NwaZqx61vhoVkwXyriaDRkMkd3akVu5xE+wmMpYrAvQWjv2CKVNtM7ECWUaIljyQp9+4Gp3Qx8L64mgB+J2XHa4jr0gy5cZkcMHoi9b2WCHod9mSzg4otbVF1nTWz+65WYBf6jkqJlk/RnYHKx8ugxhEv6nUx1sP7KnVueXkIKk3J4uyW6vrCoHdvbFL6mWWWgbQjOv0i5J7H2OvRTfRlemEZINtCX4xcNqSesctfWtxsU2V9XMLyDAZXDqgxaJr60DWZkr9uHd4vXEAkI3FK29OR0Vw8jRNlqXCZKBrio0lk51a+kx2GfWjxYMx9iGjHrVtphTZGFRBsvjEyWLfWNK0sw0HRTJLP0udSnnifQQn/V42rY17NJkdcTGG/rKsBkdtZFkqKVd+5aPdmdUZLSY/hqfZsfgk9S5SOIQjTSxhZun73zHlxu4W0ZJZq4fmSTGLPVEIycGfyPdivXHbGb/cFAk50ZnREFrauiNzpgDNwJSBaKuXgOmlC4Ji6Zz3UcET2ekzlKuuiUtiGLQbMVtKbf+eysO/I8KV0r0otx7YU1ogsdSyTIvseHIWNEtBZta2E6tzHwE6c8G1XLDc72ZxjUGtZYRkR7sTL4Ie7U5T9uw79ntbrEemEBREo/fhb9m7aoZGRqn0P8zA6o1Bprv14/mgaYRuHeCa9lbuRFw3fxAH28q8pMzSv0hAqgcOGXgnXPkhgqEW9XbY+mEg2n53fHRJlIaOa271e70G0KRPrZ/ku7qfGP2Ved7XWW49sKd5wRsAOqNw+L0TiSJx9Xg+8XgzpMeaiaXm63ErdwzQ7DCQH0vG9z5T2VHAcyCK4OSVG3OJmVL0WQ1X5omTOEMDBDKdmewySzbniS1AY59iumMv3vLGPnnrmK1MZhx1aisBaAU8QuGkXkvrU+yLDvlA2t6iEFi9IAME6NCnzNKXUO/QYgmx3r0uzyFgT1zpxOpMqRiW2ZHMi7cMWT4xszbiAZMs7WxstTIrd0s+cXZQhIHTwYEMli0pgjQA6DIPmJK6JH3yIGqn01qp2Kd5Hsu8t53wvQ2gE++DKdNDJrsx1dA4ZLUCYSF6yxBr66guCQ1xGcAJrU773QOxVnXo2VgYv6z1eKbOWFmEQLDJbtF6odo22R0f4Ve7YRAB2isu7K/KjlrsN5TTfquAnU12m6SE781OsWXcPLsV0FuBWLzFTgN5SeCHp+y5thnf275HshoSWoBlQ7SxzJbzxtIt5/HvsiPhFAhSb2fctldAWR6yUQje0r+icmsBQOp9HNtOwIkrtzHoe4udWo/N6uz1VB4HAjKNYiD1pgScGqhlYyEg6i3mzEjZZHknQGvqOY6dWeztGVT3WTQ4zL1TlNgX3/Xsd/eq3Cpgpzxx41YTnjgJRmUpe9zSzxaz4yyh8QwkvAKiObFJQHdTHnJSjy22jIbQtmhgLFFAPoUON1gWqAwHPghPzPje4t6hCoGAWutTwtvOCQiz8fnvZgHktB55p9NCsb/NemwABt9zVi6mYKq32cEQxues6sxYMd9TgyBVbmOL3e8h0yeiSLT01N/4TgN9YiA2OVb3M3xbh5Arov7sBOykMOvx4nj5Vnasmae0jSWcgWEWyIspe9F6ZFH/fgQ9Wp0+bRHBScfAsnCyelqebhkzV7O8LxLQzwKOaiF2i7a/kx1d19JoDGN5W2sMiwc1K3NbjwN0dO89NUFll4zlkCi3LCPk4NYdzewgVufs5MPrxXF6i50GDhMZMBD11jGban3dgqj9nv4G284u+vKeDL7iz4kUUo+dJWl9IRx7be/IsN69LrcK2BlYZEfQ02yIZBOll/0nAUBv6WdHu3mWAQNoa+GhDHzQLPVakkM21PLWfiZUDN/QY2v+4GRHLS5KCxTzXcv3OqvTbBylIdbrbQVoz+XyQHD8npau3CLwZVZnV0DRYvdWrgnWOUXCDIpDAy4wGvRfb9EKAGQC0LMDcazH6ItWj4BhA/K2NhjQ6r9jpcjumvFjQaOq9zO27bOG2O88+I/6dy/KrQJ2aj1uyOygCmEfrcb2uwQMPd+LxaeB8ZzqpR56Hz0HNwK0LoMtecisKECzAyYZLZBZsp4SqZvBSUEmKgYfoGIbILO8+ybq76tizXjpNFMnsVY3pewxmddx/Ywn9lY5UwjsBLKnCjg46TsSftdAjVidLOga3mH1vAWdvGO/p+/HepnF7i1+1rY+MjTohnoe/PF3TCkmDum1llsF7NxiH0sqs7w3pZ0lPPGlAymRyIOjrtEFw9rOUgRbf4llECLzZIOxFEEPEjzrIwFox5HiYmUcsP9u7zf5HrECPRhmfC/ll4mH4JXEVr7XZ4TQrKENNASbz8D3QulgZttb6umzUC1Y1cYSbuBNQF82gBpRCKFPBNT8dxmlwuQSgJa0nVExWZ84N8/7m/WJfo8ooHtdbhWwZxZ7xnlnFnu+oWMfOn0Rv+t5Wyzt0AJzpb3VCSSppyHYPRfs8Ilvm6adJVbOVQKAPKWNbRSrSDg1Yq0d7EOTPVr6QSGMZWeANlEk/rss2Ja51BwMxbaVbHoqc/ddXGIeDEtSb4sFjf3NwJCBWjYWb1WzPmXeg58rZulzesfLLgHoxKtDPPByMbJz9WYiu3tdbhWws+Bplk+cWuXOcrL14ob29SioOWvOWI/+Hfidz05gk885UtsWKr4W/MryiZMgaHZHiAdo60Uk/LKnE3ATJSATQC3jM+F7DkMHLnhUJP2d478JGLL6mYXn680MLJ6p1dn4BGjT0x4b+mTrjRUJoyFC26hMB/WyPrG2OMURwbd918mABl0ZQCdKsbrvUWu+zUts816XWwXsXLBjgE5ToDIXPLG8dWLU4ucAvfzLpnBPlI3/nrmvYrJtMxqCpd6VZq3GTestWBoEJRa79wy8JWXHMgYnVi9sOpMqyv/FcTGFoJ4PUyQR1EJ32zt7U8/2M28b+uJAjNMJ+mzcF8Yl+wNj2y123ic6vsRiz5Qi/d48HgunYmw9DrQS6kUFRGTg+PqMUqkb+oTf40HeE7CHwmmPBKAd0LK2WADQW9BYOg0xplSoBaRWLvldv2da3+3fmxxAMxoiO9hCrWPndjIQzqwiv0G2cp0ezGjbBJw2BbGozN1YGEAnMmCuu6cYtlAVrJ8sG4fV0xIs/UR29Yr1+npg9ZKxeE9qhRrx1FZvW6Ce/T7+f16P9cF+N6vHFHweBPVyhXobAtb3utwqYM9c/uwUm7Wcqqs3DpDlB0UimHZLWPvL6sXvMRfR/27ThibWShpocu+wtCzaTw+0h/F3sfgTi2wT6lyxtLMssJYqMCdDbIsphNCnxJJlVEVmjet7WygO3k87NlYPhxKDp+x3bP3w75vxJcqtku8N18/quuPzuBYEHSmgLXvB9JMGrL3M4xph6+5ksZOC9IWnCviJzvWUsiyDIbOAqTXX3tG0RVbP/ozfZlRMrBfbpMeaj/9m4OuzU6y1EvvpqYItwIfF873WJY7z4MfJ+2S/i9WzIF1Wr78zrhd41BVwauCQgAxTTiPFvplO8FZn4rUwQ4T1c1uf4v4YWbmZfM2zbP1kXoNXpswAcm0zGdDYzhbZkfHd63LXwF5K+QOllFeWUt5eSnlrKeWvX0fHWGE8sbcCsbB7LjxvmgZGGMgkG9q72XhoapuVEtv0C5ZvzOPPlNNlber72QKMzzyFwzIftliUuQxiQNbLznDerk/M9c/maovVSumExGvhc+Xr2XfHz2w92qfE+/Brg3st8XeR9rhqn2w7vN4YaLM4CgfauKaCAiL575nFHvYQ9cr8N1bWzw1Z7GfX0MZeRP5GrfWNpZTfKyJvKKU8XGt92zW0bT+UbAKvlfH/zeIK1s3yM7Wgkxxln4K3tGUXh7krxlmGWcCIc50RLPwfEOB0QlxwowwUaiWRZ54GWzvaXesiCx9DyJTUVssp25iBwkm9D7se8H1GJ+z9+mF9Sq3VBPzdGM4kzkfmfVjPK/sefwfHntIXTiGsGQZ9z67Xo5TjhrFsUQhZoPuq3k5qWBAldWvy2GutH661vvH4/0+IyNtF5PffbbussCBW4G3Nol7+PZgAlavnFIOpR+Zgi3vG/g5ioH6MJWzrpVQMWXjZRUSV1BttWkqNsHzpFKAZEHhZr29ourEJ1xmtzt5fL2tKcTjlxhR8SidsGAv2yytTejiH9NOPIacTtq2fkReB7zG5jOaKxQsyQys/bDWWwRbFie9toc22UCqZl0T3GV0/ciPlWjn2UsoDIvLHReSXr7NdLXuy6bYdwJDwzP+L2QnZQZGMH2R8ve/7FiuFbzCySEL0PXw2BbXQJ7ZwE+uYWjJXcHc5QMdNlMU1PPhyELbtsD5s8Qa21mNgOBpfJif+PW1bhvUo150oKRZEb4qPjs/1iVKIUQb+2RbAZP1kF29tmWMesLbvmvfc+snfiW3T+bwhZL82YC+l/B4R+XER+V9qrZ8mv39BKeWRUsojjz/++DP6Bl0kByv81QCOsxKoJTwr2MeJ9LnxWQDHeA/zeJH4BZgF1PbJwk03b1rv+C1DK+mzvlJHViMPBMPYN4DhYTAv9n37DfYsUwgMMLNTuLkL7r6fAJGp55UbXQcyrJdnqdg62D5XpjoWu27WjaMr9InOowXmLFhsvpfsWUoBhjWVyDxRSnQPDYyqtT7dKou9lHIuC6j/cK31J9g7tdaHaq0P1lofvP/++5/RdxgHuCWViYFhbkHH7/lJzi74YQFdb1nuTZaKrceO0zMA8RuDH4GP9UZpjnRRwkIcxQKoNbfB6mTz4vuE72UA4uVUazX/P6znxpLlKrPxbeFfqQwSGiJT+jlVNQaVLW239ZSAKr7vDZEMjDePheyd4V5PPBPbftYn3k+7fmx7rA/+MNzSZqx3azj2siQcf5+IvL3W+h1336VxoUDgQY0EM3NObbzpUzcyyW3lFqn7HvsdsVZjDCECdMY9ZpkH4UAV2bxsE0SvRcI7Nq5h28yCmVtuYmTju6qi3tYn+44d8zrQcgrQ/czy2InV6etbb9L2jX5ng+Wd9mmDoqbeTiX9dEDN14+Eet5Loopkg1Lc4lnM7udhP1OlMV5T97pch8X+J0XkvxORLymlvOn43399De2GQkHGLy4CmIZO8EBAUgXZEf0RXbKF4mDPsgMtmbLIXH5KjTSgxO/5tsW0M35mv7OFFsjGkNICuLGazCXU85Y2BznSpwDMrG0yvpELzupV0s8NCn5TLGBVCdt6VwLopN9LW7b+lkNMpu9+rpJ30rGsrjs7nswD8wHrrR5RSP2lewj7dDPIftfpjrXW1wj/C3PXXvZsMbtJtkHQ47sgS78oO9DnR7v9+9sOfMT+VvK9CGrr48S2QmYHtUxjZlDoE/N2NiiuNZc4cp2sn3x8pZR0Y448MBbEyqzxlH/dArTMeryq7NzaZPUyMPR9RzohvcPc74krAm1qQdPxjdveRJtRS3gdfBk1wuehiqwFQcP6ie9knum9Lrfq5ClNdzz4ye7v+xOd+J5PLaQLd8NEUi4QNqFfqFmQb+82r9mYZAH2+0bcRmObl1nxV91gG05P0n4ON6b0d5iV60Bpy+EV2g55tuWoPt+s9j2WFppSgIlhQOeq2v6xthnYL1Yn6RMDWrfusjMPWG+LQrg6DRrHEhRBloXD1o9fr6tzdTfKJm/7hnD9dgF7lu7IrdU4kdHSkvDOFkutgfEqp2/7lX4vLECyuBKw2MohdiCwVy4wpZEGKkm9rA9ecWVei+37+obe4t7TA24bZHdVy3sLXz+0FOFfNp6MdvHGyZqSCm1uVBqjdcD6zT2EbD5ZW4N+bpgrFgRdCw6PZFeTfqoRt/Wg2r0utwrYGffnrbEDyTbZslEMFcMAa8MmaBPJTqOmp+34N67Ov0p450BccK/wmJwyF9zHNagLThWsbZN7Se2RHOZlY2b88rDtRL7Ld/xY7DyJ4Frpz0ZKeM29H+Y9bwSC3IBh9Tg4bsljX1MaHmi3yIn1c2vAOtaz3zB9cGuKeS2ZshFZqMPsBkg7Zvvsd0vw9DquFLixgkHQCDJjMESr2gMkO7RwpcW10QX31/xusWjXXFT/Xv79MdAxz2bTMWq64OP3vPXW56B/r4/PWvFsY/L70cfvbEqTzJRpYuFlFu0mTjZRuOx7uTcp5tka0Pp5YVZ97ye0FQLIG/sULHwJ9TJq1BsNrG2/P1a9ycF+RPz1a8y05Z7R/UnGcq/L7QJ2ItjogpMgIbUM7WLZJ8FFVi+z9P2BKMuVj/u0ZfPiwhi60isbJfL9TAb2XdZ37iEQGYyszrVNV7lLnAbInJzUNkuBAAAgAElEQVTx/6nVuaFPW6xcoxCyes6goECLfQ9zNR6LeTaLHKZcdhzsvczHQJvFm7K7V/IYTaYksv0Rv7dY7PEdprgQb72ln3kW/hmnlWJb97rcLmAnwONzsbfmVG/5oxg0AOfeZwvHp0laq9O2Z/rpF/6KRRJAhrzDgl8hoDvH7+XpnFmfJPbBtX8Vq3NtXihFNQ/4ZadwzViS+WTrIFitGy21YLmvWI+1LvPVQd/2ezS+Q60yrcaN7Hf12VpuvZdLZlBwD9N/n41FYj2nhDNlo22txz54Pcanb1FAlIcnsrvX5ZYBO/4/B1jOlfd6I44zc3/x2dV5xUr7lFEFupnXLOjq+smtzvVnaY6zATX7uy0ekelXVo8q2KuPRdugyi3ZdFy+443pDQnu3gs8s+9lQOv7uRWcPNB5K3Rpmz1z3yNznvUhU2SZ9Z+tuwwgc0Vix7eWOcP6HoKgmQxcW1kevX//XpZbBuzIv9rJ9ZaUyHjBYz39FYJop1nG9bJUSk/FrP0JL2Z11rq+UUJf6GJbl8EWpYH19C58bq1y4LFBUNY260OV7EoDMxa3gakMEm8n91AIgKTgxMbn69nn+LuQobECTiPlhiXNY3cG0zoV4/pLwTj2I3grdP0k42vfG7ft550rQInPEks/s7z9WtxqGNzrcsuAHf+fg5MB6AZEY+3LaA82ISMPYS2Aczcbk2ZoEFDzufzMUqR/nCKMJb7D73NxfSIL1wc42WbaEkOgVmfi7eh7Ge2C/cxODvI/fGLrZWNJr8AgczUay3qGBnsWeeO1sWi9DIztWHyf8n5u8ZIyisMrQ6bgfd+zpAHTB9cW23uZodfqrcjuxLGTkmvxOLHMKgq8dLLgmcvm73HfEnBk6Y9ri+RQK/Ui2A2MHnCZcmOLObtfhVn/fnzMyh3FLNYt4WOf3PgwP4Hy6WSOIzhpvSyrSszPSztMdn792L6ZehuszrUc7rkOwLhGufgxozLlSpitfW51Zt4jW3ds74U89qRte0meV9TxndHe8+2I2D+jyb4319r+oL15h8Xq3FqkbSee1L0qtwvYCRWTW6tsQixA9gW/TsWwINYaLXCoVQrbvGShekWyforNbbD2L6m3wXJiSjHPilnvk37HyIlYwowHR3ljP7dYeDwlcgwE3N1OQM2feN5gqVGrc9XbeaagNhgL+54DIyrf5DZSHR87rHdJvOW923tsLH7+2HwyJWyAdV7PdmP1auUHD5mS8M9YP32fbqLcMmCH/w+W+vKcWatb/igB/0tBFni2ZHFgm0tbHuQkvMNdRE7hsEM2HpjXgkNDgF4Z3+juDR7MFPNsCxj78XmLnVqPxIr3c8W4ZG9V18pS/ZZ/2RXLMU2yt82DtRbk9FfMwvPrxQOvCAcsT7dxyzQaR5cH3881y5T1YWDRJn1n7bC7lLaMZWR5b1EabHyrMpiJIqleQRDwl5sptwzYx8L2vJ/INmucu+n23aUtq427QtiW++1/XuNklyDWWk6+HOvN5ufswi/2LL8aAGWn31NQSwDaKRKqbOb4LLVWtd6KtVrn0YVmORjWygEMga+tH5cuywDTA88aGI/Afq1P+r2LvbU6PWCLiFwy698BK//eOGVY37tkFjvxIrwhkln1IsyokliPjK8GZcPGwuaKewP+HZG4XqjXQvpwr8vtulKA0CwelLjr1esNU7WYxeUXJQUn7F+sd5j5RllPDePeAOeA7fjW+uQ3DwPHHji0faDH8tkG8+B0iPWsixqfjWR3SZ6FDWY8N+1THJ/vJ+dI7bNaOwer71wQd9sG7etAQUQZeA/B0wSh7QHIcOsxysW3tYWDFsnnmH0P5Wb6Tb53eZjbH4P3Sspb9VOxAK31UNnc2U1GdlN7R47fq3K+K62eGgFnUzFK487Z5OQyu3rH751NxqPVejcVPL1lwD62YJkVyINKYur5QCC+f+naWrNoR7mtTNMz3tR7CMxKutzH8QUwXAUCuzHYRm1u+t7K/IL06WINCOYtQaXYzyC7gfsbn1W52FtZsnf89w7e4mIAXRcZq1wz+XpZsXWwalUPrEBq/aeWfv9eB8wulwnAUL83lS6Dy3mW3fElzJg6a896n853xciuA1+X09lU+vcRDEFJ3dl1ENW27+ymYOnfObOgbeodsG393ixnu0lKsV6n1quoEKDe/jDLfe2d2E9UwthP2ye5kXK7gB0A02eJdJBji9kCH25Mzj1KeOZdzUMdf+8iuKhr7mDs59Ad3BCZZ0Dr28bFm1vCLqi05qJS8PXWo4R6B9h0th4D/5zHXGSXg/goaLY2V0hVGEvx+L1SrJIygAn1/HwulmGXZwNRkN35DqxHAhb+mcrJgNM8G+BbnllLtIHo2RicNB6B9dA6xjU1BlqV0/K9+3aTWVO9bRnUY99z9VDZwFjOpyJTKVYua9+bq5wHi93LLiqbSzOWm0H2WwXsf/VPfr58y1f+ERGJ1iIFMAIyQ0vYgX+sx8Gf1du7Z2vKZsTlrvWp1XNWNbv0LFiBzJrbR7lc+D6Rd2hbG6xObRstJw+GOt5SrJzUUkS5MKvzzm4yVmezMHGzgivNrMfLwyz3waa/JIBpXX4GFjWv50D0vhXA3B9mOduVoCRYvfvOJhMbQABrz4zViWOR1jYFcXh2MApB23b1BkpDn6HBdN+K7PbzQruc77jsUAkjNbLIbjoqZmnfo/00yoYoU6cQqMU+nyz2tPz79/8e+U8+/98RkWhx4QLUMnK310B1BLSUciAWbBYp5+lqWxQQ6WcDbW4do/V4ubfco35/N5UQRDPPvPUIFqb3krzbiouZu6gL0O7AclpAdNfqWRdc4B1vrW4DUQQLvVPGfI9Zj96iNUDbvy+ygCh+jwHmfWe7aHWerVudhnI40iDG6gzWY+9TN0SsDJa2ZrnTZIBW9c6OxQBmn5c+x2itRtlVtISPSgN5d1QIh7nKOVAjjbve2bGcT5OUUkCeg3Xn+nQ2lWM9WK8ql5nPy36e5fzM8um1gnKbB9QPtH3i2AdlOqKTRsrbQmjWa7QePdfJQZxb42fGJe4Wpi40zUTw2n8CHlProTt4cVizHj1gHt+hoGatMLU67ztz1iMs+IsBEMVnM1hAfcH7d0QWMBwtZtM2BC7Pd9MRnOJYau1cuRnLIQaxFsBChYAgA7I7P74zW0DRtkYb01iPrd6u9bPJ5XzXg5IBMJmyWTyNs2mguGZrHaOHEmU3AKedBSdUNpqlgmsKFcJYUfc+dRko7eHqDRSu9mFVuc1j48F7LT7AyZTw/lChnvR6hjKKysYqahgLyFPnB2mzA+7rGzLZrwXYSynfX0p5rJTy6HW0lxUFdtTGHjBFIvDgpBlrzlEcuHmsS9ytueed7QL476ax1dmtuR3fvECpsMWFoDa2OmF8e/yeysW5qHThDhbzYGMiYIqInA/ohCW75dinc6tIznbF0Czo8uPp2ztnO6CVogVkqAKgse5z7r3loPumF1nSJC9MPR2ftbxRuen4Lo1nwdcPozg0kDc5OuoOWXc4n3uUXfO4xnNFrXqm3JwiQaPjvjMOfK2fhMbyNAS+49s6H9IeA95/rrJTrwVoFksP6ffA+5hnOZuOShE8066ooZ8wn9aLENN2kzmsVzufHQ9uolyXxf4DIvKl19RWWhqPajbvrrl1DXzPd3yDzbVZ2fc5AOvPAKDP+6RdsI1ynGz8o8tLPQA1Y61Cn8B6FFlARZ8d3PjQUtN30I1toA3ZCRacfNsWnIzVObBWbT3cKMsmROvxYsRLE+vRbjCcq5GHYK1z9Qg47eG+R6y5NleoTM93xgqk83IelQQq4QunvLklDIG85slYy5ABpgGnkXdFvBZLWRHlBooE+eVQb7b1hsFTA2rVKKkm8yM1shsCdJ/Pc7S8Z/VaLDWK3kBfd0jXLBb0VIpU6V7LHYi1MPrLU4BddlCPKMXL+ZYGT2utrxaRT1xHW2tlR6gRahU5S7QDdF+UCP6WI13qmY3pwBCpmLPdsiiR+2OBH9zQHvhEln49DxblhQEnGMsAaEWWjXIBgFXJpvdehLZ1ceCWIfU+zoEW2JNAnqNU6LzsF6tzN40UAvTp3HoILBOhA+3YI0EZoHXu2zLgtOdBSe9dlSJy5lL9xpZwtxTPAjjNxsK7JECr4KSya8ptzWIfURXgJdH0wzCfRHbGw4xz5dfB8uzIeTvPDSkV72GiDDpXDv3ckfl0tKSmO/qALpvjUQwj1uO05P5QQ5rkvS63l2Of+eJCC7ZbwoMIu+OEd0qpzGwigYo5h+BXszrF0Dr3gVs3cokZpcJcVMuVc3qIglqwnAgYMgvkzPLgI6u+bd4ZAnkKavOYjuqW8GJ1GirGu8QDK1BdYgxYoXK7JBt6GABkGxpkgJb+AkTOSzoq4XNnQft4CM5nz9RZADqA0wB85wE4YXAx1HNBXpU58/iGsYDZc9ARoFXmhmOfkZID2TlqtMtOZWA5fR3ffWYsyzqYpgElN9h7mlU1HY0xPFTU52ogc5BdlIE7oAT741Za7FtKKeUFpZRHSimPPP7448+4nelosdfaA5fGTYZgG91gSMWcT8b9Pd+xDI1oPRqw2GsgxlExaAGRbAgb6MINBuCEvL9a+nsbkLsAENdnhtM3AA0UTuBIKwTyQAZuozA6SscyFXH95LSHT73TDdZcYkZfBGXardVARw2UW7D0CQ3BvKvs0Et/Nrd10F3wcY6zsdgnC06Xh9kAn+WXdT4tOEXu2sd2rKWv/WTgZA0fIbITqhQ1y2kK8SauOFtbGsycuMc3By8ClVvPDFKP4HxFue0PlsLxHphX+l3BVzPHPniKbMD5rsBYZpMUcRPlxoC91vpQrfXBWuuD999//zNuRzn2Q41Wp27M6ZjbynhpQ8WcWSpGuXLrukNgBKkYV2+aHBVzzsDJguHCkVoKSeuZTJLznVlc91HAVL7etoXH/tFy8t5OrRjIs9Yjd7d3cSx4yAYtNee6W+ux9A3mrLlDBSXsvCukLzRYPLS4iOxQBqNA3ki50XrqtRhwWk/B7Bka3TDYz3b9WEoOwQLBqQftlvn0XovKANcdU/B2/QyBNtBRCHxi1o8BTCo7XHddBqMslVq7xdxTPsGqJxa09yJa0JX2yQe6Ya5gvbJA8Bb66ybKraVijGV6bqPSWQqdrWcB+s7RrbNBugEVo0pjrp2KqTE3ulYfrO1tn59NLZtm74DWWlyj1LsI0J5q4hw75+Yv95XIYKzcZtxgu249tnpESQWX+DhXJl4A9ZA/71SX5VYjrcTjBRf70ViAMpotdy0yBrVWb7bph/aQzYhWkrZ+znfT0VOUTsk1qslSOB2gq7FWvfeB6zXKPMpuTOGAciOZQeh17oGOGlEj/nv1aIice67c0V/RqkZjbAC0sxgevMJ8nhMZnA8UiTFEwPjbO2WqsmPpq6hwb6JcC7CXUn5ERF4rIn+olPLBUsr/cB3tsoLpjgiYIh2cPO8WQG1vN0o91jtzVMx+7uCEWTgmXW3f3UFMe7O8YrQeF16xL64LZskQPjsqqQhqurjODYDYINbYcvJ0lLUefcCo1g5Oqtx8lsoM3Ko9YLJQKtNkN8qQl24bs1M/NJA3yKaxGRPEWp2VTpjcSVcfZ+DKlOaVDyiV1of9fAQZ57WcE6B1Ch4P58R1Z+vhJXk4V2EdHOeqOK93nCaJKYJ2L+j6GVm0fXzWA2OBYEr9zNagiNRIEp+Y4vqhQdeQ7tiTBJB2URngPTScypMbKddybW+t9S9dRztbCk93RO0/Nw4RrfHnkbRFH+DEjamL6z7DZ3fr8VDdRlG+lwAt4zr79wYW9FxdTnW0oHExezrq3C2uy301bYd0xxnrDcBpFqMU2/iO4LQ/uEDXwDr2AcC+waIVeEFBrbvgLDNIgaA45RZ5YjZXloPW7xnZERfcHnphLjhmOdmTtWfTJLXONiCHh4Eg/hLjEzGovawDl0gA1mpPLeRXA+iJTqQXh3e3uHTHDmpQj50dcPw5AnQMBPd65yiXRmfOAcTxnS6Dvn6ed96D/SwQPLorZgcBci/zuS5/LQ1jH30ddKbhJsqto2J4uqPd0IyKMRa0mxDNoriji5IAn6Fi3MEfBCefpxvpi764zs+mtsE6zWMBWttCbhVz5FnaYssWwPxw9D7QugkbOqYtItXVvY8ITl52I14arblzpBPIMXUbBNX5XBSC0hfM++jHzR21NnC3sZ/nMJ8qc5YdZcEXZF77H+9m/H3o5y4DGQv2zduZO42lVJDIem69N2BYhg+e6OxUzNoBpeXZ+TGzjJ0E5ZlIx3gB8P70ENNAmeL6id6kzRZCmsd4LcQLVcPgDIKg6mVPzkvSdaDXTC80IVjsx++JyO06eXqTpUC6o0/1W4JtNWzMyC93rrw92y/Cb5w3cW1x8zRKZX+kE47W44XfYLO3GqT1ySySgRu5axtF6/ELoiIt4PLDXS72fnYKQTfmWee869G7GeXkYx/69/IUOksrWZDpGU1RkWCapAbWSvB2LFgw3t96A1EJm2DmHMHJBA4dPYMpn6FPM7Moe+BwnFPtlCIoG1XCLJiJh63Od5MJzGLOOM3Jn8Ejgu8hRYXXVmC9nfN2UCmid+XXufL+uM8YdejX8NmRjkKD7c7OHn5SGWAas8ZoDgPZNa/FyKDLzlNdIiJV7Jru3k7t1I/cTLl1wK4We62dl35esNg9z2dT/TyfrdYxBqM8yNQ6yE5wHgLnvMHyhsV1TsHJnnBUTg/Hwi45MpbabDnLpZ/VWGXtDpZzC1iYruazDDKqCa3Vy73vkw2MIvfYrBuMT2BGyD6CWpsrVSRuPtFy8jLQA1F497o5hDb3DB+8rdMEDh34HkAGpfirENArjLEApJVs4DKCE6bZGjCkVmftOfIiAWTKMcDpLW8zFl8PFKWvp3vofFrqzbWG9NUqUXa1du+jHN/xeeUYpzJ8NtBfIoPcelVuZ1MDVT0tPk3LB/33mqLeRUpFvSRdF7EeOWx10DhKj13d63LrgB3THX0WhclOAHCaK1j1jIqZa3OlVfszy4ml+jUqZhpTMZwnnltWDD+mbrnOHvyCQOWM4ATf26tFy5WbAQKSZbC0DZkIJiAXqR8EGeSuGzihFbbbxVTRQWaH3jHDDo6dTd0q8mMRWZS35sijDM6mDlgBZKQH+47LLK4f47VgrKPTAgLex4i31blqaYtTPMy21LOWt8JCS02dbNvtLhNBK9vRCRoLEBvYx37qfDZvx9EXNN3R1fNUhcpp6ecRoKVTOE12PgNGrAza+OYZaA+2ZzFLxeaV76YiRbhhUI9rGC8AlPC97rWc73y9SYp02R3mCvEXuZFyC4F9fPJUqZgzA04kc8ZRMYejlXlnN8lu4pSDPzBkg0PdbR1G/Y/g1K4igMVchQQzawd/HbNIz8JRq5MvyhksmXjcHAHE8/XNjZR4aEpdTfM9cZt+tgfAtIzuLWm02aDtnprW3diFUok0hOWXEdR0ruwdIaOrATD41eREwInmYuv6cTn5ZnwGnI5gKEoBRqBFw6DWbhm29UPXgQYzmdUJniko3FbPZY2o7M4aGEaliNRIz7iJ8j04pY8KQWVHYy0Djh3z33vbsF8a1WVjbmdEuZ2DIsG9ILB+jEHhlE33lnucSrNndkewvyGK/fYCu4gNLor04Ew/lRitKxmAIVIx9hpWu1GWtvDZbCYbr5nFttVNb+7goXsIhlIJQUkfxJqbNW6AJ6FGRtkfIhZ49NrgznlzsDDjm/tYFDBZbvQFaWt4kg/euWiBNcuVn02TiPj4RL9UjWWp+GyhmO6om34CkOGAEucYgY/RApXKpQe6vVdIDIPSaYieIlg48M2dtjPrBxQeBlj1nnFUppghpjTEsoZr9BCc18I8FJpmO/ezJ152+teKpNqDRm3dKdXk0x1dlsr5VJoHphiBVxiE+Na8GDo7v/fA25kH8+kPTTWlMZUmu5sotxDYl39Z5koH6AncLA9q9nKtpR5ypC6vfIfZLSywtWS3lGKtDXtPCj/Ag8DDFqUGgnUsSz218FyKoE87w8XlXVujuDoYLheaxQMf9iI0ltlhFdAoN3oH4KT9yg6YIC1guPIBt4rgZLNULDgNgUd6jEb5UG+ViXRr1fK9C/WjynuUI2++N3ewKEel6K+IaO790WtpMp+7J2Mtb2/RdqpCZdc8IGYdzzGgi+AUcsYRRAH80Wvx4I+ya/M5lSg7YrF7Cif1EGZNUSzNIMQ1Jb5eSAiwqZtKqahBMeL0zZo+voPr7ibKLQT2DgyUl97jguepcJEHF2c5jQMxu6m4P74xGytlBHzecrrABTDHRWICwdPSRz2V2DwEgXQ1lyZ5B5RNyNQhVooCCFpAzcp2Vr1/NlKKTXEKWoFy/N7RGt8ROsEFh9H7EOlei2ag8OwWvbeEgZNTUiTOUMQCkbEwR5kdxFJjf5zCn3BE2oODTE8H7PJ08QKWoTGDN3m0FO3ZARYMB++jUVaLnBpoz1wJdwVf2pzbscT0ymYYZN4HWvFOnt3IQXrIUoe6r408gWbZewXU5NSpUqRUmkFBAtZdCZc2B0vb9u78e11uHbAXN0Ei/kg48sTsatYEfM8KbGhWrwMf1sMc3EjF2BQ6tDo1uFclXuYl1QJmrTY45PnsEcggFzj6YyNmLDsAGbXYQSF4SkVB05zII+DUKJWpK2aTQkeC2mg5CVhOyJGOlfBiPXrryhwJJwBiLtcSoR4YO0rewckFPF1GiAaC9dmlWxs0pxqCkmaOAXhiMLzneet42yVZU/Y9COiWXkdETHyirymwoGef6cXWGAN760UwhXuYlzug8BwLUioYp8K/YYteC37P3FFEKSN+bQXene8VtRo+qKgPuu5gTd1EuYXA3hcAAxkPTowXb8Eo50rjHy5gR/yNppe+6c7P4kYx1vFsLVoRa90g8FgQ9Yef7CJBKxAvOTPZLYbisB5KHF8FGTCrvgZw0vTGDpgx8OzTwPB7Jlg7iJkogMkRnNRr6ZvOBiq7l9TpGp3PHohlOfliQE09QLYO7DPr7Rgvyf35PAUUbUvBSZUbternbtG29YpptjNLX/XWeeep0ToeghPxek1GCAPoI72ohkhMURzk8rs4VePTzxDE+5yrDIzBRLxeDAS3etLz7UfxENxDxe89Q7OMEyVUdmqM7aDeTZRbB+wicsyTBVAjVoq/uc1exqQUR19wBmRqtB6Rh8eN2SmV8UEjpHkwUk6tG7eh75ix9E3oOX1vjZ8D0KoCPDNWNeExDz7bJFJdKF+Vp/K93r33LrhXija/l9fzSgrfGSt4MWCI4IRzzLNbXNZI4NMjqIl4K5dTaxeHHvhGeXbXfXTopbbgm6kHQNvXgT3xjCCzb8Buv+/72b1QG3w38Ynh96LXgodz9ETnbvLeFTtsZf9U3Q6UIu5H5crj9c2YaogKoStTO5ZivocpihhXaVlVhG6z+f6wZyd7nuBel1sJ7AG0gV82QDvzy7W85a0AcueMBKMQZPYW1DzwLCATLeimSAiAtWyaNhbm3lvuWpUSA7WuNAos3L4Am1XdrKIOTj5jwm8w7QPyvToPqEgifWFPBLd6s1V4oz98gQd4Gt9r6ARiqR2su42gFoK84PJf7O366V6SBScfCFZQ814Ez/OW9j2ruGLud7MCjez6YZmQ3bJzANbAuMvufOqHc7J++riRDdZGmePhQOsN4PxVGy9QjwRpj4NTpgKKTIDKA0qlSszlb2MBpbifl3MJmI3F41vVePD6zg6U6aFhBCj9luEjpu0l06rn0t/rcmuBfbFkjvc5TOsWHgsO+YAKuoORv4dJc5aT+cMFwWrAtoWAk7WcfAodbiYM8vhAkz1VOhvrqge/0Ko+XjvgwMkALXU1exaQlSeAE81qmFu+tsjSb7SmLDh5LyICWKMYJKYIIqgtAO1ALVEII77XX5HL1g/GGTrwoRVoKZUOICTNlsR2tJ6moZ63PgzOT0Agz8rOAq2v1zLL3Nq0OdxMkVQDfLpe+1UEQNcAQLdsGpgD26eFYzdnMxTsQXHxu3isl91khzSLC54aSkXnQMeym4LBZA4oYeDZeXx41cK9LrcS2HXhXByBSOONHmh1g4u4Y+oAFvqsgdqUKwS09A8H1Ox5wOj8bBIJm4Lw2QPPoh49BpEOaspv23rLgruzi306NxZeVzb2e517HF69MPVAXq0IasWAGuYhI2cpIqZPLfBMuPmeH76UHmTuG+WCgBPyxEi7nJ91aoJSOJilItbbQUvUKPgZXXd+NUC3VpFOOCrYsw58/KBRV1IogyyTpLU99VO0l6weDdpXuXNcB0hDGG9n7rdnquwwHx1l7nl3VIoa9FxLpbyc+82K9nvIZzugFTsvRnYQzPRxBpEYsEZKxa9zf8/OjsQLdk1JnCz2YcEFcA4gY+iLyYIFnqxDF3V5phuMcfOwwfbdKhNxNI9ueqVidn6D9XpPHw4i4l1bFsizKZg2daqYxew5dmaVeYvSZPjMGgCcWnoly3/3yk03K1NS3Es6brB9H4sU7hI3RQLfQ1DrFiVLk3RpocD3dpDxlEoFvtdRP5j3PM9m01dDY/Xgph1Lp1TwIjsLatEwaGC4A+W2x3UgdL36zBkRkafB0leQ4Tcp+vvKEfhAuYHSWKgJoEagTzaZwSqphe6ye5Z5r9q2fnB/mKVW9CLimYOWfjh1mOvxJkxtZorEYkSmFM9xrg4z3ddt3cnNlFsK7NIOfOCRe6UhDJdMsk2Qclie9VzzkIsNf2hjPzvLCTZYW7gkK2bvwPDpSx/IYx6CtYSD5S3OWsY0sGZR8sAhc1F7dgu4kcHbsafvRCyl4l13ny2gaZoiHWRMsI0eU+/eB8rc0BcEnDRm4ik5n5Z5RgwDXAe4MS2NFQ2DThlxTwrvHVfZibgUTGIFIpUnYq1OHw8xh4HmnlVl6imdINZzwvXaL1nzHqaIBouNcjPfs16EP5HLDlvZa0Gu6F8AACAASURBVBU4QONBI1S4Xnb2Rs3ZGhTI33tvR2UwR24e6VNVisyo2oPMF4Oirzulfm6i3FJg7y44D8gRKsZlJ+AGuzwcAyrESrGBQ7tRLDgpXRKzBfr3lnoXHmhn7Ke9hfKsRf0xkAdgv48gc3GwfO/FwfaTKRu1Mu/sYCzU1bTfuzCc5QaLfWJKcfQ9tdhjvXYAa46Uis67CX7NXgY1KHhB6xjA3/YzUiqWd3d/XHrnrDlKjYCyCcpN5yrKwKTCssCzXik9lPnS76ngrak2loSeokkt9JRK84SJtQrZQnhoSovJUpHIXSvHbrzlfd8LauQEOkowEGxlrhk2LMtJFR7d6y1dVprS6EoKvXO37kAGN1FuJbDjwrkDINOBgPPL+o5ew1raIlGXCgN5kZvXjRIs9jP8nvUQ+v013dJXi/2MboI+lv2h9vGBF6HfszSItH4i1WQpDvs9lldu084iQGtuvZdd2+TE+2gBZJTBXumoOFf+gJJSHFgPT1SGgz8wFq9szs96xkL0dvA8g01z8y44yqBdE0zGEi+tWk7DmvUzYZ+85d2pEe4pWqVo0nrn2dAQlhaQ9jdBz9wewgyUiha7iyUZSqXV65SDV4pGSYHHN9euNKpZr55jJ1w5WschCAr5/l65kcu8vOV9Tva6oX6cBybSvVdvjCHY30S5lcCuaUMenMJfD6o9sOYPLRlq5AgWLODo7+Fm9frfLo25tMq7swCOpX6qU1LqgsfMGfxblxfNSkJw6mDoeb4GBLP93qHa6wrmAGoxpc3IYBfrodcypFSANuP3Wyul4q1V64L7NFQ8+GN4YpgrPMDT+mmynCyF0y1YG1gzbYvb0JhhA1TFIjtrCeP1tOGCMeDYDUCLneOQIriLsjs33zt6Hy3OrUpRARPy0Z3MfZbK5Ww9RfSSbHxi5PGN997BKXgc7yjDx9BtXna7GGdQo0oPc+3M98BDKEjXxAyxFuQFz20Ha+Mmyu0EdrDCWIDjDgkq9awGG1wUQUolcpY2m4ZTMSYPmZ1UnC2f/fSltVaVs7R3qRxjAWe230u9Yu5/9ymYGAhGa85snr1VioHiCKBmuflWz2ywo5wINdKUsDCQsdZqzMKZQuDwDigbnlvvDpi0DI2ew62UikbkfCaSsfRbXKNTBwGc4IoIVIod7G2wPyi3OWa3ZPSXz8KZikv9deuO0WbN+zD54S4ffcZ10EEcvQgL9jyPHbOcdEL9uQSksQwlNmsqZbSgO9AyY0wB2itTsNiDx2cpFdNPVIqzxZE+V9Hr1Tm+VVRMKeVLSynvKKW8u5TywutoMytIe5xTCyjmcCPNoodQAkCfdbBgudj4Z/BEyIY+1gsb7OAtfb8xO4UU76GJdMKZUVLOHWyLMlpA/u+LnhFwYneR44nKUXZLfgeLzXUXEXnaAGZpFh49+INK+BBl127rGwI04YlrPPijQMBzuPuhMG+tXhwO7R0FaLTiFQiaUhStN8M7VpmGA0PAZxvZ6bwMvA+cKyu77kUYg6IdBopZRz6PHevpxWCmnvf4xB6aiusnfq8pxcCxWy9ioWK87NBg8lSMDfajnJhHwmNJdk3XCla8v4IYjLGbKHcN7KWUnYh8l4h8mYj8YRH5S6WUP3y37a5803KkfrJho1y0jWmvIWX8WXZ4BcFQMRTTx7p148ECsy9sP20qYzyq74EnnuTrIJ5yliSA4w+9mA0mEdTQwmNWIEtz81kx7JDNbirmrwD5LBWkVHw9BBAK0JOlDuwcxywVm/ngrMdQD8EJwN/JwGeJMGrEHHqBWIDKIHimKLvjHB+C7GwgmMtuqbcL9fidL+bUpbOgqeVNgv3+aoCLVOZdKbbvqcd3YOvuOMdnfSy9n0cZHOxYzFpZXoF+TuF7SqksdM1ClaJBYf7Qhngq5nZx7F8sIu+utb6n1nohIv9CRL7yGtodFp3ItcW15NZyV283IWDajWI5UpdyRYDAHjN22r8F1kb1itlgxuKabVsh+AUyYODkuXkEuoPbYKlLjLfVBZl3t9krRbxfZbTB/KZHBdjquQymLoNiFEJTioJztbS1dxuTzRVu+j7eYz1wr3VDKxDs2zsx1Y+Boe8TApiXnUifq5bDPeN8igFMlZ2QufJKCvO8Jy9z0087x2osnDmDQuWE47UyOJ4yBYDeU5n3toxXht87RMA8HA9N+XVwxmS+I0qj1Vv6b64iMJ7bkfefxcgJ50rX5sHvIbmZch3A/vtF5APw8wePz0wppbyglPJIKeWRxx9//K4+2EDbTVoDTHCXjvPRN+a8/JFdCpi4KGufSJFlUbZ6k4R6U/ueBVqtZ0DUWCTHxVx74LLVqwPwhQ0WAXN5ZwLAVCthcpvHAPQM7xwPisztGcjcjU9BbSp9886wmHUsw3pO2UxOdoe5yq5gFoUCdKfkDrXKVFCZHmVeAEAcyNRjn7Tftu3ScriP4lzaQqVY4qnEXekctMrcWNWunpH5se25VilFOh01R5mP5tivOy+7blB0r3euVabjz0u9CvXsXO0KgraYuWptl255HyrKvMtgmth8wh6auzyl9D6Zk6Cw7hRo5ypmHVQdH5N5sTLHPun+mEBJmPUqXea6N7zs4prqMr+Jch3AXsiz0P1a60O11gdrrQ/ef//9d/VBtIqWiXUWVyEWF0xkryemnk72sgDj9bR7t+BtPQeYx7ZNPe3nPm7MZcHb++Z9wOhp44KPLHZCnzjX1gL08j17NWtXUsuzCTadOGXjLWhVUta1bf10Mse58oqMu/xxrubZKjc6PqMQFkvN1/MAjQr+mNix9HNWZadjiZvXKgRpfyRlIjK3gGnBf1GUfY2xflaJIK6KeYJ5OMx2vVoQP9abjwCJHonKfFIaQo0OaQWt8zafc9wfc+VKcSoSZN7mry4UJ86V9QZgTZuxMBkQL9vLHOs5mXfZSQDxJjtQCNrPAnN8E+U6gP2DIvIH4OfPFZEPXUO7w6ICmmcb4PCurdlgMJFqeffFFekLBTC7wZxrGygcWCTHvirwTFAPLZKplP43Fo3VWaU6C8RzwNSCPoAlQzeYpWKY1eldYlU46u4ai+TgAFMVJ2wKY60Kkzlw0OgSH60wpoCMcqsSNn2TnVP6E8o88yKkKw2cK7VE1apf6nEAU9ktikQiqDVwkm7ReuuxUQVd6TMrV61OXK8N6CY3xwDQ/XtkbU5WLk0GQqx6lO/kLG9cr06ZorLRIGhcr+idx+81ZTNb76MKyMAZBjoPFWTXZR7roVy0n0F24Bl7pdjaukUW++tF5A+WUj6/lHJHRL5aRP71NbQ7LAZAYKNYF7UDkUh3rzH7InNtFcBEJKRcRUuttGwI7xJX6RZsq8esBu9ug5Wtm+CSBNsO6n20tqtr21s33Uqx9bgXISJNUVnqh7m2XXHqBlDZBcpotpugWZjO6tw3S81+z3OrFqCJ7DbFJ3S8xziOWJ54crJrm9e8E+mE+L1jPeLxoewWGTAFhGAI8gWFy+k2VPrgtWgl8cpGZefGNwMYqgdG17Qcv6dxsYUaNRTObBV1PRpVXp77lh8ev6eKk3ktjTI69hMt6CnIDmQe1qvuj44Hs5O5AXGn9G/6ErCzu22g1rovpXydiLxcRHYi8v211rfedc+S0gCk8uBXTzuDRQLWqnc1vRZfLOh+zLq5/E5DezpBA5We61QLttVzWQ2HuYbgqaVrlmeRUmFBHh8E7RsM+d5ez1pA1tuBDX1UirPbmNFribKbiez2Timi0jAc6dyBKNaz3kDwWpiFZ+gSx99DjAb7rW1hPebx9T55hVCa92H6RPlllF0HEDRg9h5owcMM/DJRJAjQnSLjXosIWNXOMLAeWDX1jOUNslvAUMZ9qmD5FmlgeJj7dcrYpxBrcV6LUkY01iJiZI5K0XtAbD694jyg7Jx3NUG9myh3DewiIrXWl4rIS6+jrS3FavFoQU9G+MdIOVhcerF/2GCQ36uW9/I9COAYoLVB1+baghY3FqxoPWJBV7d53aYXsSlX/YCSXYBmgzWwICmCiexav+sSQCrFgr3NMoibXut52Q1d9yCDLrvDcWMGICigFGeuAHXz4vh89gXKjinqntXQZZ5b3jYIitxqVwhivjeSneGOyVypXDqIdzlR8AW5iPkett1loMUERmEdUJkXu4d6vTyTbQI5LYoM/wwmcux2rqbWpx4zWWQnPTPIWNUk1tJk15ViDDzHtTGU3dQBnMVtbqLcypOnuAmYi2o2Zu2uZrcaZLiYl7alUQBYT6kfjReHQJ4g7w+LZOBZoBup9fymN1kUJl8asiEAMC8NHypELuB9FNIneGc/c+WG9cL91lXzoCcju5Be6V1pteact+MVgqGMJlwHMgBoZ3EVkHn1GQxIQ9iUz6l06ifIjinOI4hrXaTNWGxHOf2Dk7mp5y1oB7QI4uiZeo+kUz/RS/IeGMqlKw4WBLWAiZa3WedOeVNLuNb2bfUUI8dOPDfY603Bzt5TdLKbI4gfwDPtMo9ZMUF2dO8hhXO7smJuvCjF4HN3jVU2dTD2rm10pSNgjlxiZq0aLjC498dAHutnsUDALNqU01cLejfuk3+G1iPf0JECEPEADYE8VEACvKbWE26pWT7SgmHf9NqnqeVeHRxo0xQz8D48aNv4yzjwzIEHMlBwPveEdgHZ2e+B90GsXD1jgTL36bLeI9FMpInJrkSA7kBbG+et9rnn783aKH2dB68F6jWLljyLYGi/19MI+/rp3vkUvV4wRNRzMzIfKm/nLRsQt/MiEo2/RZEMZMeMuNbWzSD7tVAxN11aAGfWgIp163wGjOV7K3DeYuqpW75s3tlp/3gn9J5YaoFLZhuM8NKBD4W2tfi7wZkF7SmH+MxmdjRwagt+MlkNQSkGLynSX/PcU+HUUgvfMweGCpVd92y6osDgqYg42bl3isgsTi5o5TqL9tIBn5krBDX3PQTRrtyQFuBnHGLQ3svOypzRF81z854NyI4ZFOZ7RaLCnYrUY3yKg2HfL76eANCiPK3s+Dpo6wf2nt7LgieXfQomykBEGs+fKW/1WoZBUFh33mBaMq86ZWTeKSJ14JGcLPakWCs3Cj+eMoVFUnvgsKXCDSLzwwCgXrtKADNcH9ACcr2eBRDuRSD/Ovm2MoA21rlrq3jrkSkEm0/cwEmKUZSqbgytNEX6aypyzAt2QVAHFpbXZEpDvxezIfxdKjYI6urBxVJzWD9EKYIVZoyFwpRiOVIHlhZQq7oHQbWfuey6zC11aE6CTp5SWd6xAc/inmH6qre8SdDexW06GEqoZ/ugICqd03eUkQ+C+r03ovJ8DAr3wlLPG1pkrtRrGRhs01S6p+gCo5mnaDzT4/duYx77jRd0BxnwobvtAyqLWzfbzetcaQVxv0hyCscGVHQiQx6rxKyY5kWUCGAW7OfwPZ8NsXfWnD5Ty2JEX9jDK5zvNVTTFMHQbMKdVYo+5TMGsbor7QFl7O0ADVHiZqI53MXKXGkPrOd5cAQZ6rUc4qZXblfbQ+AJ8YKyIB9ywirzXg9iCOYkaAnrzr/T15RV+shnj2ItrK0sVdQrb1x3khgiU+n0qQ0gd4rK/Gk8XK8TylyszJsVr4rEe9nLf7pfw/ppMnDUjyD+2HVHD6GBzG+i3E5gn9a58hbgqD6IdQyMGCCIC9db+vPc6wVX2gHmGYBapsUxgOMDh54CEFkUUCk+gDNTHtXncJ8R3pYeCZ/c5lXrESy1tWwIWw8ooyEQ2PTVEAQdAUGxXkS06iVysk0hdNqjH2JCZdq/Z72P5T8GauwQU6sHtJn26XCwfRKprl5p9+RjDjc7CTrK1BmvKfBaJlTw1nOL/bTBdy0GMEHmXgZp+qpEGqspt5bHHuupUrSJEtL3LPM+nFJE2sWf0LXPXPIGfQfmGPaHju8myu0E9tJBm23oDhY9o2B5DumO5GKpZnXOnpu3B4ZGFpDep43fuwibl1ve4W86OgpAZPFIVEkVGN/OvaP1EHwnx9v6y5HiRWg2iNWUabXZEKzewdQr4a6RpU8DrwXGZ2MYJdRrWQ3BguaWNz5j62dv6K++yY0VqH+xqSAYdooDPSltIwRB1YtwVmCbTwAnn82DYzEB5MB5x0CwD+7NVQh/T2TnAMtY7FOsh4qyyQAySRQcvQzwRKePi106jp0FJQ+z9ZKo5e0VnqEApfVb5VTI2vD0kMI48xQ7pdppq5sotxLYY7YAB2gRjab7DWa1MQsu6uYVWd7197ssbUdawFA4UoSlj7F8+7lagL501+GKLKCNC15qT2kL1lWxQBfSwCrPf++WRc9Z7/XiDZf+aHebl7ahPY96lMHezlW/uEu6zIlSZBk+gbt2PKpIjGs0rwXlNEhX8+myMW8eg6c9fVWDoKX4wLO4dQAWbUXlBuCEQVDvfYiXnY9hxPEpQPvTvowrHwXfR8F+BEOkMxkd5Tn9kI0l3ICxyhToU6IUUZGwQLffC12ZynhtoDKddLw2CSPWKyeLPSvGyoUN7RegiHUHFQh8Xvnorhi+wcbfwwWo3/MUwPI9eDaBBVRsv0WiFXZ+NB97iqA9bNX4UPzTbSYn339Px8KAAL2PYkC71TOHtGx+r9a7JJawz/2uoR6PF8SrdZnFHsGpt3XMmACgLUQG2tbFPmYGDfPmcf1Uq0wX8PVeRPQU0WI3MkDgcfJkWTjW+xDzPQyCeos9S5dVgGRBUJpJMhODwlN5QVH7PdSTIOzhQPQsLNUlsigEmn4YvKvoJaUJCLDXffC9K1xOAZayrIGbKLcU2DGtDni3gbXh3bp+4lDfgUmb7FH9pZ6nT9xElgjQ+j2fZof1TCxgxBM7ULNBychZRhf1WM9tlOHhCrehmVI07qcPmpGNybhrnyusLjjSF1QpkguptJ+Zgr88dABb5ioe6mGZJPugFPMrDHwsQtcPKuGw6Y/gVMXnYhfZs/kEGcgRLMyhnuIDnokC8hk3hCfGIGi0vJN6h9lQa+x2x3hbZ7VjmZY5qFUE/8Sd94RrdbKbfHxLeD1H4RQ/n8ex+BgC7tn2DgsgH5aT76Uo1XSy2IfFWOzksiKTymiAPaZE6jsifYN5vhcBehj4IRY7fo9ROBNulEArcSXVOXZ7YMhbpt6CZYG8te8ZUBM55u563pYEXetABliPWJ16+lbrMaWId754F5wpXMyiMBZ0rcF6tHeKQ72J19N3fMaEPjNB0Jncs8POAFSkcMogt97KAIOgWg8t6D4Wa5yYAPIAMP06UDCM3kfuLavXEimVuIfMWAQMgx0CbbTYUXYq8zaf0CdcG+3ivlJoPS/zFgs4zlUxc04UifPA5pvB9dsJ7Db4FSPlHkCoBe0WID7TBX9GwAlPedK/uFPFLGa7mWy9/pdzYhoYc4kv9hagK8iAW0D9exac1tPOlu9VNxamFNFrgTTUFZmzA0NI/SzxCcKVEyU8PITink0AmPo9ehWsU5ToRcwqu6GC7/VUdlJ4EJSnr3qemAVBXXaLxCwcf+2AGR/Kzq8DB5haz1Ny3qC4NCCq/YyJC3N1HibWW6q5PeT/9KCrd/Qem1Js68cmIGg/VS4KyJECLLRepHDQ++hywnnR7+Gavqm/eXpLT56KW1wWoM3flXR5yP7oswhP9fNBswujNKKV2ykV67pfJOCE+fZSI7+91LMnBzVtcbjB0LqB7+38Bpvt5sUTcgggqBQNYLqAEf5JNE9j0WAmHtwq/OI1K1/tU1TCVbzbPKCjQAZ4g1+go0C57Q+2T8YFp0qYfW8QBB2k3k1GBhbA4ve6t4MAsukEsvQAcqOjKL9MYjSOm0fe354BOE7KYL32NWW9Bv09Gkc0jlL6+jGUCqxXdogRA88aBD0KzwZY2/6wz9DL9nvIJyW0rKpJpB7kRsotBfZl0nx+r/5xYMMZHjhYjA/+AHcNE7J2d0tXNn0RIc83lQK0QOT9axUbNIN7UjxPrO/woCu38HDBswAyiyHozZFLvRFAW0u0Vl+Pg8xaqh+CISqSSzhspSCzyC4CpqcT/P0jjWNvJ4m7YYDxELRWO10CdN8cwdDO8diL6PXEnFxuMhikr+qzouvO01+MwmlGTQTo3rb2qYOvP8+gdInl5uPeY8p0eC+LV4poHO1xz9o5no7nEJCSYzLohoHNYOqBYGn1suApcvMjas1SefFqkpsot5SK8YEnb+VaYeNGYUf1L521oRsFreNRcFF/304lhu+RoKtPwZwtrziV0tMBiwVfz7HH+3I4J2sUwhxPzZncXZH2DDnLS2cl2XoKmNVssFIiP4n1ejaEt7jcRqFjsX9Hs532dce/tZ5P3QzBL1SKovVsnAHPHLB6Kk9PR/lAqQj3PmwQ1OViT052sO48XeKvw9XvdaVoT2b69ePpEvQihFjs7F6fAyhTw+mXCND+jAPSX90LnTg3L4P1w9IPjUKIWThRmXYlXAqe3lbZRW/Aeh+9TyInjj0tFmgJDQETeXGwi9L/IQqR6NbpM7TwLsgCtIeW+DF8rkjmYMn4QBP+xZbmkTiQ6fXgwIfjbUViEIvma7P896PS0H6MUiKxHkt3HAGt/h4t4Qa+U6FgYTd939BGmQIny+kE+MMXBNTGSrF7V+N6Qr9HYzsuHoKnaLXYNEL/TGkIK7tS4uVTS70YRA9/25Pmo3sPsyuuDrSR4qTXPjtvx/9Rc22L7SH7NxSssbDQSkI9PptRZNcwBkFFXJpksWmL+m2996YZFPBOkwHcC+UDyDdRbiWw26h/tFb9XQ2WK8d3OviLiDl4czlwic92RCFMsDEh8FOKD56OwKIfGFrqYfbOBNYNSd2sfmNaL6LVQwva8L3dKtOxmA19/H+UgXFt/SEtsZylcW2dFYjfY5lIvW28zGscCG7fI7QAC+SFGM0h1vNnABgl5/8otc4xGgYc+EiMBscyWYpDHH2hyiXNY/d9KkR2JVq03hPu/L1VQN7j2y0uQXvWg5l6I6L3iLhninN8sY9K0cepqpddsbn8XWlYrwxTMPW5idEcEdLThALUD1WKcDcN7r2bCp7eSmBfwAItdrsxPXfMuE725+Q8xYCW4WijlHJ0zybNtLDfU+vR5s3PZvNSqxPGR2mIqd93kv0RD31mvAGjFI99opsH084st8u+1917S0ex8wV4YMi76fo9c/VsO7bNAsE2cLhuraJCmIJSDBY7bGjOp0cX/BJ524G3488cLFanp1SINxBoiH4CWeViU+9KqxeCoLOtN7rIClMwUZlOTnZGcbkYDZsXqvBAAS1yWsZybqgY60E3LwK9nZWL0LhSFKoQMAiqyq0HT+N8Fviep6NuotxSYMcsg6kFv/xdMSLROvbpciIuyAKb1fDLhGO3nLfNNgn1BhYJS/VDxeVdfluP/YEODqLWmiNKkYGv4zrZGYBR7AGVor89E+ellO7t2GtX/dkBmM9dB/FYz/O90mTHKBxzzYEBX60HPPE0sOoZReUUCaOj8Fi8B0xfj1M4EAw3AUBUbl0GJgiKKYKwzof56EGZ2rW5Xs/n5B9lZ2QgvZ6RAaO6rJHRYzvS2hvNi4hTpqAU0cseB0G5Mj3Mtk9e5iq7myh3BeyllK8qpby1lDKXUh68rk6tlcUSjqlM/o9Zi8T7Vejx9kGetQmeMksYgK9ZgWB5F5EBGFoQ9wdMcDGfgVW9d9+zC0nSsbCMCeN9GH6wy3N8YOhYj+bNWwuPHdy4JNkCVin6FEwYi74Dd5GjXPyFXyo7z3W2rBhQgL7eARS89QZEvHs/zF8uJfxZw9FcGaUY5Gvnsx2OcdZqwXXuDAMEInYT5+gchD1BqgpIWhlloFiLnczL6EAUzDFSMTif+CzGt/pclSJEmR4TJWavFMcZRcby1r+VCsqUnmeY+zsit+cPbTwqIn9WRF59DX3ZXMyJPPLHFKYpWoZLvb4g2MGUQMUYK5AHHDFfO95QOEhXOyD4x+uFCyguH2FnQSV2RDu4g2g5kSDznihFpKPQ+9CrF0Ssy98DYuPDMgbAWo4zv3SMBb+8fGego/R7qBQbyBx4cNGmqyH49u9Z4ONehMpAv2e8MuG0wEjB4uVh7I4iZgXGTKRtQdDZyROVqSBP7EAtXqvA59h6A6gU+1hULpz2KGb94Hy2etLbMgqeALRPpcSsKq3HjT8IapeeWcZiV96T8cr0Jspd5bHXWt8u0i2Xmyreesw2JtIlcdL6O6VYN9mnxzXAdAd/JrNR3LHmSeSpy2h1XhxmubNTbyDyg9PkudzeT3qKFqxq/8dzF7k43j85vGKi/I7GYhkamJap5XI/y27F29m7sUgVZxW5fqrSILnR3nW/IErxcp7lvrNdlF2JMvDeDtv03isM/XTKjcUn1mgz7KcH/wnAX9cP+57ma4s44JuA/gJ50j+G7uIMjIcfHae/72xq/eT0Ill3YSyc5tFxGOoHlSkzxpynuFAx/LSvP6Bk1qtEpcgD6zHYfxPlxjj2UsoLSimPlFIeefzxx++qLQQLc2hhED3HxTw6oOQnbVQvnGZ01uNidSKoxc17MBuzRJdYeCDYb2hz74V44Jno4jIZE84i8bLDMwDLxoS4hvQ+qVK03GofH/0DFm7zeqU4OtjEMoMwo8gc3596VoNXCKMAWX82qsfAOLZ1uZ+HSrErau5xTWQ+A7VWxjLwnD6Cb6dP2N1GuDasUmQZW9hvfr+KPRjHDiNRr+zgYjsmXsC+B/UGXhmu11EQtH0PZIDrYHLrNVOK0r5ns2J+13DspZSfLaU8Sv77yqt8qNb6UK31wVrrg/fff/8z77EcrTJCu1zO3LqhC7fYxYUT258J1GN86GwoAH1GN6azUvzVAHOVpJ99MdNAsAFfYq26vHnu7XDFxSicsQXU5cks75hpYcEpuLYAmCI4L10BNnAyFiVaXACYuHkN5RAPRKFcDHc9E4UwoIyYd+V5/6gUZ9on/xexsE8YL2gyQIUw8HZErEIowq+WOBxwXnhsh4GhP3maXY9g6D1Pnx6UK4+py90x2wAAIABJREFUhT4mxKhDHxPq/P3gGgd3OlXExoQafWrWouX0+V64uTz2VSqm1vr8m+jIVUoBy4ml+o3SHU3E271zDtSIiFq5U6vHbtm7PFT5t2Ehiagl2vuJ4KvKWo/F6/ciQPssnD72CTYKbl6tRy2ZJMugbULi7dhsGtz0NgUTF26TOQOZYq3j+867fD3f6xXJfNwSMaOoH+3u8mT8+SyWu+7rx8vFgxMqby5fThUMvQgCMgZ4VrwrT7v4i6y0331Oanv2vHNy19BQefc1PaLI2EliphSNF1FK/+MULnWzjW+oqLucVC5i6kmT54gCDIFg4y0Xc1+ORjsPLNXYKQQTa4F1/ryzvg5+11jsvxuLLkoRazUYS6KlQK6fYvN53iKaTSNQj2d2MIvd8MsERG0+erQeA2gDsKOFgJvC1zszd1f7dECi3BjYD4AWvR2f8qnf02yaAhZQcO/bZuLy5PnoNhMBgbbJHLN3BGSOAAb3CgWZl5FyK8776H3SttpYNnhJlr/fUk/fsQpBPT6kDu3aUIVgYy14lbC25e+h6TLoMvcUoFH6AL6RXuTz2esxhcDXAduzUZHEsSAF2FM+7R4ytzuqt3OkHJfxju42GqQMo8V+Gzj2UsqfKaV8UET+UxF5SSnl5dfTrbygBYtcuf68vNOf+QUgcpxsGP3ObbClrQ7QOiHo2lazKZZ/l0XS+0nBaUZvwL6j32MRfREAsMlbZZ6i8kqq95P+AQt3YEifIWiz2zPX4hM2/53z/miJGkXCsgzC5Vq9T/o97l3NdPMy0Dacvsm08EAkrW3/vRB4ZoDpgKE9G3oRfV580B7TMqdS3J9IBJnDvMxeKU7cYjf1BNYKU/rGiq983aHCJUHQS9dPXeelYDZNpAB9PfYH4Q9uPvWAG4I2zcJxFGB1niJ6ZZE+lfa925IV85Mi8pPX1JfNxQC0maQItCISFoCItdjxnTJoG98zPztr1X7PtjVNFerFsbB+4qIU4eMbKTNsB8EflZRVSMfc6GPf5mrHwhaup0FExhkh4c+mBWuVpxae7fqteIZHnex1ylqPBjhDrMWD0yBfenDQCHniLhebxmevR0BQ6zLQ7ChGbaEyRTC8nC2AaV45cuXGS6pHwHRB0E73SavHr28exCeOv59cP9FYsFQF83ZYBooFUS1e2WAiAatnA8G9T3cgO6qKmCCoWT+lyAHXAe6F2udcn10SLztexyA3Um4lFWNA24Gap0ZErDWnBekE1s7yjH9vzRtAbr71AYBAhIM4VUoOoNn4cLOO+snqBSXFFBKrBwu3/SwjpRif+XrIk2JbqoAm2PTmHdOWvts3DwK0PTAE77RNV2y9Y7tIcSzKhng7c781Eakm5Hvxb9Ea4NN3RNuaudU5eSvQzksWPG0A7dMWHU/tnxkvCdYri+3wevY0rPmboMfn9DTqMf1Y62nxNGj0dtwpYUIP+XMC/NI6ohDmAZ9u5MK9HWNUnf403rhwoFv+HVnLod5uBNACzybyzIEMs7IBZLDemiI5g0Vpv8eAlikJ92zle1NoW0IZtc3H0p9RRTIVC+IAmK0PVJ5cSW1RJNy7wnnXd8XVi+9sUfD6yNN0FRQJejtedv6ZjSWBkoIDPPrMZtOQeMFsrfrIlfNMEn/yFA+q9bZyL6mU0UE1+zdB/VhwTZbicvLdfMZ8e3ZiNXo7m4Kgrt7ey2CgEEKwf5YbKbcU2CPQNQ2fgozd9MwKZBs6WA1SVupNvB4Fghy0MTgswoEWj/1j+5mcbFvjtnc7Us+DWoljGVnxzNPgnoVXJExOtk+hnlckzNthY55KUK7+nU3rZ4OSYl4L7ZP/HmmbKjdviEz9ue9DKdK9JLh4zX7PKil9Zj0gAF94h9E8SPfp1wyIur3WATPSNZ4Hp9lRJGA9Au0sR957LVNBytEqIDRgTrc7JgU3nQcCas2NNn1icS3POEBvtQIjGMIY3GIWGYD2hn5OxfZBLSCrEOy7IgInVotpewtor9FR0xTHF7wkIgNGR6EFhPVW568U23aiFEM9B+LLv/COV9SFtDNQCNgpP3dLveVfszYcQOMpZf89PxY+n7EPXsZr65wp2DBXG/o0Mmi2joXuq2AYLP+PHpGuDutdcS/JJkXEzCCRcarxBF7hDVHstxPYs02XURW4SEbB07V6YXElVmC0ViOgcIpBhvVGFib2gQdTCa3k2uJ0FNu8HmQk1ttidQ4Urm8LUzf1+6N+mvZ3FnxHnhu2qf+/HkexYx4ZDyIRoJk3yZRbpJWurkiGCmF1fCXIBMdr++BBlAG0l29sx3hJZP0sBgsZi1FuEvo5lfX4Vrb/w3jhGZML0kpLn3rbpzz2pJiN4QCSbRRvmer/UwAzLriEelusVRrAnUYADeMaAIHdvBLruQ3Wf14fC763NQg6tpwIEIA8R4DpFWdsi29CWemDtzoRAEIfBj+Pnnm5ZErKW+Nr1uoItLnSiHOVelcpOGXgK7RP7BmfT/ssS4DAPsW2t43Fe3xbFYKfd76msQ8S+h6UKcznKSsmKQwgixMw0/4o7LPd+uLCXHN8r1DLwr4T+ukm+4wski38MgMLr0hykOl9GMrOyYm1tWYtX40rXwd7M14SU+DgxAEze3YVOsrUI2uMKu/BfK6tn92A4rDPhDxbj2t4y7Q/2wa+uI7KcHzjOSb6NjGq7NiWdth82vdY0B7LULmt4Mgm+gv6eSsOKD1bhVrHx39ZamPqgjsw5BkTfnHFxb2FKljjQ9lGCbRHsriuaj2OvJ1McWpba4HS0TPcUpgHvVaPeSjXQpu57+WK2raN37tKQJdbtLnMPTc/5O9JvcwIWNrSMdj2ryrzTLnhvPt1vjXw7McypV6obSv7/qjveObAjoXtddsvJrsip+BpWqjG1J/ZRkksyhIWF7bN0x0n0g7nyu2zZ7pR1iyg8UaBfg82ivmXur/2XW2ft5M/25QRMgDfZ2blbgg4blDwWzyL1KpHOqrwPiHyjA+c5SA+lIHEd7alisaxYPFrg8tJwvc8hZKtaTHAmitXbMMDOesTBd8MoNN60K+BMTaV0x+zTksGdDz4lgEkb0cEeUW7KFetwAHQGfAfpCiujYXSJUOQgXpbwGmjJeMtpyy46EGFXeNg3iEgczbYKAx8GxwUDWJh2/13o75P7J0GXON6qQUdUvaYYiFjSdbPZi+p3IVCWFk/wetN28a2fH1Sb4NhkBsUXgaxbT4+2xaX70pboZ4+P10ClpbMdc8tWgnPruqCj6iRLbzi5noTH9twfK6trdcOhMNd1GKPKZ/enWfZPKzeNnDSdyWpZ8fNxpJlBrFzCKlyY9ajVwhUkYlpW/u1tjaHWSorwEeBzscZMgUPfV+sY6aAYj/TvTcA2qWe71Nsm7XVx5asH6+EKYhjn+I3fBYX9Ww2eFcop/mGoqe3FNjHWjvX/mNw4ouEgxprZ53/HFnLY8Daspns9/zY4gLkC1Xc92KfUit3q7czbaCjyHUMXimekaC2n3fetv5uPJatXtKQNltJ2QvZJsmdQV7mltsl9ZgHNDQo4rMuO+LtNKODrCnp9Vi/R8+ixb6yXgfeDo6398nLnMmX7cfkexv2rLZv+gTyPVExScks0VEWh6/nFxW38IQ8u3pmhwYb1ygcD2qjNM1Yz44vA6JtNFY+vsDpp23bfq6/s/xrwKlc3cq9qhXoFcKWIDr2M+fm7VjWDjaN1tTabaSj8RUm36uuAwKio4yitRTMeFpcVtvWZ9OanAaGHTVy4HvcsBtQOFhvoFxG++yUFZMUvpj1d9tAZkRDcKC13+PtwDuDBb+2SEIgL7U6c0UyeoeC6CaFEJ9p5zOqi1Ma4/GN3F27USW8413wzScsvWGQrANqxU9jMNxiGVJAaePDMXN+Wcj3ttECcSxRucH3U4XAfx4+8/tjMOehn9NG2iW0FWmmcZ/ss7UMmDCfcVrMXj9x7EnJrPHR4SBfzy8c7jKS7zmLKz/J1+uIcG7Xurb2e1vud7HPrAy8QlqeYR+s4mKpoluUBPsevdCsrIOMjzPkcxeBTmU8ssDiMzF9yHjbTC6c32bgNAK1KIP0tGYCoqm1mho+hf48+p63xv26HfZpoNxWaawysqDJOkhksIVW0v5cde+NKB3992SxJ4UucHe50ZpLPLoVcu2OEN9WfmjBTTbLCIGKZ84yZIeYRifrsN5VXFRTbyPIjICAAq3r52rbA5BgY8mAbgvtYsfsQQ3aTixY/z0EmdEhrdV+EpDxHPt2K3d0jUMcS5ZIwK14+yyPT5B5cGOxho/2Cb43MKr4+Ox6xb23hVbSZ1vPnnSLnY23j+VksSclA5nMYs/cSAYgZy54egaNRg9h3PbdUiNsAVrr2PYzPeJPFcIYoEdWINbLrEDs+/IXlFifJNbbYtFKUm+DMtc+4fi2ym7XzMtEITCg8yDKxjIAGW75j+tlc8esXL8Xtqa0xvkfv2Pamsb9zGI72+bKtpWtlfH4bD39VRarSz0UOVnsaeEWkN9gY+GzZ9mJvAa0aDknCmELyGyyZJLFzTZYAAsKaji+kexYP+GZs6a2B/JyBWH6Hn4eA4Nta71tNpZnTn/Z+hk9hPXC+Ngcm3t2+LpnVu5VueuR5Y2glo4vWwf4oUE/qWFADJjgWSZzJU5xZZiBbWm1UiRJeIgyT9uGtXJTf2jjrv403rNVzMJx167mHGKv5634zCrTetFiryvB03GfuGfBPYRVK9CD2kagDRuabN7Myr0KJzuV3m4py1WpuaUm5t81GmukpLDklqiTwYZj41wGsZ9ZcI/SX2QePEBuCoImqZTZs617aAs3771P089kLKlCLyKH0VhYSquT3RblnSlzJvNIHZKxlFtyCVgp5dtKKb9eSnlLKeUnSym/77o6ln+3//9VAofZ0XVeT0zbO7M4bTtb6CFuJRGwSOIFPCffAm0aXEza4hZtBmq2nWxeEGhTheDnc8cPOo36GUCNWvXxmQ+6MjDmZxW8DOIay+I9qcLtn1toAdoOqefWK7az6bBemt7J1pT+LOZnO774LDOq6BwX38/xO369ZiCO/+/3ED/YRMbnDhV6j0H/vS13xTwsIl9Ya/0iEXmniPyfd9+l9XJVrtxbBGh5p25rsZM2pfV6/7ZYMnk2zbjtLHjaLW/bR9s262fSpwS0t5z29e+IkKBZtuk3yJf2aQP4s2d5wHP8vfxO+v7Mt+/XGBtfKUoLkHYMXbNlL4jpt/2eHxu0/QzXeeZZ9O9JeCcNfrvxCrHq/XpdN6psPQ/02s7oWTMMEjmVUm6HxV5rfUWtdX/88XUi8rl336X1koEhW/BhMxmA5u0s9eyBIaoQUlDTn2O/twR5OYBJqOcXqs+uGbY1kl3yjn3Gx7v0k4MFtrUpVfQKJyxZn/jGZP30fRq3zcbM6SE59jOOXcr4nU3egAMUNpbWJ5pGCGMZAKYFQ/s7075vh4L4GHy37CH2XrYX/Pdo2wL1vOyo5Y817LOrrNebsNqvM3j6V0XkZdfY3rCwjXkVt45lt+SBNf0Z64mtlygSBrTUZfRg+IyBVlbfwWftoFFmQbc+IlfuN9h483LZ9TZH4+PBt3W50Nz65FRyP9ikfVyReQDRbI6T8aUWrV3T+F5+1YPtI0pwC3WYjReL72cWXNxyBoABrVVm9t/Uqm/zuLRh25ZQz8dkeMrpuF6gv5L5vAk2ZjV4Wkr5WRH5d8mvvr7W+tPHd75eRPYi8sNJOy8QkReIiHze533eM+qsFk5pLD9TS9j/bU9T37bD88otWJi2mFvnNnTDzwT48FngyhN3G8eTgUWep6/1JNabbF8sV67v8BO6rE/mWWaJOoWwSg8NwYK1vQ5qq2DobufMlRtTCPZ9CjIOoPW9Q60pqGnF3GsZP9ue8mnnYQvtwsbMYmBe6StAL//v/41tR0qOjIXuK9enZO7s97SfbF7sv3OtMpkVdv1lFdhrrc/Pfl9K+e9F5CtE5L+siY9Ra31IRB4SEXnwwQfvSmcR42EjWMQFFGmI3uY4T/dqVjXL0MgzO8T1iYHauN7oj3Av9WIfrmIBsd9dxZLB722JM7B5ybwyDzJ8PvGZBXIKtIlFmSpT1s8tADLIuNG+H0b9HFjsa+vApeSnhgEzKGI8ZNu6y+YqzCc04J9x2enYxmPJ/g5q+x5Zd9sO643HexM8+12lO5ZSvlRE/g8R+dO11s9eT5fWC05kf3b8l2wUT4Uw6zGjRrSkVie8F7nyyHhtyRbYGgCM9WT4zjY6YX3hsmdbaCX8XZYq2hRJ1u8kBXJL8M08uwLQmvEU+7stysZ8L7E6vaXX69UBgNj6HMTH687PFavHFEI0DFjbcf6isTBedzRxIfUQvHxj22wskdKFsbD96OYo8+D12U3kst8tx/7/isjvFZGHSylvKqW86Br6tFqIARwCnBOZNOYWqvy35Gtbbn75N7fK9J3Y3y0HaPqtd2MwZvW29AmfRa68vzP6e6pYrsIlHwfkvp/UI8BHA44BDBM5Teuyy4CWtc/qBdkxq3OLDHDdbehTV4r2W9iH9BqHdI2Nn40OB/nvXekMQGg7Kn0K2u5nbH9LYD0/3EXGJ7Y+N9iWn39XcOxZqbV+wXV15CqFRqevMGk4MX4BZBYJflcnhy3K8BeNtlrsG8Aic1u3gcy67LJ8abNR3EVdaRoq2QzbaCX2TuT0h9cjrB1suorME4WQHYHvVif0140vs2gzLymlBTJw6l0ZzjG+kysEbceObdiWlwvrZ7DqJfyOGSJ+Dft6c62AA6SfG7wBXs+PdzwvN3FfzK28UgAXlxZvVRtQc7cWFlIvpXAm27ZIvzA/D57aBYgl25heIaxZDWPOkm2UddkxZcO5ciu7LRsF299CezBFxP+CEqfbMs9m6bttazMvXXi9zDremvIZwWmb7KLFPlYaKTVyRY9P3Byvfc9TYixoHyzhZ2p5o7dT/DtE5u1nCe+k994EJRXHos9uwmK/lcBODPZN2j/7Hb1m1i04rKdx4k1cOelwBjwZNXKdR8K1pAqh0R76c5TdFrA4Ix/ONoF+8IxoRTqPbd4c0EK9DLTj3fJkXqA1ryC3KJI0RkPWT1bPgyPrAz88p+OM9dL0VSsm2j6fl2x/8H6bfhIZpHM8aBvHR5VUMMbIvCSGof9edp7hZLEPCrPiPIBk1hxW96CdKQQEGY1s79xisX04vpMCO3tW6M/LMwn1Rqc86YamsnNgAa+E6wqI7LZYwobGam1nYBh/F9+BZwNLbY2GCCDKwJA8a21mlloCMtlYRtQadj4HaP8zA9ox+GY0Fr+Ww/bTKimJzwYGCLPYeQZUMr7QJqnXnsG6a9+zY9v6hzYyGfjx3cTh01sK7OzZGNQ8yGB1VZ75pVy2vkjXupmLSjfmsaRXGISFC/UIPdPr+3qF1AvVUjAM+cQkMMbqBWrEeDvH+nR8XAZYtgQc89OwZEOHjZlv6P47HS+ReQLQYU0a2dl1mlv6Y1Bj/WWe0+iaA3xzU6AyAWgT1yr234yX1id0n6TegJUTtp9hRHo1QGbFO4zIDMQ6yz0vtxLY6cEEdwkPO2jUF9QYoLdE70WiQthyCAXLlgBn6tZRgLaLmS1KVnGLtUHT3VIFxOuLRBorCzRlJx4p+LoTpGtc+VUUQiZzHpQU00+22a96BqC3HdvMaIjQX9IXr0gotQZtja9jGM8njiusV3JgyF9MZsZH1kFG4bT9uEV2TE4JRvhgLVc2y88nKmZQssWZu4Pxdz4IyqxOxpXXYLHDInHBWmYl5Ry7hN/5cV7VYmcZC6Ees2ifYZ88+OLGbDJPraK42bVsObjFrbIxyPQNHcfJ5Nnqt3dIvUTBpwphAJgioBSvALRYNsV2UroGZWf/3WLRYvEnSK8qu4zGaut1ir/L+pkq+OR7Ye8lFM6JihmUTguw341BJqNU8vx3Cb+LPHH/XeaC97bjwvELfKtC8PXZAvRceV4vbnpWL6Z8jhUCpnx6KobztrbfWFIXPHy/v8QC5Fr8YTKuEJJ6mVKUWD+zDEe8vwgoRXLvjQ90Z9lYfL2KaZMDX2zLewgZf4+lg6HtN2ubKRTmuaUeXwP7cT+L7xNVNmyOXZ+Si9dOFvugsIWnZcsfsDD1FGSaBTSeSKYQNh00Ih29Sl45lpwW8N/HPtl3sH50UeF77lnm7WTWDYJMpkx9umomgy1BLGrNEZMggEyyoTPZmd8F2QE4uX5b2Yl5Rs9PJECbGhRsDO57Wygr24di+plZwlji9wBoE+Mogm8cH/VanOyyfmb0V/a9bO9pxROwD4rK7Opc+bEetNUs712ctCybZlMA8PhvlrLHAXp9cXELKAMgKxfm0WTxAh9UEgHlliqEuDGj7HCu/AaL46TAM7Q6I/inMt/guRVWjwAJezb6bhZsMzSWj08woE0yoHjcxoEh6xMZ/GiuttJY8XvjflL6I/EU28/wu26ISOhT8BCoUszq2fXD1khr8pTHzosP9oiIzMdIc2axM1BrCoFsQh8YQUsvTdlzwJptsIwaYQoh5XvLuO3gtYTafIN5xWXWe7CAmAxiX/SujKtkqZh+blEIzNJPAC9VCO4ZsyzZhvZgYQyD479b/roWAydKlwxABkuejWUVSTYvWDwYGlAjz3o9Jzti5eqTrL8U9KlBsZ7J1jJn0j0LzwZ7PaPkbuISsFsN7Ln1GAG6PcMNtindMW4iv0gKWSQZV76Jitlg5WCJ9+UwGei7OJbl35SXJqDWQIZ+z8qTWexX5VZjn8Kv8kySRHbaCxqfCH9XN1TbdLDJfNfHGShw2XdENsY13M9YrnJgyM6LhHpa0jiDGwuKwIN2BtBC6nFKzvbTYkRSL5G575MNjNrvpZllx2e34RKwZ6VQcDr+uyWnmlrsm8CQbLAMaFs72zZYq0fAUAsDntZmcFFZPQbQY0rFL0r8bI8zSKiX8dKNitlgPW49earFW49Yctlt71N26IXLXMLv4nqN9YprW+Rq5ye4pyimbWw/O6iWU4d+LGQPUYpiXC/EdvCDW8CX7CG/zk1fBldLYNlCOfIYhv3dyWIfFOreB+6xvx+DoAL17Dtb62X8spZ0kRCQCPU2utK93rhtPz4L0GK+R/lXssOaDMgfnG4pn6LVIjjxuIatl/PE4VdNBllGEa9Xhu9kIOPvMtl6CtIH31PO24CTjOu5dZoHTxF8eb2sT6xNtkZ8WiZTmKmVS/ri2+YKgcjOJ0qQvqRG1aZ6cZz9DMnyz3wDyH4rgf3/a+/Ko/QojvuvVntp75W0u6wOtFp0IUASaAU2EUJI5jSHCYc4YuxgHjZgTCA+4JEY5+EDyGU7kOfgg5A8H4lNXhw7B0Y8XxhsEBhhDoE4xCmxOtFKQqur88dMz0z31JTm0x7fQf3e27czPVPd9fX0VFVXVfdwL5hkBY5JTaXTD036bBonDKVNwHw+JV95oQqB24rVwhZJlj6nFH1fueOGEOhyKUXmpfX9xKwLp0pQbqKvPP1C+zyxz8pf4CZsPcv1vewrR1hnDCn47rc3hntWnOUtPWPvHqnv5IyQ9LVo3AmzFq7v/PeRUyR+GwBSfcDlqvv73gDyjqzpOApSyLPZHRfDksbrSKFMBXu6g6UUOsnKTQVPE/ekp6ZpoSb5yg/2BROtBoHOVzY8ndtGQCe5o9yBygmnXFkGTt95sx02rzysRxBA3Gsi+dgPdg1Ayo3FzD6kdMCoX5g+lzJJIhdOVZpOchn51mMS3MImi+hZSUqRpfP7Lj1+uL7zF6/JrsP4WioDjtuKQHiH4neA4zObjhPQ/nPIEzPRdMcM+NkqwIECeQKdEDyN6AnZ13JY3pyizmN1FuqK8S0Srk5OIEhKSsrlT9ed3R5nsfuplGydzG/xt1N26cK6uWvCs7JF+RahJH5LaqaYbM+9n53tCIqETzE9MJ00fqR9b/JlhGT3OTtb9p6/OzNx75GEIZeRlieuwcWScgWQc49zy59LJ83c1MeeAS63Nb13S3xN2qFQyn+3iAVIYYKWe6EjOuZjET6dnOqXppMs9tjVFJwn70j3Qbo9ztL3feXJ3+krymR7qZWnQh+w2/2K2yNkPw85Zc9VJK4Astfce4FkQC59LW3px0j3AUeX/n3p7aKTwtD/L/VBuu54hpBPqFn4MwROqHHjJ6q7Kt0/YqqoEBPyn78beIZTJi1Ck4LvSV58WcTy5D0r4dPQw4ayFOy8ayR74YavvSXrUbLmuGtiQI4ZsBGdYM2LLhyBzg/I8XRpSZsOyMXXxHTHVDAqpktNwRmeuAVREb30ggl1+kLGoeO0TI66cwloznr0rU5HOB34WXHGhh+fSCK1uZYonLL7XOq7PM+KU4rsWpACBC27GFGyjr3zoD3JgHH5LTThId87SyH/qVuGHWUq2N3/gGytpndeYxQCY3X69JwwzeMrl4J1vEvFjRfkbc8fuG572cJpvzfbyR9AzrYefXBXCnGNJJErACh+jjC77zih5isg1jAQLDVEfc4IaMtTVZLOIcuIawh9bpUbO+6yxzIJdIXtKiooRUe5uTxxbizu3RFz+T0hKmUUsZkrHt9JSMJbMij8d67kLXYiupWInqLgQ9Y/I6KJw8WYhLhj01qcG5Q+HSfUZJeKoI1zCBmOJWlTrv3egHfqFIVTyBPbnqfcBEtGCg4lmy2k7/Kmq1lIqyfl1Yxh3YLVybo4vLq59ji/rZT2mnJHMUJN/Bwhs3lYxFMkuKRrB9fnsitGEmrp5yKtIE3llXMGRUSXalZUCNwMKk9cI5crJs2KOEvyeRoFg33IFvtfG2PmGmPmA/gpgM8PA08HRHtDLQBg047BqKyuegwA/oFYSD5Lzjfv08mCPZuOCw5KK/mkIKjkMvLzyh1ecrgTxOwExmKTXpSoXULmtTxKkUMeRVLoytM8rhFO0IqLrQR3lJhN4yvhgg0Kl2/nWgHbN/PtpeliesEwYFYZixuaeWXuuPPWAAhZMawQrkrUqO/UAAAU40lEQVTTpXzl7Hh1f1MSnOLyf4u9VPJZMcaYbYnTRoyOMsK0CY1h+3HZpLaxAIDtu/Zm0nW21AMA3t4WK4SG2moA8cOuYYSiHSz7GOeY5HezkKav7FQT2a6YXFNUYXAhelHSlhPHk6+cWDcW86LE/EoKc2jKVOjy3MEvC7vXkGxxIaRP9l123emNyRLt+fGJRHvSClALfyGXw28Oi73QhXESL2If+ArI6YPsxAXORRXT2bqRas+31AuNU9myPKtauWt5Uo3L4mPWRPQlInodwKUYJYt9bG1gnS+Y2h6VTW4PBPv6bbsy6Q6b0JS6Z1JIt2df8GZzD8be89Y770ZlTXXVzv0tY6tTdPZhcwrBumK49nwrMAnJYo9fsPS19NLndHvc3ue+X1lKMSWmXS5oZsGtPLWQFILsKw/vyWF5O3SeW4ClY4QwclmP3G/JdkNYxEWSwsy2HgtdfSsJNYlPmOx7/GdUxYyf6EPZjFK0RW58wsvlr8rb5/YenrdkmTRr4cerUKfHU0lY7ES0goieZv7OAQBjzM3GmCkAvgvgk0I9VxLRSiJauWHDhiEzvuqWU/DdK46Lzj9+4mE4e95EXHzsoVHZ8r4pqK2Of2JrQw0A4MrFvVHZrK5mAMA77+7JbOuwjkAhvLppZ1R2SGtg/duHZF1BSXS3BgphXUIhWNiB0NFcl7rWWBfUtW1XmqdcQk0MHAbnrlvA3pMii/llXTiuUBNTPiVfsCDUOItUWtjkW4FOewfr/hJ85VJeud9n7mzHLeN44tJ6I55y0BWax55HqLHjzhO0Ep3rUsmu2xeGEl2y1bQ1nv07eeWW3XfSOAczo03x5PE/kkibmR6MMR/IWdf3APw3gFsy6rkbwN0A0NfXN+Sf1jq2JnX+9YuPdspuP38ubj9/rlO29rYPOuc3nj4bE5rqcOoRh0Rl9111vCOMO5vrMG9yK65aMj0qW9jTjhf7t7PWuIW19N/YEtfVOrbGUSLW8k9icntDis7CDvjxTbWpa22h4tq9N/21XDu4GsLZzs7B2GUVBU8F34YUPBV3W0S2lSxmFHl7sLC8CHmL/I6a2QubpDTCtHWcVIp5fOWMUsy4x2mXUSQxTwK/krUqKIS47uw62dhO5MZi2hNjOwdeMCT1HWcYRLMOSSlK444xfCykoH3km8/RByUh2CUQ0QxjzJrw9GwAq4fO0uiisa4a131ghlMWuHhiNw8R4cefXOTc84Wzj8AJMzowb0pbVPYPFx/tWNkzuwJL/yPH90Rlf3nmHHz6h6vQ1pAWzBY944MYguTTnxjOBpKw7qg3t+5MXbOD0iqNHbv3Rdda6gOFIMUJ6muCEbtrT6w0/Kl7ocIpj5ARLXZOyJhsl4pssR/YFcP5yuU9X9x2kxf99ErJb8sqxRx+4kKVqV83Rye5sfIoGd4Vk6479fu42Q4zDvxZjuTmEwPIkoCWFJ94LTgfDVfMkAQ7gNuIaBaA/QBeBfCJobNUHqirHoMzjup2ys6a52Z7NtRWp2YI5y+YjPMXTHbKHr15mTOo39c7Dp8/cw7OmR/X97FF0/Dth16Jgrt2AM2d3Brdc+TE1qjdLHQ0pV0/U8c34NG1m9E/kB2fOHRcoBDe3JJwR7XUY+vOPdHAratOvw1N9QEv2wfTQW070LmAdbRnNfPySQub8sQnCg48p+ITgnByrMdsqzMVPC1UKeZQprzgErRoVDdTJijhg3VjiUpR+GpVFJ8Q/HXiTFFQivmyqpj2rEJgLX2Xp5IX7MaY84aLkfcyOpvrnXMiwuWLpjllN59xOD59yiwnZrD61tMcl8Oywzvxrcv6sHhmR1T2n9f8EX71QhzTqKoiTBk3FqclXE/LDu/EDx9/gxX6FlNCwb4tkXU0uX0sVq8fwOYduyO+fcQKwXVt9Q8MRi+9dSElYWMWO1iFENDZWYRzLWSBe3WkPGQpj11e3p69sCmLN5dOmLVIsQsxCJp9jdv+wSJPfIJ7xn6cIYm0wksqRS8ImnRjeXR8XAOZv4V7VlHdOQLP0j5Ekt9e7oPgfDRSB4dqsStGCVVVFGUDWdTXuOdEhA/M6XLK5k9pw/yEuwgAfv3Zpc75aUd2Y8UNJ+Kwjsao7IHrF+OlDTucts49epIzS1m+8FCseK4/Sj8FAjfWE69tic6njguuJQV0d2s9+gcGozLuBesNeXk5wQNRYOXZF+WQlvoUXVcY1F7PBqwDuub69LC3L6RsraYVgzRDiNplrHJpFWRMFx5wwkmkk2YBrlsgiTyzFv5nCq4YL3+ds9i59ROphXEOn9kKwaeXXFXiwjhx5pau02/XKfMUyWisPFXBrgAATO9scs5ndDVjRpgxZPH3y+c75yfP6cILXzzdmUX8+8ff79zT2lCDz5w6C0tnd0Zlnzl1Nj78nd+hd0LcZhW5e2hYt9JRCVdT39R2PLZ2S5znz0hhKfBsX1Z7TxITQ4Wwafvu1LUqb2axZWccR7GzBs5Si+iZF1pyQ/h0XM15cvlFuoNM+ZSzsSRhmBaK4ocvvNmVuHYg1eoB4gxCv+bJVRfjEzliNKOxV4wKdsWQUOv51bmBfc1J053zRTMm4JWvuLGHJ285xfmyTEdzHR6+camTDvrNy/qw4rn+KI0UCKz29sY4EL2wJwh6n5yYuRzbMw6Prt0cuSH8mQ4gKwQLO/t4d08ceJ7UPhar3niHjSEkfwsAbNoRK43GujHYPrg3cqVx6bK1oeKyayySEFfRhqJP3sYhzafv93faE6xVf/bB8un5mR06KZjJrFi14yRP6mah8QnJmhcXuEnB93LLilEohgs2MyeJiW1u5k9bQ20q8PzITUsdAdbdOhYvf/kMx0d6z58uxJr+7Y7w/NpF8zEhEVM4a95EfP3BNbigL65/yawO/OL5OD5h10EkX16rENa/kx14njo+HWeYOr4Rb28bjJQEJxDsWokNA/FK6drqKuzeuz+633fPAbFC2JVQQBaFbK6VRJ5dRfMsbGLjEzlcRw6dzxO7RgLZdYqzpPCgQIUgueT82UfJB08VimJD3j8/QGNddSrOcM78Sc75Ia31+MNfneqU3f3hPgx4i8S+d8VxkcAFgKuXHIY3t76LC/umRGUTmuqwcXssjDtDi/3ISbFbaensTjz6ymY2I8iiuzW9BcbE1nqs3bQzsso5hdgVxh64LCcrnCYw6yBqwxkNO0OIYhBMBpOwwM13e7CLtCSndQjHpSIsUIoJBDeW6H/PdlXJ2ypYv7/UnlrsCkXRUVtdhfFeptDx0yc4520NtbjrkmOcshU3LMbGhK+eiLDihsXoaIoVwscX92JhzzhnW4zrls3AKxvjYLGdsSTdUQumjsPaTTuxd1+2dOhi9kSqr6nCrj37IyHjL/BLtvfW1myF0NWSzpxqCeti9EEktG0KbtKNJaUIRvRMSqP0OcuYzv2fRK6trwU6Nj4hzHb81E0NnioUZYi2htrUArTpnW4gmogcoQ4A15880zmvrxmDey8/FnO6W6KyL517JE6a3eFY/19dPh87dsc+/p7Q9TP7kLjNM+dOxI8ef0NM9bNupde3pBe4WeHrp+YCceD5bWGfJjv72JyIM3Q012H1+gEx/c/GQwYTCqGmugrYvS/6LdysJ0+WkuRSYX3zUkaR0K4fHNbgqULxHseJiTUJQCDozpzrLoT70NGuW6mzpR4PXL84WnsAAF/546NwxQnTnBnIvZcfi5rE/rSLZ07A4d0t+NTSeCX2Ze+fin955NXoPmmTPC6jyKKLS01tsZlIg6lrFlPG2dXUsdLobK7D1p17ItePH8AHgHYmg8kiVghMTCByR6Wlb56P4+TJijGjkMmugl2hqED4qao1Y6ow+5AWp8xXGs31Nfjf605wyr5w1hH43GmzndTSn167yNnj6MK+KVi9fgBXn3RYVHbfVcfjoTUbo3MrfO02G0C8VmE358MJYRe4JWMWXS31eOHt7c5iOR929pHc86mtocZRCO3Mth52kd7GgbSyidc6pIW3nTXsZczx9O6OmWwPG1SwKxSKTFRVERq9jeqSbiAgmEV8+dyjnLIFU9tTrqZHblrq1HXFol7UjqnC8oVx4PmO8+bi96/HC9waaqvxwaO6ne01Fkxtx6/XbGQ3erOw32fYMRi7cLpbx2Lrzj3YGe6TxAVtbWCc2/7bWurdrenZhw2QbxAUQmdzHT56fA8mtaXphxsq2BUKxaig29u4rra6Clec0OuUXbhwCi5MCHoAuOtSNzh97dIZmDe5DYsSgeyHPncSBhIWfFtDDW44eaaznuFzp83CR+95LJoFcLAKIbnWoaO5DhsGBiNrnNuLyX7Ep59RCNYVM2VcA75w9hGZbQ8nVLArFIqywpgqwkmJlcxAejUxEeFTy9xdW5fM6kxtyrfq86c4eeW9HU342wvm4cRZsZvqm5f14Ru/eAldieykcY21zpYW9rsOnYmyhtox2Ll7HxtsHWnQaKTe+Ojr6zMrV64c9XYVCoViOLB/vwGRm130u5c3Yf6hbdFCuBf7t+Nnz67H1UumZ1VTMIjocWNM34HuU4tdoVAoCgTnnz+ud7xzPr2zCdM7h0+oF4IiTBIUCoVCMZJQwa5QKBQVBhXsCoVCUWFQwa5QKBQVBhXsCoVCUWFQwa5QKBQVBhXsCoVCUWFQwa5QKBQVhqKsPCWiDQBeHabqDgXw2jDV5aMVwDsjVDdQvrwr3zzKlfdy5RsoX94Plu+pxpiOA91UFME+nCCiDXl+6EHWfbcx5sqRqDusvyx5V74z6y9L3suV77D+suR9pGVLJbhito5g3T8ZwbqB8uVd+eZRrryXK99A+fI+orKlEgT7iLlKjDEjLdjLlXflm0G58l6ufAPly/tIy5ZKEOx3F5uBIaBceVe+Rx/lynu58g2UMe9l72NXKBQKhYtKsNgVCoVCkUDJCXYi+g4R9RPR04myeUT0CBH9gYh+QkQtYXkPEb1LRE+Gf99I0CwnoqeI6BkiuqPUeA+vzQ2vPRNery8G7wX2+aWJ/n6SiPYT0fxi8H0QvNcQ0b1h+XNEdFOCppT7vJaI7gnLVxHRkiLyPYWIfh723zNEdF1YPo6IHiCiNeH/9gTNTUT0IhE9T0SnlgvvRDQ+vH87Ed3p1TXqY70gGGNK6g/AYgDHAHg6UfYYgBPD48sB3Boe9yTvS9w/HkH+aUd4fi+AZSXGezWApwDMS/A8phi8F8K3R3cUgJfLqM8vAfCD8LgBwNpwDJV0nwO4BsA94XEngMcRGGXF4LsbwDHhcTOAFwDMAXAHgBvD8hsB3B4ezwGwCkAdgGkAXiriOC+U90YAiwB8AsCdiXqKMtYL+Ss5i90Y8ysAm73iWQB+FR4/AOC8A1TTC+AFY8yG8HxFDpoho0DeTwHwlDFmVUi7yRizD0XgfQh9fjGA74fH5dDnBkAjEVUDGAtgN4BtKP0+nwPgwZCuH0EaXh+Kw/c6Y8wT4fEAgOcATAJwDgIBh/D/h8LjcxAo00FjzCsAXgRwbDnwbozZYYx5CID/heqijPVCUHKCPQNPAzg7PL4AQPIz5tOI6PdE9EsiOiEsexHA7NBVU43gQbmfPh89ZPE+E4AhovuJ6Aki+mxYXiq8S31usRyxYC8VvoFs3n8EYAeAdQgsrr8xxmxG6fCexfcqAOcQUTURTQOwILxWVL6JqAfA0QB+B6DLGLMOCAQogpkFEAjO1xNkb4Rl5cB7FkplvGSiXAT75QCuIaLHEUyhdofl6wAcaow5GsANAL5HRC3GmC0ArgLwbwB+jWDKvXfUuQ6QxXs1gmnepeH/c4loWQnxnsU3AICIjgOw0xjzNACUEN9ANu/HAtgHYCICt8CfE1FvCfGexfd3EAjElQC+CuBhAHuLyTcRNQG4D8CfGWO2SbcyZaZMeGdRQuMlE2XxMWtjzGoErgsQ0UwAHwzLBwEMhsePE9FLCCzhlSZYAPCTkOZKBC/0qCOLdwQv6i+NMRvDa/+DwOf6YCnwLvBtcRFia93SFJ3vkI8s3i8B8H/GmD0A+onoNwhcGi+XAu/CON8L4Hp7HxE9DGBNeG3U+SaiGgSC8bvGmP8Ii98mom5jzDoi6gbQH5a/AdeanQzgrTLhPROlMF4klIXFTkSd4f8qAH8B4BvheQcRjQmPewHMAPCyR9MO4GoA3xp9zrN5B3A/gLlE1BBO504E8KxHUzTeBb5t2QUAfpBBU6p9/hqApRSgEcD7AKz2aEquz8Mx0hgen4zAWi/KWCEiAvBtAM8ZY/4ucem/AHwkPP4IgB8nyi8iorrQjTQDwKNlwrtUV9HHi4hiR2/9PwRW4DoAexBo+48BuA5BBPsFALchXlh1HoBnEPggnwBwllfPs+HfRaXGe3j/n4T8Pw3gjmLxfhB8LwHw24x6SrbPATQB+GHY588C+Ew59DmCzJ3nEQT7ViDY4a9YfC9CEIR+CsCT4d8ZCDJFHkQwk3gQwLgEzc0IsmGeB3B6mfG+FkGQe3v4nOYUa6wX8qcrTxUKhaLCUBauGIVCoVDkhwp2hUKhqDCoYFcoFIoKgwp2hUKhqDCoYFcoFIoKgwp2RcWDiPZRsBPlMxTsjnhDmCsu0fQQ0SWjxaNCMZxQwa54L+BdY8x8Y8wRAE5GkLt8ywFoehCsVFUoyg6ax66oeBDRdmNMU+K8F8EWuRMATAXwrwi2aAWATxpjHiai3wI4HMArCHb8+zqCRUNLEGxBe5cx5p9G7UcoFAVABbui4uEL9rBsC4DZAAYA7DfG7CKiGQC+b4zpo+BjFp82xpwZ3n8lgE5jzBeJqA7AbwBcYIKtaBWKkkJZbAKmUIwA7K6DNQDupOArUPsQbCLH4RQEe/ucH563Itj3RAW7ouSggl3xnkPoitmHYBe/WwC8DWAegpiT/1GFiAzAtcaY+0eFSYViCNDgqeI9BSLqQLBr4p0m8EO2AlhnjNkP4MMIPtsGBC6a5gTp/QCuCrd9BRHNtDsuKhSlBrXYFe8FjCWiJxG4XfYiCJbabVv/EcB9RHQBgJ8j+MISEOwAuJeIVgH4ZwBfQ5Ap80S4/esGxJ9/UyhKCho8VSgUigqDumIUCoWiwqCCXaFQKCoMKtgVCoWiwqCCXaFQKCoMKtgVCoWiwqCCXaFQKCoMKtgVCoWiwqCCXaFQKCoM/w9xcrNxXrClbAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(sorted_data[\"CO2.1\"] - sorted_data[\"seasonally.2\"]).plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En zoomant un peu, nous réalisons que ces oscillations se répètent tous les ans, et oscille d'environ 6 ppm. Le pic le plus haut arrive vers le mois de mai, tandis que la falaise la plus basse arrive vers le mois de septembre."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa56c18d3c8>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(sorted_data[\"CO2.1\"] - sorted_data[\"seasonally.2\"])[:30].plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La courbe correspondant à **CO2.1** semble au premier coup d'œil être une courbe exponentielle. En testant quelques valeurs, nous pouvons trouver cette courbe exponentielle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa56c0aab00>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "a = 31.5\n",
    "lmbda = 0.0019\n",
    "b = 284\n",
    "sorted_data[\"approximation\"] = [a * math.e**(lmbda*x) + b for x in range(len(sorted_data))]\n",
    "\n",
    "sorted_data[\"approximation\"].plot()\n",
    "sorted_data[\"seasonally.2\"].plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Maintenant, essayons avec cette approximation de prédire la concentration de CO2 dans l'atmosphère dans les prochaines années."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fa56c006c50>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "nb_months_to_predict = 400\n",
    "\n",
    "last_period_in_data = raw_data[\"Date\"][raw_data[\"Date\"].keys()[-1]]\n",
    "dates = []\n",
    "for i in range(1, nb_months_to_predict+1):\n",
    "    year, month = last_period_in_data.year + i//12, last_period_in_data.month + i%12\n",
    "    dates.append(pd.Period(year=year, month=month, freq=\"M\"))\n",
    "\n",
    "a = 31.5\n",
    "lmbda = 0.0019\n",
    "b = 284\n",
    "approximations = [a * math.e**(lmbda*x) + b for x in range(len(sorted_data), len(sorted_data)+nb_months_to_predict)]\n",
    "\n",
    "\n",
    "data = {'Date': dates, \n",
    "        'approximation': approximations}\n",
    "\n",
    "df = pd.DataFrame(data)\n",
    "df = df.set_index('Date')\n",
    "\n",
    "df[\"approximation\"].plot()"
   ]
  },
  {
   "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
}