generate-geoparquet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c5df7296-090b-4eb3-8ccd-99198ab466a4",
   "metadata": {
    "tags": []
   },
   "source": [
    "# 1. Convert Sentinel-2 Grid KML to Geoparquet\n",
    "\n",
    "The official Sentinel-2 grid is sourced from this ~100MB KML file:\n",
    "https://sentinel.esa.int/documents/247904/1955685/S2A_OPER_GIP_TILPAR_MPC__20151209T095117_V20150622T000000_21000101T000000_B00.kml \n",
    "\n",
    "The following notebook converts this file to the cloud-native file format [GeoParquet](https://github.com/opengeospatial/geoparquet) for easier visualization and batch processing of other datasets."
   ]
  },
  {
   "cell_type": "code",
   "id": "a802d46a-70dc-4331-afd9-e6eeebc74209",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:54:09.573787Z",
     "start_time": "2024-04-15T13:54:08.850060Z"
    }
   },
   "source": [
    "import geopandas as gpd\n",
    "import dask_geopandas\n",
    "from shapely import wkt\n",
    "import utils\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)"
   ],
   "outputs": [],
   "execution_count": 1
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:13:44.686149Z",
     "start_time": "2024-04-15T13:13:33.705982Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Load KML file directly from URL\n",
    "gdf = gpd.read_file(utils.URL_ESA_S2_GRID_KML, \n",
    "                    engine=\"pyogrio\", force_2d=True, \n",
    "                    columns=[\"Name\", \"description\", \"geometry\"])\n",
    "gdf.rename(columns=dict(Name='tile'), inplace=True)\n",
    "gdf.head()"
   ],
   "id": "38475971a295b1d8",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "    tile                                        description  \\\n",
       "0  01CCV  TILE PROPERTIES<br><table border=0 cellpadding...   \n",
       "1  01CDH  TILE PROPERTIES<br><table border=0 cellpadding...   \n",
       "2  01CDJ  TILE PROPERTIES<br><table border=0 cellpadding...   \n",
       "3  01CDK  TILE PROPERTIES<br><table border=0 cellpadding...   \n",
       "4  01CDL  TILE PROPERTIES<br><table border=0 cellpadding...   \n",
       "\n",
       "                                            geometry  \n",
       "0  GEOMETRYCOLLECTION (POLYGON ((180.00000 -73.05...  \n",
       "1  GEOMETRYCOLLECTION (POLYGON ((180.00000 -83.80...  \n",
       "2  GEOMETRYCOLLECTION (POLYGON ((180.00000 -82.91...  \n",
       "3  GEOMETRYCOLLECTION (POLYGON ((180.00000 -82.01...  \n",
       "4  GEOMETRYCOLLECTION (POLYGON ((180.00000 -81.12...  "
      ],
      "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>tile</th>\n",
       "      <th>description</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01CCV</td>\n",
       "      <td>TILE PROPERTIES&lt;br&gt;&lt;table border=0 cellpadding...</td>\n",
       "      <td>GEOMETRYCOLLECTION (POLYGON ((180.00000 -73.05...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01CDH</td>\n",
       "      <td>TILE PROPERTIES&lt;br&gt;&lt;table border=0 cellpadding...</td>\n",
       "      <td>GEOMETRYCOLLECTION (POLYGON ((180.00000 -83.80...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01CDJ</td>\n",
       "      <td>TILE PROPERTIES&lt;br&gt;&lt;table border=0 cellpadding...</td>\n",
       "      <td>GEOMETRYCOLLECTION (POLYGON ((180.00000 -82.91...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>01CDK</td>\n",
       "      <td>TILE PROPERTIES&lt;br&gt;&lt;table border=0 cellpadding...</td>\n",
       "      <td>GEOMETRYCOLLECTION (POLYGON ((180.00000 -82.01...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01CDL</td>\n",
       "      <td>TILE PROPERTIES&lt;br&gt;&lt;table border=0 cellpadding...</td>\n",
       "      <td>GEOMETRYCOLLECTION (POLYGON ((180.00000 -81.12...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 2
  },
  {
   "cell_type": "code",
   "id": "1e68911e-a5c7-42ad-a72a-44d5b299be15",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:13:45.892242Z",
     "start_time": "2024-04-15T13:13:44.687033Z"
    }
   },
   "source": [
    "# GEOMETRYCOLLECTION contains polygon and centroid. If polygons cross antimeridian split into 2\n",
    "# Assume first geometry is main footprint polygon:\n",
    "gdf['geometry'] = gdf.geometry.apply(lambda x: x.geoms[0])\n",
    "\n",
    "# Extract UTM_WKT and EPSG from \"Description\" column\n",
    "gdf['epsg'] = gdf.apply(utils.get_epsg, axis=1)\n",
    "gdf['utm_wkt'] = gdf.apply(utils.get_utm_wkt, axis=1)\n",
    "\n",
    "# Drop description column \n",
    "gdf = gdf.drop(columns=['description'])\n",
    "gdf.head()"
   ],
   "execution_count": 3,
   "outputs": [
    {
     "data": {
      "text/plain": [
       "    tile                                           geometry   epsg  \\\n",
       "0  01CCV  POLYGON ((180.00000 -73.05974, 176.86462 -72.9...  32701   \n",
       "1  01CDH  POLYGON ((180.00000 -83.80855, 174.71288 -83.7...  32701   \n",
       "2  01CDJ  POLYGON ((180.00000 -82.91344, 175.74819 -82.8...  32701   \n",
       "3  01CDK  POLYGON ((180.00000 -82.01866, 176.55270 -81.9...  32701   \n",
       "4  01CDL  POLYGON ((180.00000 -81.12317, 177.19616 -81.1...  32701   \n",
       "\n",
       "                                             utm_wkt  \n",
       "0  POLYGON ((300000 2000020,300000 1890220,409800...  \n",
       "1  POLYGON ((399960 800020,399960 690220,509760 6...  \n",
       "2  POLYGON ((399960 900040,399960 790240,509760 7...  \n",
       "3  POLYGON ((399960 1000000,399960 890200,509760 ...  \n",
       "4  POLYGON ((399960 1100020,399960 990220,509760 ...  "
      ],
      "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>tile</th>\n",
       "      <th>geometry</th>\n",
       "      <th>epsg</th>\n",
       "      <th>utm_wkt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01CCV</td>\n",
       "      <td>POLYGON ((180.00000 -73.05974, 176.86462 -72.9...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((300000 2000020,300000 1890220,409800...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01CDH</td>\n",
       "      <td>POLYGON ((180.00000 -83.80855, 174.71288 -83.7...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((399960 800020,399960 690220,509760 6...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01CDJ</td>\n",
       "      <td>POLYGON ((180.00000 -82.91344, 175.74819 -82.8...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((399960 900040,399960 790240,509760 7...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>01CDK</td>\n",
       "      <td>POLYGON ((180.00000 -82.01866, 176.55270 -81.9...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((399960 1000000,399960 890200,509760 ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01CDL</td>\n",
       "      <td>POLYGON ((180.00000 -81.12317, 177.19616 -81.1...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((399960 1100020,399960 990220,509760 ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ]
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:13:46.310023Z",
     "start_time": "2024-04-15T13:13:45.893686Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Add simple UTM bounds (left, down, right, up)\n",
    "gdf['utm_bounds'] = gdf.utm_wkt.apply(lambda x: wkt.loads(x).bounds).astype(str)"
   ],
   "id": "26f3a11e1d5ed680",
   "outputs": [],
   "execution_count": 4
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:13:46.370712Z",
     "start_time": "2024-04-15T13:13:46.310911Z"
    }
   },
   "cell_type": "code",
   "source": [
    "print(f\"Number of tiles: {len(gdf)}\")\n",
    "gdf.to_parquet(\"sentinel-2-grid.parquet\")"
   ],
   "id": "b63ecd0312ed2f6",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of tiles: 56686\n"
     ]
    }
   ],
   "execution_count": 5
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## 1.1 Only keep tiles that intersect with land areas",
   "id": "e14b7a350b5a31de"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:15:53.437150Z",
     "start_time": "2024-04-15T13:15:52.129641Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Load high resolution land mask\n",
    "land_highres = gpd.read_file(utils.URL_NE_VEC_10m_LAND_GEOJSON, engine=\"pyogrio\")\n",
    "land_highres.plot()"
   ],
   "id": "bba4855166b195c5",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAisAAAEbCAYAAAAbPMQYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAAC72klEQVR4nOyddXgc57m375lZZq2YGczMGCdxmKGhtmnSNGlKOWlO27RfT5NCUjin3KYcakNN0jCDncQQM5Nsixl3tQwz3x8ryZYl2ZIstOe+Ll/J7s7OvLuafec3z/s8v0dQFEVBRUVFRUVFRWWcIo71AFRUVFRUVFRUToYqVlRUVFRUVFTGNapYUVFRUVFRURnXqGJFRUVFRUVFZVyjihUVFRUVFRWVcY0qVlRUVFRUVFTGNapYUVFRUVFRURnXqGJFRUVFRUVFZVyjGesBnC6yLFNbW4vVakUQhLEejoqKioqKisoAUBSFjo4O0tLSEMWTx04mvFipra0lMzNzrIehoqKioqKiMgSqqqrIyMg46TYTXqxYrVYg9mFtNtsYj0ZFRUVFRUVlILjdbjIzM7uv4ydjwouVrqUfm82mihUVFRUVFZUJxkBSONQEWxUVFRUVFZVxjSpWVFRUVFRUVMY1qlhRUVFRUVFRGdeoYkVFRUVFRUVlXKOKFRUVFRUVFZVxjSpWVFRUVFRUVMY1qlhRUVFRUVFRGddMeJ8VFRWV3niCEcw6iVZviMaOIN5gBE8wQjAi4w9F0UgC7b4wggBaUUQQQKcRsRm1xJl0mHQSNoMWu1GLUSeN9cdRUVE5y1HFispZT1RWKG3soNEdpMEdoLLVh04SCUdlZAVEAarb/fhDUbKcJlq8IapafSRa9TR2BEGBVIcBp1lHgkWPXiOi04gEwlEsei1xJi0rihMxaCREcfD9q6KygtT5Ppc/jDcQobzFS0WLF08wyu4aF82eIGXNXnLizcRbdGQ7TVS2ehEQ2FvnpqbdTyAsY9JJiIKAXiMSisYeCwgEIlGCYRmtJGDWa9BKIlpJICorJNkMLMqLx2HSYjVoSbDomJfjxKxXpw8VFZXRQZ1tVM5IXL4w5S1e/OEoLn+YQ/Ud6LUinkCEmvYAaw81YdFLJNsMmPUajjR58IWiyLJCKCLTEYx070sQQFEGd3xBAItOgygKZDqNvLarjuIUK1PTbTG3RiUW/dhe2Y47EMYfiuIJRmhwB8hymrhyVjoH693kJVqobvOz5mAjB+o7yE+0kB1vQq8RMes1GDQSigJ7a90EwlHqXIHu42slkVBE7jEuXygKgCcYe1yQZMETiJCk06MRBbSSSEcgQlRWMGhFTDoNCVY9eYlmdJLIO/saSLUbCISjXDg1lXpXgFBEJtGqx6AV8YViYxAEkAQBp0WHVa9Rm4yOIN7O8ybeosdu1HY/H47KaERB/e5VzggERRnsNDy+cLvd2O12XC6Xard/luINRqhq87Gjsp06VwCLXmJWdhwOo45/b61iU1krLZ4QggCyopBqM1Lr8tPqDRGVFRQFQtFjF3WrQUOq3UC7L4yvU0ScDKNWQq8VMes0mPUSMzMdzM12kuYwYDFo8ASiVLX5cJp1bDjSwmPrywGQRAGzTkJWwBeKIHf+Ei16TY/HXcfwh6PD/dUhCvQ4zvFoJYHCJCveUISmjiC+UJRUu4GCJAuFSVZavUE2HG2hIMmCJAhIgsCGshYC4dh3me4w4jTHojA3zM+kKPnU/T9UjhGVFZo6gqTYDd3PKYpCQ0eAVk+QmvYA7d4QClCQZCIr3kowLNPmC5HhMBBVoKYzItjsCXHh1JTuCN1QcfnCINBDFJ2KQDhKdZsfs14i1W7EG4xQ1uyl1RviUEMHcSYdq6ckYzUMfJ8qZwaDuX6rYkVlwqAoCoGwjEErIggCsqywp9bF81uq2V/vZneNq/tCqZNEVhQnkpdgZkq6jcpmH3EWHWkOI099WomiKIiCwKUz0th4tIW6dj/BiIxJp0EQoLEjSJM7gC8c7Y40HI9GFNBpRBKterKcJublOFlRlIjdqOHZLVUcqu9gb20Hbb4QwUjsDjfSnyqYINiNWmRZ6Y46WfQaUmwGdBoRSRRIsupJsOhZVpRAttNMbqIZyzheKnL5QjR7QjR2BDja5GFymo30OBObylopTrZ2Ckw9CRYdRq2E06zrFaWIRGU8wQiKAjajtlsMKEpMBA9k2a8jEKas2UtVq49IVGFXjYupaTb21LgpSrGwKC8Bp0XHtoo2kqw6WtwBRI3AvjoPsiwTCCuUpFjRiAIHGjrwhaLkJZpZnJ9AWbMXg1YizW4gyWY45VhORWNHAJNOM6S/ayAcZWtFG79+7xCby9vQiAJGnYRJJ5Edb+YHl01mSpr9tMeoMnFQxYrKhEVRFGQlNrFpJAG95lhyZ2NHgC3lbWTEGZme4UCWFV7bXcfL22uocwUob/ESb9ERDMu4A+Fu4dJFqt3ATfOzKGv2EorKfHCgEV8oSkackYIkC3tqXERkhXZfGIgJnniLjoisoJNiwiTVbiA/0YIoQLLdwLKCRLLiTT2OEwhH8YWirDvczL3P7SAcndA/sX6JM2n54rI8CpPM1LYHuHRGGgkW/VgPC4BD9W6e3FiJJApY9BqKU6xsrWjj4mmpfPPfO3Ca9Zw/KYlQREYBmjxBqlv9uPxhatv9tHhjkTiLXsPMTAerp6Rwzex0TLqeF+lIVEYUhCHlIvVFVFYQiImccCRKqy+MQSv1imQoikIwIscSozuX7qrafJSk2E47enI87kCYdm+YVIcBrTT04lFZltlZ7eJgfQeiEIsqvrCtmi3l7d1RzRSbAbNeIhxVeOqOBWTEmQiEozR1BElzGIf1c6mMD1SxojJu8YUiNLiDHG3ykOYwUpRs7TEJeYIRnttcRW27nw8ONJKTYAZgTnYcuQlmlhcloteIvL6rDptRQ0mKjSSLHlEUWHekmee3VrOnxkVjRzCWICtDWJYpTraSbNOTG29BqxEQBYGjzV7qXQGisoLLH6am3d89DlGAZJsBk04i02nCrNNw3uQkLpuehkYSu/ND3P4wla0+ypu97K114/KH8YWjBMNRDtR3jPr3O1qYdRKXTE9lVlYcSwviEQSBOJNu3CTd7q5up6rVzx/XHuZAXQdRRSEn3owoxCIgO6raUZTY5yhOsbK8MJGceBNLChMx6yVaPCFsRm23SGjxBHGadbT5wmg7RbRWUvNBBktUVhCFWJfdsmYvRxo9tHpDNHuDVLX6CIZlMjtzth5bV8Yzm6uIN+vIcJpwmnSsKE7kujkZaE5DOKmMH8aNWMnJyaGioqLX83fffTd/+MMfuPXWW3n88cd7vLZgwQI2btw44GOoYmV8oygK6w63EGfWUtbs5XCjh+aOIGaDhrwEM5fNSOtxt+oPRbvDxbGLRpCqNj/eUJTXd9Wxekoy/3VeEZ+UNvPM5koaO4Ik2wx8+4JiEq0GQGFHtYvadj/hqMykVBt1rgBTUm18criZnVXtpMcZ2V/XwfojzfhCUXLiTZS3+HqNfVamA5tRy6RUG0sK4lmSn4AoCgTCUR58dS87qmJVOG3e0IRf4hkMJp3EXSvyKEm2kRJnQC+KOC2x5ZLxcvGWZYWadj9GncjWinbe2dtAXqKZZk+QvbVuDjV00O4L85m5mfzkqql4ghGe3lTJusPNxJn1XDw1hanpdjKdplMfTGXQhCIyOk3fgqO23c+j68qwGbQkWmM3Ila9BqtBS5rDQJbTRGNHAAWBdIdxlEeuMpyMG7HS1NRENHosKXDPnj2cf/75fPjhh6xcuZJbb72VhoYGHn300e5tdDodTqdzwMdQxcr4oCMQ5u29Deyvc6MoEIhEURSFe88vJtGqR1EU9tS4+bSsBbNeg92oZUVRYq878b21Lj4pbebaORnEdy4p9DexRWWFbRVt7K934zDqEAVYWZKERa/BH4ry+IZyipItzM+NZ3N5K//aWMFHpc0kW/UUJFmocwXwhWLJrznxZgQhJpa0ksiVs9JZXpjA3Jy+z8XGjgDBcJSPS5upbPVxuNGLPxwhGI4JpJd31OAOnDwxd6KiEWPLHjMzHOQkmMiMM3HFzDSy4s1jPbR+qWn38+zmKt7f30BtZwTtZ9dMZ1VJUvdd+oF6N43uIPvr3Jj0Gq6alT6uc26GgiwrtHhDSKJAnEk74uKyzuVHAJKsekRRJBKVe0RFvMEIda4AeQnmXktp7b4QL26rYd3hZnbVuHD7wxQmW7h6Vjpzc2IJ9JlO07gRyKPF8VYGE51xI1ZO5J577uG1116jtLQUQRC49dZbaW9v56WXXhryPlWxMnbsrGpnT6e4+Li0GYNW5IIpKVgNGhrcQZYWxKPVSOyrdTE5zUZJio2ceDM6jUgwEuWFrTU8svYwP79mBovy47v3qygKz22uQlZgQZ6TeleABneApo4gzd4Qbd4Q0zMdXD0rnff2N/D6rjrq3QH0GpHVk1PYV+cmFJUpSbai1wjUugIIgoCnUzy0+8OEozJ58WY0GpGOQJhEi56KVh9bK9pIcxhZlOdkXk48SwsTTvodNHYEqGsP8OK2at7b30ity0+8ObYcUu8KkGwzdC8luPxhWr2hU1YXjWeMWon5uU6unp3OlDQbZr2G9/c38Jl5WaeV0zBaeIMR1h1uJsVuYHqG45TbB8JRdJI4bDkpI42iKDS4g9S5/NiN2u5z7miTl53V7aw52IQATE6z8eAVUyhJGZs5MxSR8QXDtHiCuP0RMp1G9FoNrkCYw40eDjV0sLPaxQf7G/usgku26Tl/cjJ17QHa/WFEAUJRBZtBw9Q0G5fPTCM33oJhjAwNOwJhdBqxO+fuRJHWF3UuPwkWfY/fUSAcxaA9c00Zx6VYCYVCpKWlce+99/Ld734XgFtvvZWXXnoJnU6Hw+FgxYoV/OQnPyEpKanf/QSDQYLBYPdjt9tNZmamKlZGGUVRaOwIEorIeEMRatr85CaYu/NMXL4wT26soMUbpLY9gEkr0RGMTUrXzs5karqNdl+YpzZVcqTJQ1GSFU8wgkkn8fQdCzja4sXtjXCw0UWHP0plm4+OQIS1h5roCERYUhDPz6+ZgT8Upbrdx95aN5fPSCPTaeL/3jnI7z44TKo9FjK+bm5m9x1yusNIKCpzpMlDVasPV6dwCUZkcuPNZMWbyI03k5NoxjbIUkpfKEKLJ4RRJ3Uf78SJJhKVqXMFOFjfwZaKNh5dV0bwBC+U8Y7DpMWs0+ANRbh+biZ3r8xHr5FUp9sxwB0Is6fGxZu76yht8HC40y9IK4kUJFnITTCTZjeQ4YxFwHITzJ1RjtMXX+XNXpxmLQatpt8lneo2H/WuAEebvAQiUWrbAxyqd7On1k2LN4SsKKTZDThMOhrdAZo8IXQaEa0o4A0NvVQ/L8HMbUtzuWVh9pD3MRjafSGONHnJizdx91Pb2VjWgkWnYVF+PHEmHa/vruO/zi/icGMHsgyrpySzoiiR0kYPSVY93mC0zyTmMymK0hfjUqw899xz3HTTTVRWVpKWlgbAs88+i8ViITs7m7KyMr7//e8TiUTYunUren3fVQUPPPAADz74YK/nVbEyNvRlPOUPRTnU0EFHIEKbL+ZlsrfWxbbKdoKRKJIoYtVryE0wU5JqpSMQQRIEDDqJpQUJJFh0fHCgkZVFSdhNxwTD+sPNlLf4EAUoSbFSkmrjwwON/Gd7DUeaPGglkZe/uoTNZW387K0DCAIcauggJ96Myx+mzhVAFEAUBDSSQKBzyeZok4dgRMask5iR6eC/LyhmVlbciH93iqLwcWkz7+9v4IVt1XiCw++jMhKYdRI6jcgtC7MpSbFi1EloJZFp6XYcJt1YD29QdE1/E20pweUP88c1h3n600rcgQiiAKl2Iwtyndy2NJeCJMuI3JF7g5HupVtvIILZcGyZLBiOUtvuo6Y9wIvbathf38H+Ovewj+Fk5CaYmZ5hZ0l+PBdPTxv1Zbw2b4iXd9Tw4cEmQhGZaRl2BAHizbHfhSgIJNkMLMh1kmwzsKfGxa5qF/vqXLFSd0FgYV48F01NmTDRvNNhXIqVCy64AJ1Ox6uvvtrvNnV1dWRnZ/PMM89w9dVX97mNGlkZW6Kywi/fPcgLW2vwhSJ0BCPYjVqmpdvJSzAjK9DUEaQo2cLlM9MoSIoZgXmCERrdAcx6DQkWfb93C5FOi/sT79Q6AmFq2wNoJKhq8eMNR5if4yTRaqDBHeDVnbUxkzdFoSDRgiAIJFr1LC1IQOpMivUEIjR2BBAFgTizlgSLAUkUiEZlvKEoZp1EWFZGPeza5A7w1t56fvL6fgLjLMoiCLGln6isEJUVIrLCBZOT0WslZmTaSbUbOac4acJGVTzB2IX+xJLk8UqLJ8jaQ028tKOWjUdbAJiWbue+1cXMy4kb8SoZRVF6CDu3P4ysKAgogEBlq5e1h5p5bVfdqFTDCQKsKk4i2W5gVqaDdIeROleAOpefO5bloe/8LQcj0R42CP3h9seWbwY7BwTCURrcAbJPyNuqafez8WgLeQlmHCYdOfFnX47NyRiMWBmVX2hFRQXvvfceL7744km3S01NJTs7m9LS0n630ev1/UZdVEaeQDjK/roO6t2B7ufafWE+7sxbATBoxZjRVrOXgkQLgYjMTfOzcJp1WA0x4yx3IIxWFDHqJLZVtLG5vJV6V4DtVe0EwlGeu2sRNoMWdyBMRyDCprIWXthawyeHm9FJIvNznQgInDcpme/9Zw8fHWrq4UK7MM/JT66ahiQKvLevgd99eJidVe1oJYGvrSrEYdJS2uBhc3krpY0eorLCDfMyqXMF2F3jQiC2hDMvJ46vnFNA4Qi6rybaDNw4L4tQROaX7x46rfD3cNG11BOORlmUn0C6w4gggMcfxmzUUpxsZUqarVuMTlQmUgJtozvAc1uq2F7ZTjQqc/6kZJYVJnDFzPQhiUVFUShv8fLC1hoWF8SzOD+WnxWVY88nWfUEQlHsRi06rUQ4KtPqDaGTRG57fDNVrb5us8SOQIRIZ4uGFJuBZk9oWD/78STb9OTEx2wMtJJAozvIkSYPH+xvpN4da/WgEQUON3rYVeOiobMFhcWgwWbQMi/XyY3zMtFpJLLjTTEXa+C9fQ189z+7EYDZ2XEkWPTUu/zkJ1k4b1IyC/Pi+xUxBq3US6hAzAJBrxF5Y3cdrd4wxSkWvrQ8f8S+mzOZUfmlPvrooyQlJXHJJZecdLuWlhaqqqpITU0djWGpDBKXL8y9z+3ggwONvV6z6mPrs1fMTONfn1ay/kgL2yrbgVip6/46N2sPNaHTiFw7JwOTVuCDg80UJln47wtK+L93DxGKyIgCPHrr/O58kbImL4+uK8OglShtjN2phaIynxxups0bJDPOyOUz0nj/QAMrixO5b3Ux/nCU2VlxSKLAmoONfPXpbd0GcbOz4rhmTgZ3/3Mrhxo8PZL33txTz0+unMrK4gQSrQZe2VHD0SYvf1xzmEynmbtX5o9Y1CWKQla8iXNLknhnf0MvQ7vRxuUPE4kq2I1a6l1+ChLNVLfHPGmWFcVx3uTkMR3f2YSiKLR6Q1zzp/VUtcYqmZYUxPODyyaflivtT17fzz8/jVlL/OXjo6TaDfz2hlnMyHSQaNHz0zcP8JOrplHd5uON3XWYtRrCssxlM9JYXpjI+iPN7Kxy9bhJCITlPm0AhpNgRGZfrZtN5a199uxSlFgLiTd21/cYmzcUpcEdpLTRw1OfVvZoNWHWST1uErpuvCCWDByMRJmcahvU73935/LOO3sbqGmPJc+2eIMcavAwK8uBRa/BYdJhNWgIRWTiTDqy401ndELt6TDiYkWWZR599FE+//nPo9EcO5zH4+GBBx7gmmuuITU1lfLycr773e+SkJDAVVddNdLDGhAN7gA7qtqx6DXMyY7DoJUGZaM9FOTOX894XK8MRaPcsiibklQr/9xYicsfxm7UYtFr+OX1M5iR6eDhN/ZT7w6Qajdg1Ek0uAJ4Q1HWHmqK7SMi89Snld37PNrkJSorzMx0sKmsFVmBb72wk8dvm09Jio0ZmQ5+fcMsAPbXufm/dw6x8WgL55QkkZ9oJt6i59OyVmZnOZiR4SDJpifJemwC//BAY48L//bKdv7+cRmJVj3lLb4eYiUclfnWC7vwhaIkWfV0BCL4O83dbEYtaw42cvmMNOblxnG0ycvkVDvFKcMTWdBrJCpbfWwsa2VejpO52XG8sqMGVyAyonepfZFg0XVOrCHCUZl9tR18WtaGIMCl09NIsqqRzdHknX0N/P2To4jQ3bZBFIQh99Jx+cL8Z3s1/1hX1qMv1NKCBNIcRr7xzHZ2VLVz1ax01hxs5O+flKGTRN7vvEl54NV9zMx0cG5JEg6TDoNW4tWdtcPwSQdGl8P0yYjKClFOnuFw/Gc/WTRza2U7BUlWTMdF4XZWtfPyjlpuXpiF06RjV42Ldl+IHKcZnVbkaJOXNIeB375/+DizyWPLYh8caGRWpoODDR1Ut8VeN2olZmc7+N7Fk8lLNI8b0RKMRGn1hki1j62nzYjnrLzzzjtccMEFHDx4kKKiou7n/X4/V155Jdu3b6e9vZ3U1FTOOeccfvSjH5GZmTng/Y9k6fLhRg+v7qwlwaKjJNXGvH48N06HFk+sGVxUVqhtj4V5dZLIjEw7GXEmbAYN4ajChVNTyIgzdq93nrh2PFL4gpEeP9IuKlt8NHuDTEu39yq1+837pXx0sJF6l5+oAlajFos+dofeNoCJBuDWxTncsTyPeLOuWyS6/GEcJl2P9ee7ntzKO/vquyeekhQrT9w+v1uwRGWF331Qyt5aN0eaPFS3+TFqJVLthpOuqes0vTsWS6KApvNfolVPboKZuTlOpqbb0UoC09Ltp92MzReKOfjqtRJGrUSLN8A/N1ZR1uw9rf0OBodRw0XTUqlt86PVCNxzXjFNHUHufHIrZr2Ezajl/XtXDDk/ostCPbmzr9B4IhSROdrsQS+JNHsCbC1vJy/JQrxFx9pDzWTGxSbsSam2WD+eUSrX7giEqWzxsuZQM+/ua8DtD5MeZ+Sa2RlcODVlwBe2DYeb+M5/9lDRR/TDadbx77sWsa2ijf95eS8RWSbZZsBm0BJn1lLT5qey1dfjIv/N84sIRKL84cMjw/VRxy15CWYumZ5KcYqVB1/dR1NHsM/t0h1GGtyBIRtFaiWBcFThmtkZLMhz4g1GultAuP1hQECvEUmy6ZmWbueCKSlAbLlpOFs/QOy3+qv3DrEwN55zSvqv0h0q4zLBdqSYaD4r0ajMW3vreGdvA3WuIIcaO/pslHcikiiQajfgCUYIR2TMeomCJCuFSRYkUSTOpKUgyUK7P0xHIExugoV5OXGDqs6Iygr769xsrTjWf+dPa4+wpbyV3980m0yniW2VbaTZjSRa9bR4gmglEUcf5lKPry/jcEMHrkC023PAoBFp8oTYVtE24ETSZJueVSVJzM9xcsn0NHQaEUVRqG7zk2zT89AbB3hyY0WP7++KmWnce35R9xpyVFYobeygrLN88tOjrYQiMk2eAAIC/lCUqKJg1msw6zS0+0OcPzkFly/Ev7dWU+cK9Dm2eLMOlz+M0xyLRMzIdJCbYOKOZXlDFpLPbKrg9V31xFv06LQCeknEHYjw5gkh7ZFCrxFJdxjIdJpo7AiSbjfS7g+TFW8i1WagINnC1DQ7eYmWCV9SqSgKu6pbqW4NEG/R0eYNsbWynXZfzJvko9ImpqTZOdLkRSMJtPtif2udJDItzcrKkmSumpVOrctPptM0oATO4aartPWDAw18cKCRJfkJTM90EG/WodeIvc7DBpef2x7fwt7a3lU6Oo3Ip/efy8s7avjR6/tPOSdB7MKck2Bm3eHmU26rMjI8dNU0blqQ1eM5WVZ6iZYdlW24AhFmZtpp6gix7nAzCZ3O03FmHYoSE0pmvYZEy/CUt58KVawME7KsIAjDV9boCUZ4bN1R/ved/hOIhxMBSLHpMek1OM06zp+UzIri2PJJmzfE3z8pY3tVO9nxJgQEjjZ78ASjGLQi2yvbe6zpOkxaLpqayrv7GpiUasVm0PL67jogdtG+YGoKWc6Yj8Pc7DhkReHZzVVUtfnRSgI1bX4+PNjU71iTrHpuX5rLR6VNHKz30OwJEm/WUZRs5XOLspmV5UAQwBeSeW1nLb95v5SiZCvnTkrivElJRGT40Wv72F3jIt1hxG7UcMvCbM4tSWZjWQvnFCd1l1zKssLNf/uUDUdbMGpFsuPN3Dg/i6tnplHnDqLXiuyobGdxQTw7q1wcburgqU+raHAH+vRE0UkiFoOG2ZkOlhTGs6wwiYIky5D+ZlWtPtYeaqK6zU9+opnVk5P51XulPLelCt8YJd7Oz3Fy4dQUlhclkp9onrDVDLKsUOfyo5cEdte4+PhwM0eavOyt7UArCf2K0r7o+m3oOsX6jMzYMqTdpMGkjTVOnJxqG9KE/+nRFv7y0VEW5jm5dUlur8glCui1PYWINxhBURSONHv40hNbaXAHO0tm9RSnWDh/UjKzs+PIjjdxtMnLPzdW8sK26h7H1WtEdj9wAc9tqeLfW6tJtRnwhiIcbfL26JulMr6QRIHF+fFkOQ28t78JTyBCVFHITYjNQU6zlhZPiKgsc9OCbC6dnkqrN8wtf/+03+iQ3aglJz42n8/JjuPyGek9bCSGC1WsDBMvba/h528d4PKZ6Vw6PXVIYd/DjR5u+utGfKHouHEu1UoCRm3MG6Oru2ya/fRClydi0IroJJFgROaiqSncuTyPF7fXIIoCWlHApJOIM+lJjzOyszrWVK6mzU+dO8BN8zOZmmYjEJFj1vlhmdwEM2VNHpJsBg41dPDvrVV0+CPEW/R8blE2Wknkbx8f5fENx3pRTUmzsTg/nr9+XEZeopk3vr4MdyBMWZOX+/+zm6NNsaUVi17iM3Mz+dKKfJJtBtz+EPWuAEUpNiJRmX9vrcZh0hIIRdhc3sb7BxppcPf8kZt0EpNSrVw7J5MFuU4y44xoT+NOu7Shg48ONRFv0fHwmwd6HW+4yHKaaPeF8IejpDmMWPQa8hMtZMebyHKamJxmY3KqbcIKFIjlzZU1e9lZ0UZZqxetJFLT7ue1XfUjWnnlNGlZPSWFy6anMjfX2Svy0tgR4ONDzVS0eMlNNHPxtFT0GolmT5CyJi9FSSaqXUH217rxhaM8t6WKxXlOrp2Tyas7a2nsCGE3afm0rJX9tW7uv7iEzeWtfHigqU/XV4gJklUlSdyyMJs/rjnMusMtPV5fWpDAH26ezaPrynh0XTkuf6x/UmGyhR+/vn/EviuV00MniegkAc8AzueufkqDEaBfXJrL/7t08pDH1x+qWBkmXtpewz3P7uh+nGDRcfXsDBblxbO8MAFpgMLFE4zwSWkzT2+q7E40PVtIdxi5aGoy8WYdkiggI9DmDVHR4kUnCiwpSiTHaeZHb+xjf11Hd+g5zW5AFAWWFSaS7jDgDUVp9YSwGTVUtHg7qw68nFOcRJrDyO8/OEzHScTgeZOS+Pm1M3h1Zy2PrDnSo/Qa4PIZaXz/0skknpA8+oOX9/D4hgr0GpF/fXEB9e4A0zPsfOHRzRxp6juPpKuTr9OsY0lBAjfOy8QwCB+PZk+Qe57ZQYs3RE68iWZPkJ3Vrl45NKfLvOw4bl6YzQVTUugIxJKl9eMkqW+4iESiNHkCOEw6Dta7eWxdORuPtlLfMbpJyyadxOwsB3evLGBxQaxEWFEUbn10M00dQS6amsKK4kSmpdt5cVsNn5a1sOFoS3f1z2hg0Ip864ISblmYzd8/KeNnbx0AYlHPTKeJihbvqCd7q4wPxoNYmTgmA+OAZk+Iv3x0lL98dJTXvraU57ZUkWwzEAhHuWBKrEtrX1j0GlZPTmZBrrNzmaODPbVu9tW6cQfCw34RGk/UtPv52yfl/b7+yu56JqdYyE+ysqfm2Dp6bWdI/ulNlf29FYD3DzRi0WtOKlQA3tvfyOpfraXNFyYqK1gNGpJtBg43erDqNSTb9H0me967upgvLssj0arvTmL8pLQZjdi3UNWIAoGI3F22vavaxSs7allSkIDFoOH6uZk4zSfPI2r3hUm06jlQ30GrN0hUjpnlDReSKHDZ9FQevHwK9s6cptM1dRuthO/BIooiGgH+svYIL2yvG5HeTEkWHSVpdjoCYQ7Wd/S5XOcLRfnkcAtOs57FBQkEwlE0neH7R9YewRuKcqjBwzt7G/j9h4eHdXz9kWrTk5dkYVeVi4JkCz++ciqKAt95cRcvbqvp3q6xI0hjP8sFKiqjhRpZOQlv7amnstXLc1uqOdzo6fFarJNw7C4YYmvYl81I47sXTyLeHOsAXNXu5+3ddeytaeeTI62AwD3nFaKRxG5Dsuo234ArZM5kujLghwuDNiYmEq16mjqC3eXLJSlWfnDpZEobPQQiEQqTLCzKTxxQNUUwEuVbz+/i5R19l2nmxJtYVZzIdfOyeOrTCj453EJ1W6x6Is1h4HsXT2b15ORT5jH4Q1G+9OQWShs8tPpi5cND/ZUKAlw6LY0jTbHy63MnJTMnO47Zg2wnIMsKb++t56PSZoKRKPecWxRL9o7KrDvSjFUnEYwqvLOvAbNO4qJpqVw7O2PMS/DLG9v5xTulfHS4lY4BdsE2iLF8lKgCA10k0mtElhXEc9OCbDKdJqrbYtUb++vc+EJRUu0GpmXY6fCHeXlHLQfqO5iUGuuHtfFo69A/4GmSZjeQ5jCilUQkUeATNVFWpQ/GQ2RFFSsn4f39DRh1EvFmPdsr28hLtLCrur3H2q1eI3LninwW5DopSrZ2ihiFpo4Ab+2p5+WddWytaBvWcakMHL1G5KpZaRys76Cy1c9jX5jPA6/uxayTmJnpYHF+AguP6/h8PDXtfh79pIw4s45km4EjTR721Lh6GEYdjyDA5FQrv7lhFrkJFj492sKfPzpKklXPf51fiNOsH3CJaYPLz18/LuOZzVUIcMrI0fFY9BrSHUb+3yWTSLYbMOkktAIkO0wD3seJNHuCbClvpdXVwQs7mzAbtBztLAWH2Gc/cSaZmengmS8tHBO/iGhUxhMM88T6Mn71/hEGkooliQJfXZLMyiwjWWmpRBWFV3fW8KN3Tx7dO54vL8vmjhUFmHUajrb4yI43YdJp8AQjXPvI+lGxoFdROV1ONMn7yjn5/PcFJcN+HFWsjCCKonCowUNlq49nNlViN2q457xCsuL7rv44VO/mpR21uPxhUmwGwlGZQERmSUECb+yq49ktVSM+5rOZrouoThL5xXXTWVWSxNwfv9ejqqc42cpn5mVy6+KcHpGAh97Yzxu764gzalhenEQoIqPXSLy1p5bDTf27dOokkanpNr578STm5jiH3Dn1YL2bP3xQilYjsrPa3Su61x9zMh1847xClhcPny+CoigcqO8gHJF5ZE0pb+9r7FcAOM06ipIt3Le6mLkj4E10qnHKCjy/pYK39jbw4cGBRwrmZdt5+kuL0UgiHV4ff/24jCc2VtEeGHgSbprdwM+vnkJBqp3fvFdKdZufYFhmcpqVyhYfH5ykIk5FZazRdBY/ZMWbmJ0VR3Z8rDFklyP4cKOKlVHGF4qw9mAjkajC3BwnqY6TO/3VuwOEwjL/2lTBn9ceHaVRnvlY9bFW9VpJxKCN+YV8aUUB8WYdaQ4jJp3EPzdWsKfGxUs7apmZ6eDBy6fQ6gvx4YFG6lwBvn1hMQVJVnyhmBGTNxhlUqqtRz6LNxDmnxsreWNvPTur2oFYBGd5YQL76tzUtMfybeZmx/H1VQVDFg2yrPBJaRP3/2fPgDP3LXqJvAQzv7phFvmJQyufPhlNHUF2VbfR4g2z6WgLu2vc1Ln8TEmzsSQvjslpNlaWpI768k9UVihrcNHkCfDKrgb21XVQ2+6naQAJoSlWHauKE7hzRR5pDjP/WlfK45/WUNY29DyNFJuB2dkO2n1h1h9pOfUbVFTGISUpVl792tIe5fPDiSpWxoBINNY/Yu2hJial2slN6NnUaltZE/e/tI/CZCsbj7Z257qoDD858SZKUix8dmEOSwoTe73uC0XYU+NmdpaDrRVt7K5xMS/HyYxMR/c2UVnhk9JmlhbEd1d9BcJRFOVYMuonh5v495Zqatv9bKtsx2nScPmMNAqSrdS1B6hs9fLDK6ZiMw7cmK8vPjrUwPde2nvKypA0u4EVhfHcviyPguTRiTKOdVKtLMuUNnr4uLSZbRWtvLm3AQEGtOyTYtXwrfMLuHR2DlpJ5JODtfzwtYMcbvEPOUdIRWU8EGfSEorIQy7Ntxo0yLKCNxRlUV48//righG5AVGrgcYAjSSikUQunpbW67U3d9fyred30RGMcrBhYKF8lcEjCnDt7HT++8ISEq39N3gz6TTMz3XGTLSavBQlW7uFyoYjLaw51Mj9F03ixW1V3PHkFgoSLeQlmrEZteypcTElzcbtS3NZWpDIwtx4REGgrMXLhiMt1LT7eX9/I3etyCfFnslwFHotL0rmR1cI/Ob9UrZ3Vhn1RSgqc/vy/FHrhHy6QiUUibLhcDOL8xPQDiGvZcvRJv6zrZoPDrXQ5Al1ewQNRGfEm7W88fXlOI87Twx6A6XNqvmZyvhHEGK9hOJMOryhSI9+SfdfVMK0dDv3/2c3wTb/gL2z4s06FhcksLQgnnMnxewmal0BAuHomCfKgypWRpyorLC7xo3NqKMjeHoTYbxZx4JcJ1VtfnbXuLAaNISjMhpRHDeGc2PJN84t4ivn5A/YuE8QhF421XNz4ihKji2fXDI9jZd31rGvzs2+umNl1buqXbywrYb7zi/iwqmpWI0acuLNPZZdFEVh3eEWFuQNT87GyuIkEi06/vTREd7e20gkKpNiM+AORPAEI4hCbNkp2zn0JNrRRqeRmJvrHJJQiUZlPjxQz+v7mhCFWK+UyCDuIhflxuE4oYS8MEGPSSviG+OO1yoq/WHWSfzg8ilcMTMNrSgiKwpPb67i+y/t4fzJydy1Io8kq4EHX93XZ/+nE9FKAtfPzeSzi7IpTrb2ugFJP0VKw2iiipXTpN0bwmzQ9Lmm5wtFONLo5cb5WchKrANwuz9EVIZWb3BAoWoAnSTw+cU5FCZZ+Z9X9iArsdLcWxdlcdG0NJrcAT461MCeWjebK1wD3u+ZxKqSJL62quC07wC0kki8RU8gHCUS6T/fIRSReejNAzz05gFmZNh56o6FHGroYEqaHV1nT5bpGTbq3QEy44ZHQExJd/CTK6fz1ZVeUGRafVGe3VzBBwebsRl13LE8/7Rcc8cCs75vC++Y2GtiXk58n0Z1oihgN+txGLW0eEODDndvrXTxj3XlXDE9GY0o4PL4KauoYGa8zFGPRL0ntj+DViQv3kRBkgWbQYPNoOGpzdW4A5Gz8nemMjYIAlw2PY1vnFfY46ZIROCCKcnMynRQkGThz2uP8lFpU3cuXYJFj0knkWTVMzs7DpNOYk+Nix1V7dy+NI8rZ6WNeTflgaKKldPkxLuz4/EEI+yvc5OXaOY7F5WwINdJeYuXhXnx3P/ibnZ0nlAnoyjJwo0LMvnM3Cxu+OtG4oxaPrc4lxmZdhJMWox6LZNSbawoSUZRFOpdfvxhGaNOosEd5NOjLbR4Q2yvbGNz+ZlbQn3+APxLBkJTewdOqwmDVmJSmhOnSUvrSXxwNKJAKKpg0knMOs67RFEUvvnvXfiCEf78ublY+uhcPRRsRi02o6P78dQMBzfWthORYXo/poQTlalp9n4ddQVBYHKqjeIUN2/vbej1errD0J3o3Bd17iAPvbGfXUdq+Z/LpyEKMGdyAVMKcqlsdtEh67EYtRh1Gpo6QiTZ9CRb9byxu47FBQnUtPnZWe0ats+qcvaglQQisoJWEhEFCIRlRCHmCXXJtDQONXTQEQizu+bYjee95xVx+7JcTH04YSda9Gwpb+OrT22jvMXHjEwH/75rEZ5ghEV58eypdTPzuHy8SFSmqs3fK69yvKOKlREkyWrg+nmZ3Y+7WmwrikJJipVkmwGnUWJ7tZt4i46mjiDeYLS78kMQ4NxJSXx2QTYuf4gfXzGFwmRbv26jgiCQepyXRqrd2H2SBju7Df/2/VIO1neQ03miHqh3D6sZ21ixr9bNlvJWsuJNJJ0kX+VU2Iw6JEmivMnN7U9sO6lQAYjICv5QhF+9e5BLpqdTnGKl1Rvi9V21fHigkUhn08Q/3zKHFPvQx9UfVoOWBXm9k4gnEl3lxqIArd4QRxs7mJsbj8OsP+n7JqU5kITqHh4vGXFGblqQxdI8B7c9saPPRHaNKJDuMHDJJCdXz0kjwXlM5BmjMgdawlS2efnJv3aMSqdrlTOf/EQzP7xiKtMz7N03Lr5QFG8ogk4SMWgl2nwhUu1GAuEo5Z05cB2BCLOyHATDMm3eUC+x4g6EufOJrWw4Gqs4S7Ub+P4lJTS4AxQmx3rZHS9UIJZfOdGECqjVQGOGoigoSiyc7QtFiMoKe2vd3PjXjShKrOvlRVNTuGtFHmsONFKQbGVpH5Utg0WWFTaVt7LmYCPBTkW/p9bFpvK2CV0BIQggCjH78idvXzCkfXRVt8T+NgofH2rk92uODigiJYkCdqOWBblONpe30e47lvCp14icPzmZ+y8uJt0x8SaJkaDru352cyXzcpy8srOWv3x0lEunp/LdiyfhMA2sgqq0oYPnt1ZT2epjVpaDmxdkd3fXfvSTozz4Wu/meyatyJtfXUB2spNGdwCbQYuh8wZgS3krhxs7WFKQwBMbKvjrx2XD96FVzkoevnoa18zO6LOdx1AIR2UqWnysOdjIXz462qMVwvN3LSI/yYLDqB3zSr2BMJjr98gUT5+lBCNRBqr9BEHoXrYw6TSUN/v45buHEIBvXVDM+u+sYlKqjfv+vYu9dcciIaeLKAoszIvn+rmZRGQZk07is4tyyJlAiZl9oSixZOaPS5tZd7iZzeWtvLuvYVCJx10/7nBUIRqVWV6czF8/O5s446nzQKKyQqs3xJt76mn2BHtk4AcjMq/vruOX75aeVYnQJ/stCIJAZbOXRblxfHyokT9+eJiIrLCprJWLfvMxP3h5z4COUZhs5f6LJ/HILXP47MIcqtv8fPc/u9le2cb1czOYktZ7Aow3a7EYDVS3+XhuS1W3UPGHovz6vVJe3FaLKIo4TxHZUVE5GZdMS+X9b67gxvlZwyZUAB5fX855v1zLj1/f36tn03V/3sCh+o4JIVQGi7oMNIzoNRLhqIxWGviJ8unRFp7eVMGBeg/XzsngF9dOI8tpRhAELp6WwjnFiWTFD//deF6ihS8uy+PDA43sqXZx0bQU9te6WVPaPKEjLAA3/+3T7v+XRIFsp4nr52XyhSU56E+RgOoLRthe1UqO00y8ScJq0JIVZ6DN33eH5YEiCgKtniCNbj+WxNEpLR5rTjVhZjhNBCJRzHoNiRYtNa4Q5S0+ipItfP3cwkEf748fHmZtaRP769ycW5KEUefgr5+by2W/+4QW77Fk6Yump6PTaXHoNHzlnILu53/9/qHu3jhLfvrBoI+vogKxdhd/vHk2y4uGf3n2f98+eNJGl4kWPeUtXhbk9d1CZCKjipVh5lROf4qiUNni47mtVQRDEcIKTEmz87VVReQnWZBlGUEQCISjJ/UKGQ6y483cND+T3VXt/H19BfNynJj0GtYdbkZWwD3Axm/jmaiscLTZy0/fPMALW6tZmBfPkoIEzp+c3Kd9tF4roddocJh06HQSj2+sYE/d6QmVrnF8eLCZUGQv/3V+IXNzzrzJZLCIooBBI5ERZ+TymRk8v60Gly+MRhT5y0dH+ebq4kHdkXpDEXZVu7hkegqZDj2/eu8QDe4AaQ4DwYjMS3cvItGswWY29imkrMOUBK1ydqLTiNwwL5Mvr8wfkQqbqKzw2Pryfl/PiDPy5jeW4Q8PzQhuvKP+OkcJlz/Mv7dUkWrT4/aHuGRqMk6zAUFQMGg02DurikQxNjmPVvM3nVbD5HQHF00NUtHq48Z5mQgo7Kx2nRFi5XhKGz2UNnp4cmNFd0mfSSeRZDPwyM2zMes1SJ29MZ7eXMV/ttewt9Z96h0PgnWdxnFFyVbOm5xMUbKFqWn2AXvDnCk8v7WKJzdUoJVECpPMfG5xLjurXYiCwLojzeyrczMp1caVs9IHvM/bl+Twz42VzMxwcKDBE2uBUJTI/1w6GbNeQ3WbH7PRgCAINLgDHKjvYHq6HYdJyz/WlVPWfGpfChWVvpiZ6eBn10zHZpT48EAT181JRyOJw7ocE5FlAicRIpGogigIPQoMgpEou6tdzMx0EJGVMWkqOlyoYmWEUBQFfzhKdasPg0ZABq6bm4ndqMXjD2E5TQv24cSo12A3ailMtjI720kwEmZntQujVjpjVfrxVSI1bX6inUm1AC5vgF+8fbBHs8PhpLzFhycYparNj04jcv8FhSwsGL6mg+OZLeUtvLCthgP1Hd2lv/EWHal2Az+9ehoK8OV/bmNfnZt7nt3BW3vqeeSW2aec9BVFwWbQMiPDzu8/PMLywgTOn5zEmgONvLStmqvnZJJk1fPKzlo+KW3urp4oTraikYRhF6UqZxc7qtq54Ncf8Z2LSihOsVDZ5ueZTZUUJVu5bm6sInRrRSs58WbiLUPLhdJrJOblOLvP3S4umprCD6+YiiQK3cnlx7+nq5moIEzs9X1VrAwzsqzQ7A1yqK6DBJueQChKYbKjx2Q7noRKF06Ljr9+dJTlhYkUJzu4elY6Ll+Y57fV4Blif4mJQkcwwo7KdhIsegqSzMzIsJNq1VJ+Go3s+kMUYvlCqXYDG460EJEVvvn8LpKsBn5/02zSh8lAbrwSCIXZWdnGvvpjbScW5TlxmHQcbvJwpMGDfFzS1AcHG3l7bz0XTk3tsR9FUahq9fHy9hpavbEWB3azgT21bgTg1V11vLqrrnv7taXN6CSxl3ncwYaOkfmgKmclP33zALOzHLR4Q90Osp8cbubO5Xn845NylhUmcMP8rFPspX+WFiZ0ixVJFMhNMHPN7AwSracWQP11TR4PPb4GwojGnh944AEEQejxLyUlpft1RVF44IEHSEtLw2g0snLlSvbu3TuSQxo23IEwwUiUSDQWmotEZbyBCB2BMFa9hiWFCZSk2JiZFTchToSp6Q7uWJbHtooW0p0mvn5uEdPSrVw+I4WCRDP6YcxmH4+8srOGdl8IjSii12m5aWHOiBxHVuBwZ+O9roqhmvYgB+o9dARP7ulyJpDutBKIHBMjggBpnd5ANr2Gb7+4mwP1xwREKCLz4Kv7mPXDd9h/XMuDihYf5/9qLf/3XimPbqimsj3E7ho3itJ3E8NwVBlyUzcVlcGwrbK9h9X9moNNHG328vruOkobT6833LmTjkVgL5ySwrv/tZzzJief1j6PNHlo9Ya6I8vj1c1kxCMrU6ZM4b333ut+LEnH1sx+/vOf88tf/pLHHnuMoqIifvzjH3P++edz8OBBrNbxUTGhKAr+YJhmbxhPMILdIOA0G/EGI4TCAnaTrnsdcKLnHUzNcOANRlAUkCSRa+bl4LDUsaXyzA+RP7+1hiNNXh65eQ5Nbh/7RmlZQBIFLHoNF01NIRgen5PEcKEoCm/trkUjClw/N4OrZmUwM9OBQStyoM7FY+srSLEZiMhKj2W6Zk+QL6/MpyApZjNe3uLlvud3Eoyc2d+XyplBgkVHuHNJ2aA9vWtESYqtO9/u6+cWDsuN8ImNT7v26Q6E0QkC0U4/sP7MSEeLETWFe+CBB3jppZfYsWNHr9cURSEtLY177rmHb3/72wAEg0GSk5P52c9+xp133jmgY4y0KVwkIiNJwoSIjgw3siyzo7Kd13fWUN7iZUdNBx3BCKERyuUYayRRIMmqJxxVepi6jQQGjci3LiziqlkZ7Ktz841ndvKv2+dTnDpxjA2Hm6c+reCFbTX4QlFq2nzkxJvQaSQ0ksDVszPoCET4pLSJjUdbCUVlompzHpUJwLNfWsiCvHhW/d8aWr0hNn/vvFNWjfZFMBKlwRVkd007K4uTeuWnjBSRqDxiN+KDuX6P+KctLS0lLS0NvV7PggULeOihh8jLy6OsrIz6+npWr17dva1er2fFihWsX7++X7ESDAYJBo/ddbndI3sHrDnDlz9OhiiKpDiMNPkiHG3xk+k0sSI/jn9sqKLjDDQ3i8oKda7++8kMlNWTk5mbE4dFr8XtDwMyVqOOBncQRVGw6DWsnpJMboKFeleALz2xlbvPKcBiODtTyILBEB8facNq0PKVc/I5pzip2+tHFAXCUZm9tW4+8+cNI5b0rKIyUuQmxnyyAqEo7b4wHx5oZPWUlFO8qyexHC0/3mCES6anjcQw+2W8rBiM6Oy4YMECnnjiCYqKimhoaODHP/4xixcvZu/evdTX1wOQnNxzvS05OZmKiop+9/nwww/z4IMPjuSwVY4jzWHktzfOYu3BBn77/mFe29vIHcuz8fjCPLqx6ozoKzRc5CWaWVWcxJ0r8jlQH8ufKJmUxEs7athZ1c4FU1I4d9Kx8z0clfnZWwcIRGR+90Ep9S4/qXYD/pDM188rHNLd10REr9f1WHePRGVq27z88r3DzM1x8vSmSrVaR2VCYtCKmHQaFEUhEJFZVpjArmoX509OHlS0vtEdIDvedNbMCX0xomLloosu6v7/adOmsWjRIvLz83n88cdZuHAh0Nvl8lSZyffffz/33ntv92O3201mZma/26sMDyuKk1mYn8B7u6t4fkcjC7NM3L4gjT+trxnroY0pBq1IVFa4dXEOiVY9r+2q42+fHOsn8+0LS/jyyvxe74vKCi9srealHTXdrQKe3FgJwKK8ePae0Cn1TEdRFJ7bUsVHh5qxGjTYjBrWHmripR21Yz00FZUhc25JMha9hmZPEEkU+MPNs7HqNQTC8oBzQKJRGYdJd1YLFRjl0mWz2cy0adMoLS3lyiuvBKC+vp7U1GNliY2Njb2iLcej1+vR69WeHWOBXiNx7tRMfrOmgn/UuVlWoLqwBsIyl01P5bVddd1LSDpJZHFBPDnxZm6Zf0xIRzo7+GokkXp3gNd317GqOIlmbwiLTqLdH6LeFSQ/0cSWshbyEs3YDNox+VyjjSAIFCZbOdrkZd2RZvbUqJEUlYnN1bPT+enV04GYBf+KosTu3/NAhUooHEWSRPSnmZh7JjCqYiUYDLJ//36WLVtGbm4uKSkpvPvuu8yaNQuAUCjE2rVr+dnPfjaaw1IZBAatxF0r8vnmv3fy6QnmRGcbeo3I9HQbHxxo7FEWG4rKrDnYxPzcKJVtfiYZtIiiQFRRCEZkSqvbmZPt7NEdusMf4oODTczOcrD+SAuPr6/gC0vzxuJjjRmzs+KYnRVHKBzlx2/s44kNlWM9pFOik0QkUehlnug065ieYWdJfgLlLV7e3tvQo8JpoPu+dHoqhclW9tW5aXQH+LSsdTiHrzJCnFuSxENXTetuF+ELRXltVy0L8+K5dk7GgPbR7gth6XTVVhlhsXLfffdx2WWXkZWVRWNjIz/+8Y9xu918/vOfRxAE7rnnHh566CEKCwspLCzkoYcewmQycdNNN43ksFROkylpNkwakar24TdNmwhYDRoi0ZhD8eaK9n6321TWygvbapiU6sbtD3P17Ax+8MpetJLIpBQbpuOy+a1GHVfMjFnLf8Zp5vq5mWdlBRqATivx3YsnoxFF/rGufKyH0yeCAD+8fApXzcrgqU/LOVDvYc2hJoxaiV99ZibzcmL+SmsONvKb9wfXbduolbhtaQ63Lc0l/rjOz0+sL1PFygQgyarnx1dN7ba0UBSFFk+Qv3x2Lqn2k/d7c/nDuP1hMuKMOEzjzzx0LBlRsVJdXc2NN95Ic3MziYmJLFy4kI0bN5KdnQ3At771Lfx+P3fffTdtbW0sWLCAd955Z9x4rKj0jSAIeMNnX1WGThIpTjJxoMFL+CRls0lWPSWpNhbkOtlV3c7LO2rQa0Se21LNrUuymZPt7CFU+uJsFSpdGLQS83LieGZzFb7TMHM7tziB/1pdwjef24GiwKHTNOXq4o83zeaiabHl61sW5fC7Dw7j8odp94XYW+viSJMHh1HLH9YcHpRQ+dyibL5yTh5J1lgjPFlWWHuoiQP1bt7cXXeKd6uMNTMy7Pzh5tk9Ghn6w7HWGplxRgqTe1/borLCvlo3CVYdGlEk03lmu1gPlREVK88888xJXxcEgQceeIAHHnhgJIehMsxUtZ6dDd9CUZnddX1f7Gx6kVS7gaWFiXxpRR7rj7Ty4Kv7SHcYuWxGGssKE1icnzDmxkoTiZr2wKCEilknkWwzcLT5WJfs7AQrU9PtPHHbPHZWNHHv8/vwhE5PaP/uxlksLUzofmzSafjv1cWUN3u5YEoK5xQnYTFocPnD7Khqp649QIs3dMr9Fidb+e/VxViNsbyGSFTmbx8f4V+bqqhq9Z/WmFVGntWTk/nNDbN6/cZNOg0tHQHu+/dOfntD7NxRFIVAOEqzJ0SbL0RRsnVCNxkcDc5OYweV0+J4K2kVuG1xNhdNTWZKqh1TZ9+nfbXlrChK5MHLp/QK58qdURnxLF6LdvnCrDnUyORUG4GwTKJVT7JNjz8cxaiVEASBrRVtfb5XEMCklQhFZcJRBa0kcM3sDD6/OIeSFCtbylu57s8bAfj31ioSrXpuXZzDipI07lsd4g9rymjyDK21gdWg4dLpqb0iX6Io8Mgtc3o85zTruP/iSdy7uohfvHWwR5VYXxxs6OBPa49w3wXFCIKARhL5zLxs/vZJ+ZDGqjJ6nDcpmT/dMqff3/Sa0mZavSFu+funFCaZKUq28rVVBaTa9KQ57GpeygBQxYrKoHlzjxqO7uKzC7P4n8un9nr+uxdP6ncp52wWKQAfHWrim//eSVNHz5wnnSQSisoYtRJOs46a9t7RBKNW4rWvLyXLaaLeFWDtoUbm5TgpTom5X9a0+ZiaZueWBVn889NKOgIRfvbWASQRvrQ8n88vKeDc4mQ2lbfx2w+PUDGIiIUkCrx1z/JBL9HpNRL/79LJXDI9lcfXl5+0HLu00dNj/3FmHb+/aTa/ea+UTeWtqmvvGCMKUJhk4fYl2Tz5aRVmvYbPL8rhwqkpfZ4X5c1evvfSbnZ1dhgHyI43k+YwEmfWYTPpz/ol34GiihWVQZNsO3mS2NlEit1IIBztFcJVJ6D+mZ/rxN/H8k6os7TbH472KVQAApEoDqMWrRRb27/luIaT7b4QF/76Yz797ip+cOkkilKs/Pytg3iCEX71bikmnYZVJYnEWQ1cMy+bK+dk8fKOGv7fS3sGtNyUGWfE0keukaIo1LoCuHyxaE1BkqW7CuR4ZmXFMSsrjjpX/1U97+xr4NF1ZXxhSW73cwvz4sm63sQdT2xBpxHZXtl+yrGqDD86SeSbqwu5fWkeGknk+vk5J90+GInyj3VlXDAlhV9/ZhZv7qmjpt3P/RdNGp0Bn2GoYkVl0KwoSuS1XWdndKWrLXtWnJ6paTaynHrcvjAGu7rePFD0GhGLXjOoxFOLXsOc7DjuXJFHvKVvnyWrXsPXV2ah10hIksjnFuUwP9fJ7z84zJbyNv7fS3sA+J9LJ3Pt3AxsBi1Xz86gzhXgF28fPOUYylt8PPjKXq6anU6q3YhJJ3Gg3s0v3z3UwxemIMnCBVOSWV6YyKysOLSSQCgqo5NE1h9pOWXO14Ov7kNR4LalxwSLzajlF9fOwG7S8oVHN3GoYXgShVUGzjVzMrhzRcGAt9dJIquKkwhEoggCfG5RzsgN7ixAFSsqg+ZAfcdYD2HMuG1JNvOynZSk2sh0mtQIyhAZTPfZvAQzL3x5MXHmk5dyPrO5il+8V87kjASWFCYCsS61v75+OnWuIH/7pIzHN1Tww9f28c6+epbkJ3Dninyq23qKh6JkC22+MN+/ZBJFyVYONnTwxIYKLpmWwrR0O2/uqeepTZUE+qmIO9zo4XCjhz98eASjVsKkk2jzhbAatLj8A8uV+dlbBzDpJC6bnorZoMWi1/DkxgrO8hXEYee+1UUszIvnS09upfUkSdBT023cvCBrUPsWBIHSRg+Privj3tXFA/ZXUekbVayoDJrPzMvksfXlZ836uVYSkESBUETm/MnJzM9NOPWbJhCyrCDLyqg17TxQ30H5AJO0Eyx6/vnFBd1C5bktVczMsOMw6Ug6YTky0aonFJUxn+D6q9FoyIzX8D+XTWFRfgI/eGUPf75lLnVuPy3eILctzmbjkRbKOsc0Nd3Oty4owaSXCIajFCkWatv9/PC1/YP+rP5wtNswbqBCBUAUBK6clY7+uL/J0oIEvv/Sblp9Q0sOVjnGnOw4/ufSyUzPsMf6c4WjXDY9hZw4Ax/sqeRAm0JUVrAbtfz06mmsmpSEXjP46GlFq5daV4BfvH2Aa2anqzc3p4Hq4asyaIqSreQlmMd6GKOCIMDKokRyEyz84ebZzMs581oMKIBw3C17MBzFEwj3mVcyHPzpw9IBb2vQiug0Ig3uAKFIlNWTkslOMPcSKgBxJh33X1RCYZKlx/NKZwtnSRRYPTmJkhQbb++rpyTFRqrdSGGKnd/eMJOsuNg+99W62VfbTjgUJsGix6TXcPfKgYf/h4ObFmRh6KyK6mJRvrPb20VlaFw6LZkfXD6Z39wwkxmZDjzBCMk2A+/cs5Tf3DCLb140hae+sgpbZwd0nQRmvWZIQkWWZco6y+hnZzmG82OclahiRWVI/OuLCyhMMg8qnD8RSbUZKEm2oJdgVXHSGXlnJIlCd+mkoig0e4PoNCIGrUgkKndf7IeL71wyGYt+YJP/NbMziDPpSLLq0UoiDrOu3wvHvFwnd67Ix3xcEmyLJ8Cru+rYUdXOqzuqeXNPPVPTbVw/t2fzU7tRx3+dV0BOvIlUu4EnNlayoayNYEQmFI4QCocxjqIPxow+mlg6zXrSHMbeG6sMmHZ/FKdJh73Ty0YvKnxhSS4ZTguiGJvL3tpdR6QzatzkCfOFxzazv25wvaoqWry8uaeedYdbmJ5u49ud5egqQ+fMvtKojBhJNgM/u2Y68hluZNvqC2HSSzx81VT0Z4FpkyAIpDtM6DSxu/qootDsObWh2WCIN8eEh92oJdnWd7KsWSfx4OVTuHN5HpIoIAhCv5N9tHMZC6Ci2cvm8lbkzhNTQaCq1ceX/7mVrz2zk5lZcfz3BSW99pHhNDEvL5FAZyXSvecWMDPDRn27jwSrgVXFCczNtg/TN3Bqfv9BKb969xC+UM8kZIfp7GhsOVJ8criZbzyzgy0Vbfz6vUOc++tP+Mnr+3h/f333NlfOTGXNfy3mf1bHclSissKXntyCNxjpXjLtj60VrQQjUTYcbeErT20HwG7SkZOourKfLoIy3LdNo4zb7cZut+NyubDZbGM9nLMKRVH47n928/SmqrEeyohSkGThja8v67Mc9WwgHJWHtT29LCsIAgRCETQaiUt++zFuf4TLZ6aR6TShEQWKU6zMzorrPv7Goy1Ut/kpb/FS2uChwR1gUqqN2nY/hxs9yIpCvFnP0eZYlcxtS3P5zoUlRGUFjSSiKAqv7Kzl3EnJfZYfd3Gwrp239jZy5/I8ItEoLl8Is16DJEmUNbopa41Fat4/0MjXzikg2W7gLx8dpaLFh8Ok5Zuri6l3+fnoUDO7a1z9HmegZMQZuXR6Gha9xJ4aN5WtPvYN8i5fpX9WFidy25Jc8pMspB8XtZJlmZ++sY+/raugS5u8ePfi7nMyto2CAt1RSX8oylef2saB+o4epff/d90MrlGTa/tkMNdvNcFWZcgIgsA3zi3i49JmqtvOTDtwUQCnSUdFi7fPvh5nA8MpVOCYKZ5RH4sSPHj5VGrbfVwzJ7Y0oygKgiBQ7wrw1p46nt5UxcGG3hVoe2t7XrSPjwDpJRFBEJDE2P5kBc4tSTypUAEoTnWQk2glFI6ikTSkOXXdEZ04i5G0eCuXTk+lzRMk3mZEEAQunppKVZuPDIexu+fT30/hVjtQqtv8/GntkWHZl8oxBAG+dUExn12U0+85MTnVxvFBlMAJOVyNHX60Gom69gCKAs9uqeT9A409tokzaUl1qL5Uw4EqVlROixS7gWfvXMSSn34w1kMZERxGLXctyyEnXm0uNlL89eOjfHK4mXf3NRKIRNFrRJKtOt7Z10S9OzCkfb62q5avnJOPXhub4iQBLIaBdbHVaySaPSH21bSyq7qd2dlxLC1IJD3O1C20EuzHzoeqNh8/fn0/l05P7fbSMGqlfkubVcaem+Zn8uV+kqYVRcHnC/LurnLsRg1TEvXMzolnTnZc9+trDjbxjWe24w5EsOlFPCEZWYFp6Tb21rpZWpjI3GwH6Q4Ti/PPrOrBsUJdBlI5bRRF4cMDjeyrdfHi9toejeQmKqIAKwrj+fUNs7EaNN3JdyrDy0eHmvjcPzYN6z5LUqz84wtzSe2MfIw0iqIQkZXuCNTeWhdX/3E9wYgqVsYrWkng5gXZPHD5lF6vybJMR0cHnxxuItGiZWpWCoIk0uEPY9QK/OytQ2wvb2RSshmTQU9anIWq9gDv7m/i9zfOZE5OPIqiEAyEMJvVqMrJGMz1WxUrKsNGJCrj9oV4c28DP3hlb3dG/UTDqtfw0leXkJdgVjP4B4ksKwPqfbSprJXH1pfx1p56hvs0kUSBZYUJ/OiKqWQ6Rz8i9qUntvDOvoZRP67K4HnzG8uYlNr3deOjQ00synOi1UiEQiHe2F7Bn9bX4rTo+NaqLO56dj/XzslkeoadrHgTobDM9OOquBRF6c6ZUukbVayojBmhiIxOI3LnE1t4e4JN2JIosKQggR9cOon8pNPPT4lGowiCeNY0Lmx0B7AbNd1LL/3x/v4G7nxy66iI2TnZceg1Ynei6vxcZ68+TgBHmzw8uq6cbZVtTEmz8aMrpw7NBKzFy8r/XcPEnlXPDgTgwqkp/OozM/s8J47naH0L7kCUpzZV8u/tDTz3xbkgSsw7hUGkLMtqVPYkqAm2KmNGV8XMdy6e1C1WpqfbKGv20TGIXjBjwVUzU/nFdTOHLZoiiuJZFZnpy6jtRIKRKL9679CoRd22VrR1//9zW6p58PIpfH5xTq/tPvv3Td0VHHtr3XxwoJFZWXFcMCWF+TlOUuyGXtVgkaiMPxzFotd0/51f2FqtCpUJQpxJy/5aF09vquzROPJEZFkmLyUerz9Es+cI09PtzMtPHtAxVKEyfKhiRWVEyE0wc+viHB5bX87Xzi1iUX4833xuB2/vHb/Rlq+eWzis4uJsEioDZV+tu0fTv9HmJ6/v5739DeTEm7lwagqL8+MRBIE0h6FHuWmzJ8S7+xp4t1NwS6LA3Ow4FubFMy/Hyeu7a3llRy3BiMy1czKw6iUaO4K8epY2+JyItPrCtPrCVLV5iUQiaDR9Xw67BIdBJ3H3OYU4zX17A/VFV2WbyumjihWVEeMr5+TzzOZKPj3SxPmTk/nCktwRFSuSKPToV6TXiCwrTGBKkp7CNCet3jApDgPbK1385aOjZDtN1LT7uxMhQxH1lng4OFnoOy/Rgigw7HkqAyUUlfm4tJmPS5t5cmMF55YkYTNq2VzedtL3RWWFT8ta+bSstddrz22pGtbPk2TRsWpSMnWuAGsPNQ3fjlV6IIkCc7LjWJRlp6qpndzUU1ftZCdYMOrUy+ZYoH7rKsNKJCp3J5QlWg388vqZPPzmfpzmUq6bl8nywng+Km0Z9uN+/9JJzM6KQyOKVLX5aHQHeHJDOQYJVhQlMjv/WE+V1ZNTuXp2OrlOExvL2rjj8U34Iwpv7q6n6Cz1UhlOuoSKoigEI1EMnTksiqLwi7cPjJlQ6YsTfTGGQkmKjWAkypEmL6l2A3WuoZVbQ8z/4weXT+GS6WkAfOeFXTyz+cw2XRwrorLCprJWbDqR394w86TbKoqCosTmtMGgRlWGD3VBbQBM8BzkUUVWen5fqycn88/bF/DouqO8v7eWu1bkk2YfnnI+rSRw5cxU/nHrXG5fmsesrDimZdi5eFoqty7J5c2vL+F3N89lWk7v9eXCJCsajcTSwgS+sCAmZB5fX4bLN7zW8mcL9a4AsqxQflzZ+rbKNhQlNlm7/GG++PgW/rmxcqyGOGLsq3Nzaae4OF2xqyjw2nFLSfdfNKm7qZ7K8COJAudOTsZkPPXSjic4Mo09VQbGiIqVhx9+mHnz5mG1WklKSuLKK6/k4MGDPba59dZbu/t+dP1buHDhSA5r0AiC0N1rJBIZ30miY41O0zOp1B+KcKC+A4Ney68/KGNKmoNnvrSQ8yYlnfaxvnfxJP7v+lmsKuk72U2ni7mPnsqB1WGPZaG3+sJsKhv+qM/ZgM2owRMMoxDrj6IoCgVJVspbvHz5n1tZ9rMPhiWKMV55cmMF/3PpJO5akcc5xYmnta8399SzpTy23GQ3afnZNdNZVqgai40EXz8nj6tnZ550G0VR6PCFON1uG+Fw+PR2cJYzomJl7dq1fOUrX2Hjxo28++67RCIRVq9ejdfb0zTswgsvpK6urvvfG2+8MZLDGhJdoe3+krBU+sZq1DEpxUZ+ogVPMMxXntpKglXPg5dPYUFu3Kl30N9+9Roun5nW3ZfjdFiSn4BVH/v7dvU5CoYj+IIRIlHV2GsgmHQabEYduQlm5mQ7EQQBu1HLz986wJt76nEHzmyR3+oN0e4LsSg/gZ9dM51E68CTMPvi9se38MH+WH7XRdNSeeK2+Xz/kkk4zTEX3pwEE19ekcv8XOdpj/1sRBRgcV4cty/LP2WDUm8ghMmgxWIcmAPyiaiR+eFhVH1WmpqaSEpKYu3atSxfvhyIRVba29t56aWXhrTPsfRZUTO9B0dpQwf3Pb+LBTkOvnvJFBRF4YJff8ShBs+Q9vfKV5b0MGE6Hf6y5hAPvVXKtHQ7r35tKfXtXqwGLTVtPopSh+cYZxOhiMyX/7n1jI6mnEi6w8j731yBQSux9lATtz666bTKmPMSzDx9xwKCEYWsznYP4UiUylYfCRY9VW0+Klv93P2vbcP0Cc4enCYNL901n8xER485vOtyKAidyfqKTFmzl4JkW49thjLvq9eL3gzm+j2qOSsuV6wLqdPZ825gzZo1JCUlUVRUxB133EFjY/8TXDAYxO129/g3VnSdeLHkK1U9n4rCZCt3Lc/lxW3VNLl9CILAF5bkkDIAf46+8IWHbw35C0tyuWpaAivzbQQCAVIcZkRRIDfBQmAYj3O28Pru2rNKqADUtPv544eHAVicF88Nc9JOa3+VrT5e3l7FT9/aD8TmGY0kkptg5q4fvsvfPi7jvEnJPHn7fB68fAo3zMvkkmmppDuM2Awa8hPNzM5ysKwwgaUFCVw3J4NVJae//HomcE5xAomOWH5RozuAoiiEw+HuVASI5bNIktRDqAwVVaicPqO2pqEoCvfeey9Lly5l6tSp3c9fdNFFXHfddWRnZ1NWVsb3v/99Vq1axdatW9Hre4dSH374YR588MHRGvaA6OskVE/Ovsl0mtCIIs9vreHOFfncOD+beTlObvnbpkE1rcuON7EwL35YxhSNRolGZR68bBI6nR6DQU84KvPYugq+sDQHRYFAOHpKl0uVGFFZ4e0949dPZyT568dl3LIomySrgVuX5vP0ltoh7ysiK/z54wqunp1ONCojKwqiAE9vqmADkFXZjqwoLCtMZFnhsTyZrl5FGlHoNQdVtfqobfdzoL53F+uzAYNG5IZ5Gdx/8ST0Wg3f/c9uatt8PHbbAjQaDW2eAGadBKLUywSwi9Ge19VrSYxRWwb6yle+wuuvv84nn3xCRkZGv9vV1dWRnZ3NM888w9VXX93r9WAwSDAY7H7sdrvJzMwct3b7fr8fo9GonnDH4QlG+Mnr+zBqJP6ns5FYvSvALX//lMONA1sSWlWSxD9unTfsYwsGg2g0GiRJYsvRZmbnOIlEFYIRGVEAs0E77Mc80/juf3bz1KdnXtXPQPnswmx+eMUUBEHgYH0Hdz65hfIW35D3p9OInD8pifJmLwatyPYqV3f599KCBG6cn8XcnDiSBxih/OBAA3/56CibylpHrYxcEBhzZ1+7QeJr5xZx+9JcwpEo26tcfOYvG3noqmnctCArJvKiMtohtFk4Facz/5/Jlv3jzm7/a1/7Gq+88gofffTRSYUKQGpqKtnZ2ZSWlvb5ul6v7zPiMl4xGo1AbzV+Jp+Ap8Ki1/DQVdN48NV9RGUFSRRIsRv4x+fncf2fNwwowmLVn/6EoigKu6rb2VzeRmWrjziDRLpdy/lT0oizmpibF6vA0ImxBn0y4AuEMamCpV8ON3bw9KazV6gAHKx3E43KaDQSxSlW7r94Enc+uXXI+wtFZF7fXd/na58cbuaTw81Y9Bpe+eoS8hItp9zfqpJkFucncN+/d7K/zk27L0yLd2RL9kuSzZQ2+ka1uWmiWUubP4LTrCPFbuB/r5vRXVoeDPjZerQBnSSysrN6SxCEEREqXfseKmfrdeJERvRbUBSFr371q7z44ot88MEH5Ob233+hi5aWFqqqqkhNTT3lthOZrhMwEpXZU+Ma49GMPoIg8MDlU3pU82TFm3jo6qknedcxTtVkritgGAj1XS4oywr/985Bnvq0kgsmJbKsIIENZW38/L0yrvvzRv7fi9vZcOjYBUISFYw6CV8ogjcYRpbVPKUTiURlHn7jwJjfQY81m8rbuOqR9TR2iu6VxYloRriZpScY4XP/2DTg6jWDVuILS3LJT7TQ2uktVJhkwaAdvkuCUSuiFSHBomdWho0XvrxgVDxjBAHiTRpuX5jEH64pZN13VvHKV5f28MCxWq3cuCifBy6fQpJlaFU+KqPLiJ45X/nKV3jqqad4+eWXsVqt1NfHJn+73Y7RaMTj8fDAAw9wzTXXkJqaSnl5Od/97ndJSEjgqquuGsmhjRsUYFKqjXBU5kBdB1lOE3bT2Xnn7g2EWZwfT1Gy5ZQVQmsPNdHmDRFn7nui6bqTqXOH2FvbxOayVuwmHedPSmRaRhxhWearqwrRigLv7q1lVUky505KxhOK8MH+BraUt/GFJ7Zz/bxMvrg0j0ynEUEQiLcaaPYEkVAIR6JYTAZ1ea+Tgw0dZ11SbX/srnHz2q46bluai0YUsRm1tI5g9EIQ4KvnFHS7Rw+EWZkOUuyGbnF52Yw0nvq0kvrw0B14AXSSyLcvKmFauo1QSKYozYpVJ2HUa5mbE8cHB4anhYBdC3euLCA93sJbe+opbejg4mmpnDspiRSbjmS7+aTvjzPruXH+yT1WVMYPIypWHnnkEQBWrlzZ4/lHH32UW2+9FUmS2L17N0888QTt7e2kpqZyzjnn8Oyzz2K1nh22512GZRIC0zLslLd4KG3sYG7O2eefYDZocQfCeAfgFNnQEeTlHTXc2ke31K4lttKGDs7/1Ufdz9sMGs4pioV8j4/MXDg947httFw2PY0FmRZS7Hre3dfI8l98yMXTUrh+bibzc5xEZAWDXodh4qxGjgq5CWbOm5TEe/tVwQKwubyF25bmIokCd63I46E3DozIceJMWq6cmc4N87MG9T5RFHjw8ilMz3DgMGpZVZLE4UYPr+wcelIwgMOk5dbFOX16IN1zXhE7qlynLdzmZFr59rlZzC3KQhAEfKEoqycncdUpDN5ORBAEvP4g5gE42A4nag7j4BlRsXKqMLnRaOTtt98eySFMOHLiLbh87dz73A5+fs30Qd0pnQmIgjDg5ZUfvb6fWVlxzOjHa2VfXc+y9juW5fW7bReyHJtEUhMcfH6xGUEQuWxGGh2BCM2eEFFFYePRVkAhEJYRiCX7Jg2x/PpMwqTTcO3sNFWsdHK8Pfsdy/KYmRnHFx/fPOwGeXcsz+OLS/OG9F5BELh2zjGxvr9u6FYQZp3EeZ0NS/sza5ye4eAfn5/Hw2/u49OykzeP7I+8RDP/95k55CSYURSFJk+QsmYvNW1+zpucgnWQOWV67dgYfZ7NeYtDQbVjHYckWPVMTrURVZQz8g+kKAqhiNync6RFr+FfdyzkN+8d4qUdJ7/Di8oKP33zAE/cPr+Hpb7Y2czwkTVHup/TiAIJFh3iSXIHTixPthq0fOWcgl7bzc2O4+LffExHMHbRSXcY+eyibG5ekNVrojzr7qAEdfLt4mB9R3cCuSAIzM918vVzC/nx6/uH7RhZThNZTlO/ZbaDpTjFSukAK/JO5KfXTOeyGaf2lpmZ5eDCqalsKm8bUn7T6snJ5CTElngEQeCuJ7eyrbKd6+dmDFqoRKIybn8Yp3V0bQkEQRi1SqwzZQ4aVQfbkWAsHWxVBkZfP5aB/ICe3lTJY+vKOdzkiblJ9oNGFLhtaS5Os45d1e1sONJCltPEzupjictL8+P42TXTSXeeuloCYhGWX7xzkDZviFsWZjM13U67L8Rj68rISTBzsMHDvzZWcMP8LFJsBrSSQHGKlUkpNr7z4m4mp1n53IJMRAR0WpGIDEb9mZ+LFAhHKfn+W2M9jHHD1HQbf7hpNtnxsYtrRyDMOf+7lmZP8BTvPDVxJi0b7j93WP1/Hn5zP39ee3RQ7zFqJb53ySRuWZg94PfIssIXHtvM2kODz1/Jcpp4/Lb55HYKlrf21LG/roMvLc/DrB/47Z2iKFz9yHpmZDj47MIs8pNGN/Xg+A71ZyuDuX6rYkVl3FPT7uebz+3oXH4ZGtdOi+cn181BrzsmGI631u6LQDiKOxDGZtD2eUGoaPF2X4T6IhyVEQUBUQBvMIxGlDDoJp6xXDgqn7IZZBdt3hCzfvTuCI9oYjE/x8lzdy3qfnywvoOr/7gOb2jozsgPXz2Na2aloxtmo8KPDjXxuX9sGvD2+YlmHrpqGguGYND4m/dK+dV7h4BYPtlglscumZbKb26YeVoX+0Z3gPkPvQ/Af+5ezKysofcqGwrqMtA4tts/G+jqzqwyfKQ7jDzzpUW8d+9ybluSS05nn5TjOVVpaLLDhE6rIRKVCUdiF4njrbX7wqCVSLIa+r1zPZlQgVjydNcSgFmvJRgO4w9NvIZ+je4A7s7y1mDk5BfYf2+tGo0hTSg2lbdS5/J3P443a7lrRf6g9qHTiFw1K51Hbp7N+99cwXVzMoZdqABMThv4DZ9Vr+Evn5s7JKECsLumHQCnWccPr5hK4iBKiF/fXcc3/72TUKTnfBuJRIhGowPKe2v2HKt6OtLkPcmWI8PEDhOMPmdiSsSYcrYr5ZGkIMnK/1w2mW9fVMz7+xtZc7CRo01erpiVzgWTk3l9dx0Pv3mg1wTmNGkpTI3jqU2VWPUiZc0BBCGWezI3y4ZOd2ySlGUZRVEQRXHY1nm7lrxc/hCaYBSjc2L97NLjjonDjkAEvaX/i+Te2rHr1TWe2Xi0hYW58aQ6jCRYDRQmD2w5souvrCzgG+cVjtDojpFg0TM3O44tFf0nv8ZytLK4c3n+af1GVk9OJsGi5/6LJmE3adla0cqTGwduKPjyjlqqWn08+6WF3WZukhT770DG9cia2HLX5xdlc/kAcm2GG+ksXwIaLBNr1lRRIVZ2fPG0VC6e1tM48AtLcpmabufWf2zqEWJv9YW557ldvfZj1kncuiSHLy/JwGKJXTxGQmx2TZxZCTaqW4duuz6WtHlDdATCJ6162lLeyhu760ZxVBOHh984wK2Ls7ljeT4aURhQIqjdqOX8ycmsKknigikpozDKGPecV8Qtf/+039fvWJbbp2XAYFAUhevnZXH9vGPl1oVJgxNwANsq2/nDh0e45/wi4NQi5fhcuQSrHp0kMifb0W/10kjiD0Uw6tRL8EBRpZ3KGcW8HCdP3bGQ7D6Wik7EG4qy9lATpaWHR2FkMaITNPYbZ9aRFW8+aTKnUSeNyaQ/EWjsCPLztw+x+lcf0e4LMyvTgb6fCp4Ei57vXFTCtu+fz/9eN4OLp6WO6ve6tDCh24K+L1z+01vK7G+JZnH+0JaT3tzTdyuCvjhezNxzbhHLixJYX9rES9urqWv34guMbNuBLqKyckoX7pFgIqcpqGJF5YxjRqaDNfetZM+DF/DTq6dx3qQkrHoNWXG9jZ9WT05h1qyZwKl9gYYDQ+cE5T+N5MrxypQ0O7My49Cp4e1+KWv2cunvPqHdF2J+7jHjx4V5Tj4zN5PHb5vPlv93HnetyB9T4Xfp9P6XRerd/n5fGwi17X6EPvyUchIsLC+MCZbF+fHMH6AxZkWrl+gA2wwcjyxHmJxm45IZ6fiCIR54ZS8bTiOJfzC4/eGT2iiMFBM5TUGNQamckQiCgEWv4Yb5Wd3Onq/urOVrT2/vsV1cZ2uDoXgRdATCWA1aQhEZjSgMaPKxaGKTqkY6MyMQ1e0+RBE487TYsFHT7uczf17PrYtzyHKaWFaYyAVTkk95/kVlhRe2VnPBlJQRb8mxMM/Zb6dkzWle8MqaPKTHmfq0M1helMBHpS1cMzsDo1Ykyabnjd11J/UkURRwBcI4zYNzoY2zGPGHovjDES6YmkZRip3aVg+flDayKD9xRMVif21CVPpn4sosFZVBctmMNB6+elqP5/ISzciyPCihEghH+e37pXzt6e0oisKbe+rwhCKxic/vx+PxdN/pHe8P4w2GMZtjy1MDLQWeaPz06mkszXdinoAl2qNJIAqT0hz85KppXDg1pc/zT1Fi4qS23U84KoOicNXs9BEVKl3Rjow4E59flNPnNncsG5pbLkA0KmPQSn1GMQPBMFoh9nxxipWLp6fx+5tm88ebZp9UOCRYdMSZBn/x313dxn+dW8iUZCttvjB2g0SyVcc/Pinj07KWQe9vPHDiMk802vddw0R0LFEjKypnFTfOz6LdF+Znb8X6tMRb9AMKjYajMpvLWnl2SxV7alwcafLy98/P5Qev7OXdfQ388NV9/PCKqVwy/VjSbygi0+YNYtJJSEoEs8mE2x9GpxGH1chrPFGUaOZr5xZS1baHgw0dYz2ccUuLJ8g/N5azKD++s1IsjKIoPL2pijqXn3pXgG2VbTR7Qnz8rXOO6yE2cpxYCfftC0t4cVt1L/8T42kI0bAsMy3d3qc4M+q1WA0a5mTHMTXd3v38hdNS+VxZM4+u710plJdg5qGrp/V7s3H8Rfn4bXZXNpOfaEEjgtViYPVD7/Oza6aR5TSSE29CEibexRx6Jxh3VUdFIhE0mtjlfqI62qpiRWVMGYsfTpbTRJxJS5svzP++fZA/f3buKUO+v3u/lN9+cCwRVxIFNhxp4YkNFd3PPfzmfp7bUkVNu5/lhQlcOTONl7dX887+JianWvnWBcX8Z0cd911QPGKfbaxJtJtItJt47LZ5XP/nDVS1nl5+w5mKrMAbexr4/KOb0UkC7x9oRBSEXk7NDlPfhoQjgSiKKIrS/Zs06iSumpXO48ed48sKE07paXQyInLM8bY/Djf6uXZORq954evnFTE728kja46yr87NrAwbty/N47wpKf0mKiuKQiQSQavV9hAtnmCELZVu8pLt6HRagoEwT9+xgL9+XMY5RQlkOI2kWQ3IsjImeSWnQ38RuuOFStd3O9FEy5kZi1aZMIzFj2X1lGSevH0BNy3I4sODTdz33A4CnX1+OgJh/rjmMP/ZVo3LFwZiZbsbjzb32IfNoOFLK/KwGo7p/eo2P2sPNXG40cM/1pVz+R/W8/f1lVS1+Xl7XyOfe3QzwUhkQoZgB0uq3cgbX19GfuLJjfPOdj461MR7+xtRFPpsKTEt3U6idfQ6AguC0CPSeN7k5O7/N2olfnHtdN7aU0uDO9DX20+JRa/p9/xXFJk2XxBvIEzLCe0IHEYdl05PY35uHPFmHV9ZVcilM9MxaKV+5xBBEJAkiWg02r2NPxTltZ01fGFpHmZ9TMRYDVqONHvJdJrQ6zRcPSuTt/fX4TsNA8fIcQm/o/F7P9kxBEHoXh7qEiuDXfoeD6hiReWsQyuJTE2389BV07htSQ7/2VHLDX/dwP0v7mLRwx+QZNaws7KF6/+8gT01Lr705BY2lbf32EebL8y5/7uWjkFYhNe6AlQ0e3lnX30v47ozEatBy43zs069oUq/fLWPRpqjyfxcJ5NTbWglgXXfWUWCWcuzW6r5zgu7hnwRFuj74qrRaNha6eLHbxzo1ZG5y226ONnKB/etZFVJcq/3n8iJ5o7hqMx/ttdw3dzYOXn8BfuqWRnMz3EyPcPO79cc4bcflLHruN5ig+X4NgCDEQVdQmKw9HcMWZa7bf27vg9JknoImImC2htI5axGURRq2nzsr+vgB6/s5bIZaXznohLcvhDL/3ctNqNmRJYy0h1G/nTLHKZl2E+98QSmus3HlX9YR7NndPwrziSumZ3B/10/Y1SP2dfSQKs3RL0rwOQ0G6GIzIwH38EfjvLUFxewuCBh0McIR6IEIzKWPozxNhxu4r+e3UFhio0nb18w5M8B9FruaO3wYTMZ+uwn9PsPSnlxWw1Wo5adVe0AfPP8Ir527qldg8ORaLeD7khxuktSx0dTuiJn46E3kdobSEVlgAiCQIbTzIK8WLO5lcVJCIKA3aznvy8oHrGci1hzxu14AuER2f94Id1h5NEvzOP8yae+E1Y5hlknce/qolE/bkdH76Rop1nX3TNIpxEx62MX5he2VQ/pGFqNRFM/XacXFSTyh1tmc6TRw/ojzXiCQ1+K6RJdXf+tcYeQxN55QY0dAebnOnnh7sWcV5IEwAVTkgkN0LslMMjlov4qdI5/XlGUHo/7EireQXw3XUtiPWMTQr9jGY+oYkVFBbAZdWTEmVh0nIvmzQuyBuSEO1TKmn2s/tVH7KlxxSYnWemx1n0mIAgC09IdPHLzbL5/6WSs+qHl9KfZDVw7J4NvnFvIQ1dN4+6V+ej6Saw8E7hqdjrpDuOoH7e9vf2UyzsLcuPJcpq4Zk7GkI+Tc5ImoJNSbFwzO52b/voptz+2uUcTyKGyq7qdz/59E499fLg7mV6WZepaXLjdPoqSrcSZdKQ5jBi1EiuKkvj84pwB7dtqGlxOUVeFTl/PH98Jvr/tICZmzEP4LUmS1L38I4onb+Q63lCXgcaIiZaJfSZw/EQw0O0v//06dtcMfe16IGhFAZtRizsQRlFieQIXT0vlhnmZfYasJzIH6t189u+baOroeWctiQI2g4ZQRCYcVQhFZQQBVhYlcvOCbGZmOUiw9LwoPLelim+/sOuM615r0km8fc9yMp0jJ5T7o6WlBYPBgNncv5iQOyMTw1Ep89zmSgqTreTEm7EYNN0l2juq2rjyD+sByE808+it88g6RZfz/jjc6OGrT23jQH0HBo3AuZOSuee8IgwE0Wm0OO2WHss4Ll+YOreP4mTbSUuiR2v+PvFYpzp2JCr3OW90zX9duSvjgcFcv9XS5TGi62RTRcvoMdjvWRAEtKPgNBuWFVq8x3I61h9pYf2RFt7eW8+vPjOz10V6IlOSYuP1ry3ltx+U8vKOWpYXJnLt3AxmZ8ZhNWiQFYXKVh/rj7RwwZSUk1bCXDY9jV+/ewhfOEpGnJEkq4HN5a2DSnoej3z/0sljIlSA7oaeJ2M4y3mT7Uau/dMGorLCwjwnj31hPgatREGSFZtBgzsQ4UiTl999cJhfXNc7f+dkuRz+UJR/bqzgzT11HKiPLW8FIgrZcUbizVoEWcJhNfaaF+wmLRa9hUBYRq8R+rywj+acfeKxTnXs/m5wuvJ3xotQGSxqZGUc0J8SVhk9+hKNiqJw22Ob+fBg0xiNClJsBj68b+VpGXGNV6KyctqW5oFwbM3doJUIR2Ve3l7Dz98+SGNH3zkR450b5mXy8ElMzkYav9+PKEro9aNnB//23nq+9589NHuC3LUin2+uLkIridz+2GbeP9AI9J9svLG0noWFfXek/sXbB/jDh0e6HwvAl1fm8vVzi0/pXROMRGn3hki0GsaF18pgo8IThQmXYPvHP/6R3NxcDAYDc+bM4eOPPx7rIY0qqlAZe/qaBJo6gmyrbB/9wRxHvTtAWbN3TMcwUgxH7xWDVuq+8DR2BFk9NYXLZ/TfhG88c+n0VH5y1dgJFegq5x294ymKwrLCBJ6/axFWg4Y/rT3CFb9fR3Wbj4eumkLXKeINRroTY/2hMF5/LBI5Pz+JcORYkmgkKrO/zs2v3j3UQ6gAKMDMjLgBmezJskKy3dinUOnyKhlNukq3R4KJUsI85lfJZ599lnvuuYfvfe97bN++nWXLlnHRRRdRWdnbWvlMR5ZH/0eg0jcfHmzkuj+tw+Uf+2qdF7dXd+cJqPRPusOIzaAdF3+zwSAKcOviHH79mZlj2mkZRj+fQRAEKlt9JJi1/L+LJ2HQiuyrc3P3v7ZR1uzjkZtnk+4w8t7+BjYcbkRRFIw6LabOsmdRFJFluVvIPLLmCBf95mN+835pn8f7zQeHB9Sh+fhvYFt5a48bhpEUDmNBlwfLeGfMxcovf/lLbr/9dr74xS8yadIkfv3rX5OZmckjjzwy1kMbdcKyPOByOZX+Gaqx0vG8uK2aitahuXQON09uKOcf68rGehjjmi6TvUA4yuby1jEezcApSbHy9j3LeeDyKeMiwmoymUb9TrskxYYiCCwuSOCJ2+YDsKvaxY1/+5T39zfw6xtmcPOCLHYeZ9J2vFioaAtwwa8/4gcv7+GVnbUnPVZli6/Hsm5TR7CHQaOiKLy4rQpdZ7PFI00e1hxqQp4AF/OhEolMjByvMU2wDYVCbN26le985zs9nl+9ejXr16/v8z3BYJBg8Nh6tNvtHtExHs9IJ8PqR9hY6Gyh685nKEZKDe4AX3piS4+JcazRSSLZcaNTxjoR+6EoitJdxuwLhvGGJoZ3xNKCBP54y2xsfZijjRWyLHf3kRlNzDqJ9/c3cMXMWMl2TbsfRYHnttbQ5g1z2WQHZiFCbbOL9EQHAOFwBK1Ww9t76jjc6OFwo+eUx5mX66TW5ScQjrLmYCN//6SMcFShONlKVFFo84a4cX7WMXv+YIRtle1cOyezex8nmqmNRpFENBrt4cY7nEyUPkFjKlaam5uJRqMkJ/c0jEpOTqa+vr7P9zz88MM8+OCDozG8XnRZFA9nmHQinCQTkU1lLWQ5TSTbDIP6fjeVtY4roRJn0nLH0mw2Hm0hEJGZm+sk1T5ywmWiCRWAOleAtE5Pkjp3kMAEECvzc5z89XNzx13i9FjkY0BsKeKKmenUtvv4482zuemvG7tF55pDTZQ2dfDoLdPxR4+NTdN5c7coPwHe7XvZ50RuX5rLkoIEZFkhGJH5912L8QXCiKJIMCqjlQRMumOXxQSLjtuX5pJo1eMPhjnY6EESRCalxsSNXtN/b6Lh5GSeK8Ox71AohE43eknVQ2Hs4470Tm482QX8/vvvx+Vydf+rqqoajSF2M9zruapQGRkEBI42efEEI3QM0CW23Rfij2uOnHrDUeSc4kSavWH+vr6Crz2zg1d31HDH45v589ojNHaMj2WqsaZLqCiKwiNrjtBxGq6no0FBkoW/3Tr+hArEvsORvDCeDEEQ8IdlJqfZeOjqad3Ph2UFs15LRVuInER7j+0BCpMtzM+NO+X+Eyx6HKZYFKuy1cd7+xsJhKO8vqeOv3x0hPWHm3oIFYDaNj8pNj1GnYQoCJj1GvbXu7joNx8z90fv8Z0XduELRmjqCPDRoaZxm/tx/Lj6GqNOpxv3y0FjKlYSEhKQJKlXFKWxsbFXtKULvV6PzWbr8U9F5UTm5TqZm+PEpNNgHWCYPRxV2F83esuKp6I42cqeWjdHm7wsynPy2YVZPLelmnf3N/LwmwdY+tMP+a9nd/QyWDubOH7iFQRh3FcCGbQiP792+rha+jmesb671kkSO6vauXR6GtPSjwmTvbVuvvz0LnyhCL9+7xDh43L7dKLAr66fedL9mnQS/7l7MVPSYvt8Z189r+6s5eLffMy/t9QwP8+JJEBju4939zVwpNFDMBLFYtAQ6UzebfOFiESiJJgNGHUSHcEIz2yuorLVh0knYdOLVLf6ePjN/byyo6aXG3VfUauorBA8rprJ5QuNiOA5/qa4vxvkrg7VsiyPS9E1pstAOp2OOXPm8O6773LVVVd1P//uu+9yxRVXjOHIBo+6nDP+GKwdu0YUWFGYwJ5aFy3esa8oOdgQM7I61OBBFGL2/PXuY9GUUGcX2YIkC3oJshOsLMh1YjNqz5rz8cTPuKokiZXFiawZQ2+c/shLNPObG2YyLd0x1kMZt2TFm3CYtTR1BPjcomz++/ld3a/5w1G++vR2ipMsvLitmstnpFLe4mdSqo0UjYTDqKHd33d0wKCViDPHhFggHOWdvQ3d/2/SSyzIjUcQYr1yzptkxOUPo5NEilKPCaYkmwGDEKE4JYFku543dtdxsLaNfTUuilOsiJLEV5/Z0d0I8WdvHeTiaSnMyXayrDChT3v8jkCYNm+IjDgjWo2EzailtNFDmsOIThTQDaDMerjou3/Q+GHMHWzvvfdePvvZzzJ37lwWLVrEX/7yFyorK7nrrrvGemiD4my5OJzJxJl13Lwwm7v/tW2sh9ILWaGHUDmeJzeUU++ORVc0okCSVU+TJ8jqySl895JJY9JjZqzQSCJ/uGk2X3x8CxuOtoz1cHrw3YsmqUJlANgMWlzeENfMTmdPjYvHN1R0v/bRoWZaPEGumJlKi9tPTVtMrEiSyFN3LMQTjPL4+nLS44xUt/nYXeOiqtXP8sJ4LJ1iISIrVLX5yE808/VzC/nGMzt4b38Dq0qSu5fAjDqJ/35+F9fPzaCmzc+6Iy1EZYU4k5bvXzoZp1nHp0dbKGv2saFsL5JG5GdvHqDWdew3WtPu568fl/HXj8uwG7VcPzeDJQXxLCtI7O7L4zDpcJiORbMEQaAwyUJtu5+mjiAlqVZ0kggIo5ZPNl6vY+PCwfaPf/wjP//5z6mrq2Pq1Kn86le/Yvny5QN675ngYNsXZ6pj4Xjm+a3V3PfvnWM9jGHFrJP4x63zWJAX3+frZ6rIdvnD3PTXjeytHR/LeoVJFt6+Z/mwXXCCwRCSJI5J5c5o0XVufvv5XTy75Vhu4nmTkvjb5+cNaB+bylqpaPEyJc3K5DRH9/NlzV5u/utG3vjGMm78y0bSHEZ+fOUUUh3H2hzsr3OTbNUjigIWnYRGI1HT7ifeqMGg1/LqzlqynEbWHGziD2uO9CiBPhklKVbuXJ7HlbPS+/3tBcJRDjd2cLjRw6XT05AmWNPBgTKY6/e4ECunw1iJlWg02q3CVWFxZnDPM9t5acfJfRomIjqNyDfOLeTS6amY9ZozqtfQyWj2BDnnf9eMi15Bf/vcHFaVJA+bWPH4ApiN+rNizpFlhcc3lPPztw7iD0dZVpjAk7cvGNB7u+ZmWVaQJLFbAO2saue/nt3BB/et5OlNlWgEyHSaWZjft6iP7UPuNlA70TzPG4ywt9bN81ureHN3/YCSvG9dnM3/u3gykiR0lw/LskJYVjjS6OF7L+3hx1dOZWq6HW8gjEmv6VVmHIlEJrRgVRsZjgLHZ8wP94Rxpt7tjndWFCeekWIlFJH5xdsH+cXbB7EbtWz+3nmDzueZiCRY9FwyLZXXd9XSERy7cuYsp4nJqdbTtrE/fl4YbqESjUYRhL6b9o0liqLgC0f5wpLcmJPtvjryEy1EIlEUQeju0gz0KL8NRWR0GrFbVEhSz8axogCfmZdJOCqzenIyh+rcpMUZ8AcjGI/LLekyyBNFsfu7ieW2yFS0eshNsPDWrlounJ7G/Fwn83Od3Le6mMt+/wkN7pMnvr+7r5F0h4nzJychywov7ailzRfGE4zQ4A5w3+pipnYmGbuDEbZWtrO8KLFXsuxwXy/G6/VnfJ2ZKoAaoRkrZmU6WFWSNNbDGFFc/jDbK9vGehijxgVTUsbcJE4SBRo6Qqf9ux5IRcfpMN6ECsQ+Z1euyeopKfzoyulEFZj24Lss/ukH3PvsDvydf9/6jgi7qtt7mASe+Jm6Hk/LcHDb0lzCUZl4i55FhYlkOc0YTignP16kdAmXqKyg0UjkxJsB2FXb3qMdRpLNwN0r80/52Wra/fzkjf1c9vt1RBS4a2U+501O5qKpKTx+23yWFiZ0b5tiM7D+SAtbylt6tJOQpJHzeRlviy5qZEVFpZOcBAt/+9xcPvuPT1l3eHwlZw4n26va+81hOdNYWZzIlDQ7u2vGzuivrNnLxqOtzMo6tRfIyTjxjnc474BjZasy0jiw/D8ZWknkxnlZtPsjvLS9hhe31/D67jriTDqiikJTR5A4k5avrSokyapjVXEigUgUq0GH9gSHcK0kEgxHenS9P/H7jcoKGkkkEI6g18QqZbaWtTApzYbVGIvi3L2ioNfyXkacEadJS6uvd1WhIMDxOuBHV0ylKNkKwIqixD4/tyAI3HteIZ5AmL01LkpSrJi0EoY+KoxOl65oDYyvKIsqVs5wJqJ9+lgiigLfu3gy1/95A55xbi42VAza0bkgRWVlzBvzCYLALQuz+PYLu8d0HFpJ6HFRHAonXjSG+yIy3oUKxH6fcRY93714Et+5sIRtlW04zTqq2vy8uK2arRVteIMR/vbxUQw6iXp3kPwkCzPSbei1MmVNHuwmPfEWHZGojKIIiIJAIBylwR1AKwnUuYKk2fWUNfuYnm4nEA1jM2iod/lIshoJyUq3UAGwGHt606w52Mi3X9jNwrx4jjZ7OVAfsyAwaiUSTBLfv2IqOfEW3t/fyJ/WHuFbz+9CVhSunp1x0s+u00o4tRL6Zi817X6SbAY+OtzMwjwnBq2ErlNMDcd50bWPWBfu8XH9UMXKMDBcFvwjkSwligLeYASDVhrzC8dEYXKajVsX5/D7Dw+P9VCGHVGAVcV9Gy4ON+PlfLPox96AzRcMj4tGhWcSoigwN8cJQF6ipTsqoSgK4ahMdZufYESm0R3k3X2NlDa60Ws1mHQaVhY6WXe4hbLWAJNTrRSnWEmy6jFqNTR1BPjRa/uYmx1Hst1AfqIFgFRHbNlnaWHf0Y8ulhcmsvH+c5FEgWBE5rbHNrPhaAv+cJRzipNYWZCATqelKNlKgkXHt1/YxQOv7CU/0UJJqhW9RsIXjGDUSTR1BEmyGXrsf3aOk4P1HZh1EksLEnD7Aryzr42rZ2Wcdl5U7+9YHDfRFVWsDAOiKPaoDhoqAxEqQxFGeo1IMBxFqxF7JKSp9M+tS3L4y8dHB1yOOFH43iWTyYo3nXrDM4hAeOx7Bb2+u4Gvn1fc7+uKohAMR9FrR6fXzJmMIAjoNBJ5nSJjUmrvbf65sYLDzT6+c9EknJ1mcZGozDV/2kCa3cA/v7igO1dmsIiigEjsb2jQSjx5+wLCUZkWj58Um6mHoLhubiat3hAPv3mAK/6wjh9dOZWiRBP5CWaMOiN2U2+hLQgCJamxyhlFUfjLR7UszHdS3eYjqzOP5kRO54Z6JHriDWkcauny2DGUE2CoKldRFGRl/NztTgTu+/dOnt9aPdbDGDa+tDyP+y8qOesuhv5QlHP+d02/pnqjxS0Ls/jRFVMH/f33dyM0Xu54R5vTXU7rYm+NC5tRi82gwWbUsq/OTXmzj9VTkkfkpq6mzU2aw9pnL7zDjR7KW3wcbvQgCQqrS+JJizOjCBL6AbrYrj3YSGGytbtXFsQiepIoDngffTGS55laujwGnOwP2t9kMxih0rX/oZ40giAgnX3z2mnxjXMLeWN3Hb4J0MX3VHz1nALuu6D/O/szGaNOYmVxIs9sHt2mpyeytaK9x+/XF4r0apzX1zzSX8T2bBQqwLAtp005rvdQKBwh22nq7h3UH1vKW/EEI9gMGrSiwJR0R3dOYDQqE47KGHR9X1a/8+J+Hv/C/F5LNYIgUJhspTDZyvmTjy3RhkIhIqEQLp+CzaTH0I/g6IrILS9KJHhCFFE3DF2hx8t5poqVYeJkf9Dh6GI6Xk6Ys4lMp4kvr8jj/wbYfn684jTruGsApZRnMjXt/rEeQi9avSHCERn7CXbrAJUtPjLijGdNcvxYLzPotBp0p0ht6sqF6fCHSLPr+drTO0i06llWmEhBogmtKJBo1TEp3dnrvXtrXXxc2szmsmbm5yUMaD7X6XTodKCLyieN9MgKBCNR9Bqpl1DSdBrh+UMRjDrNhI7GqWJFReUk3LWygP/sqOVok3eshzIk5ufEcdfK/CGvv58J/Oa9Uj4ubR7rYdDe2VG362KR7jCiKAodgXCvzuAPv7mf5UWJ3Dg/ayyGOuqMdT4EDGy547WdNTy9uZouW5WDDR4+6bQ5MOkk/t5PG4Dtle0A/H7NUZ7MP3mC7omcaknKqDv5zbCigLFTxAxUqIzHKtKxP0MmGFG5Z4qPyx9mW0VrjzbfJ2OCpwiddWglkVsXZY/1MIbMl8/JZ1XJ6FT/jEf21Lj41XuHxnoYQMxV9URjN1EUewmVv6w9woajLfxzY8WJu1AZQY7/28iyjC8YpsVzzIV2V3U7r+6qR+5nCr/3/AJS7bo+X9tbG/P52TcGvapkWcYTCBONDrxYYLwJFVDFyoCQZaX7D31igqrdqGVahgNZhnBU7t6uP1EyEiG4WE8Judth8cTXVE6PaRkOxuFvd0CYTxXbPsP5qLRprIfQTbs/THWb76TbNHYE+MvHZbT7wpQ2etTf7xjQ4Q/x4tZKLvndOpb9/EM2l7WgKAp/WnsUdz99pqwGDatKUug/vS02gbR4Q/zszX2sP9w0an9bRYh1bB6PAmQwqGJlAIiicFLDJK0kYtRJaCWxe7v/396dx0dV3v3D/5zZzuwzSSbbJCEr+xZZBUFEFEEErVZFu2CrUKvUjT6tVCyI5VHrcvt77K206s1t79Zqf7d7XYogoCwKQpCwJ5CNJJNkssy+n+v5I2QkZJJMwqzJ9/165QU5c+bMdZ3MzPmea/le0e4X7ApQAoHA9+tdhGhKTdb+yURy2YgUbPnZjKRcT8cfIoAdTvZUxr/7p4tYxEGr6D14PNZgwa/eLIP5/N28mONwsGb4LI2QCLxeP94/fA6/++AEqswOOL0BPPrOd5j7xx043tDR6/OkYhEqm+0oNoSeOsxf8N3xyq4q/PS/DsRsGYjjDVbsrTTDZInvbLhLlXzfvjEUqqWic3tnRBwIRPfN1lfk3TUzSCwWgzF0C1wufG7XCqHk0swblY6dv74KN5Uak2r6d7ujZ7rvRNOZSTTy79FGiwv7ziTOsgm/v2EctBd1+VjdPnx21IT73zyEH732Db6pags+dvecQhSGuPjR5zk6WjssONPUjmc+O90tv9IZswvn2l2oaev9Yt/m8OLevx3Eg//4FpXNth6PXxx0+gWGiqae+0XjbzspV4d0NQ+re3DfBV5/AB1Ob4RLNXAUrPSht8i3qzktErN8+nLhGg2CIMDu6nzDdG3rajURi0UQi8Xn1/cIBFdQ7dqHWlciw6hX4MXll+GLNfNwx4y8eBcnLJeN0Me7CP2SiEVReY+6vUKv4wtiKVUlw8Zl43DH9O/fM12f4W3Hm3Dv3w7i4yON6LhoHZncFAXS1HyP4/WVIoEMzvYTTdh6ogU/eeMw3L7BtUYKDPjX0RYs+9MevLjtNPadaUFAYDjbYsfxxp5jVT68YIX3aAagHMehKEMNhYSDdwAJEj2+AJosLvzXnmqYbfFvlRm+UwTCoJFL4fX5IRKJ4pYqu+uLSSQSQa2QgTGGRosbWoW02wyPrql/EokEgvD9nao3IIA/v4BXMk9bSyT5aSo8dfMkXDU6A/f//RD8iXBFDEEmESFD0/NiN1w4vPFd2ykvVYHfLR6La8dl9vj+6PocvrG3utfnL54YIvVqH6J98zSU/f6DY/D4BZjtl96C4PQG8OK2CozKVGPl3CJ8+F1Dj4kZAPDGvmo8uGAk9CpZ8MY0Wt/PGrkUAUGAyy9AFkaCOEFg2LK3Gg63D7+YWwC1Ut7vc6KNWlb6IZWIuy+ReQHfAEZXRwrHcTDqFT2mol44XkUkEgVbVKSi7rMPAOoaipTrxmfhn/fOincxevXoojHDej2akZnqkN0osZCilOKNn83A4onZff4N0jWhLwIGtQxaOd1LxkJFkw0BgQXHCkXK6SY7/p//PdLrtHmBAVXm71tcon0jqVfy4CWisL77RSIOE406zC4xgOdlIYOtWBu+32RhEAR2Ptrt+RhjLOT894GuQ3JhN080dQ3IBahrKJIuy9Mn7EXFMIxbVQCAl4ix+cdTI764W39mFKTikwfnBtem6cu14zJCbr+xNIc+ozHg9Qv4zTtH4rYUw0tfnMF3dbEbRO0XGFps4QVls0vSMCZLC6lYlBCzISlY6UPndC9Rj1k2gYDQa/9wbymRe3Px3P5It3hcuGpmIiReGmo4jkNJRv8XpVgrTFNiwZjQF8LhZHSWBrdOzY3Z680pMWDLz6YjW6fof2cAaarQAeXSycawni8ILOSAzi7+OLT+JjqPP4A3v6nFM5+dxHUvfhlM2BYPX5wy4wcv78VTn55AVbN1QN//fe3bZnOHbA1R8xIo+0ki14XjOKScX+QxEQLnxLwlTHCds3Aie+qiGUhcOFA3Ed50Q403wS4IHAf8ful4qAaRtTbeac+jYdWVxfi/B8/11psbUWa7p9s01f68vLOyx7bcFAUm5/a9Rs3RegvGG7UQGEOKMnQiMiBy6+gkM7cvgNMmG/75bR3sHj8++K4hJu+FcAkM+POus+hwePDEDWMgl/fsGgw1nqWv73I/Ax595wh+fd1o6BVS1LW7gjdVanly5l6id3IYLn5TuHyBiLWAdHU1df3/QpF4ja5jXliHQCAQlVac4aqvi0U8zChIxfxBtqoMtUAFAEoy1Pj5FYUxea2TJhtOmnq2dITq5v3vPVU4FOKu/uoxGf3eVMilnSkLJGJRyBlDpHNMYbvDi9e+Oosfbt6Hv31Ti/cPJ1agcqG3v63H6RZXt/dKV/d91/sh3OECGVo5Hrp2FP7P9grsO9uKIoMKzf3M6En060HUvpmqq6tx9913o7CwEAqFAsXFxVi/fj283u6jrbvGT1z4s3nz5mgV65IxxqDiJRFroRCJvh8/Emo1zotfezDHv/h4YrGYxq1EUKKsG8RxwORcHR5bMjbeRUk4v1k0GiNSlTF5rV2ne2bN7eqOvdCXvQy8vH16/9PiSzLUSZ+RNNreK6vHr//vd3hu6+mEa/3szUvbT+NvX9d0e6903UAMtNUzR6/AppsmwBcQcK7diYyLBnNfPNEi0a8HUesGOnnyJARBwJ///GeUlJTg6NGjWLlyJRwOB5577rlu+27ZsgWLFi0K/q7T9d0EGk/R/IP2d+zvo+tLX2SKuoYi43iDFY2W+K3oq5FL8MDVJchLVUGnkCJLJ4/bDJhExkvESFPLUNvWd7r7SOhtVsnFn7MbS4344mRzt22pKhnGZmmjVrbhxB9g2H7R+U10n580Y1dFG344NRdKXtrtPTOYVk+O43DtuKxeH0smUQtWFi1a1C0AKSoqwqlTp/DKK6/0CFb0ej2yskKfUNLThYHKpczN7+15lI8lfAwsLonHRmao8LMrCjGjIBUlmZrYFyAJSWPUxZWmCq9bMFRLz7VjM6nFJEI+OFwf7yIMijcg4P43D+HJmyYiNyU2rYHJIKYd1BaLBampqT22r169GgaDAdOnT8fmzZv77JfzeDywWq3dfoazaAQV3TPnRn4czVBiccY+nf3PZhfgg9VzcOfMfApUBmDe6PSoHj9NJcOPZo7AnTPDW6V7bLYWBWnfX4wkIg53zBwRreINK36/AL0yOQeSAsCOU2b88m+H0OH0wuJK/CUzYiFms4HOnDmDl156Cc8//3y37U8++SQWLFgAhUKB7du3Y82aNTCbzVi3bl3I4zz11FN44oknYlHkYe37zLnhj0Afjuye2GZJfWB+CR66dhTdfQ/CyChOMV+/dBx+NDN/QItdyqVivHffFXivrB4GDY8FYzIGNYOL9GR1+/DTWQXYeaoFHn9yjFe5WHm9Bb977whevH1KvIuSEAbcsrJhw4aQg2Iv/Pn222+7PaehoQGLFi3CrbfeinvuuafbY+vWrcOsWbNQWlqKNWvWYOPGjXj22Wd7ff21a9fCYrEEf+rq6gZaBZLA4p1dd6CvX98R2/Eq14ynboLBKs3TR2URSrGIw8Rc3aBW5U5RyfDzOYVYNtlIgUoEyaViVDTZoEnSabpdDtdaknK192gY8Kdj9erVWL58eZ/7FBQUBP/f0NCA+fPnY9asWfjLX/7S7/Evv/xyWK1WNDU1ITMzs8fjPM+D52mq3lDh8QeCaxcB8W+5Gejr779gldxYcMZ5vZtklqGVY2y2BkfrI9t1HBAYZJTPJKEoeQlKMjQRT6Efax6/AJvLC40isdIjxMOAgxWDwQCDwRDWvvX19Zg/fz6mTp2KLVu2hDWauaysDHK5HHq9fqBFI4jMTKFYujBQSTb+gIC9Z1pj+ppiji6Kl+J3i8dixZb98AUi13qXqpJhYk7izmAcruaMNCBNJUOr49IXJ4yXVocX+6vbsWBszxv34SZq33wNDQ246qqrkJeXh+eeew4tLS0wmUwwmUzBfT766CO8+uqrOHr0KM6cOYPXXnsNjz32GFatWkWtJ4OUTIFKsttdaY7p4Lebp+TgshH6mL3eUDS7xIAtd82IWHeQQirGbxeNjnuLIOmJMYbfLh4T72JcsipzZPI4XTxZItlErZN069atqKysRGVlJXJzu6/N0TUmQCqV4uWXX8YjjzwCQRBQVFSEjRs34v77749WsQYt2VosSPTF8gJ13fhMPH/rZLooRsCckQYsnZSN9w83XPKxXl8xDbNLwmtpJrHFcRx+OCUXT350HLYYD4SPpHmjIjOLLdmvX1ELVu666y7cddddfe5zcS6WRJbsf2gSefti0AWUppLhihIDHlhQQoFKBN02Pe+Sg5Xl0/MoUElwIhGHDC0PW0tyBivZOjmKw1i9ezigDnBCBikWM4FWXVmEH1yWg5IMyqcSSbOLDbhu/ODHAYzJ0uB3tKxBUrhrdkG8izBoRekqulE+j+bKETJIRVFOaz82W4slk7Ipi2WUPHPLJGw7sQ2BAfblXzM2Ay/dMQUKWfIODh9Obp+Wi02fnIDbN7B8KxIRB5lEhNnFabh5Si5GZ2kgl4qhkonxxt4anGi04lBtG5pt0RvAe+eMvhMMDqds4xSsEDJIcumlX6w0cgls7tBN1HfOyKNAJYqqzA4M9Gt+xax8PLZkHOW+SCICOIS5WDFUMjGuHpuJVXOLMN6oBceFHpv24DUjAQBWlxdL/7QHNa3frzlVaFBFZFDsbdNysWBs36unD5dABaBuIEKCBOH7hHD+MFZp3Xsm9Kq54crU8r0GKjdPycFtYay+SwYvN0UJjbz/+7UigwpXlKThzZUz8cSNEyhQSTLvl9X3uuqyVMxBfT4ZH8cBz99WipfuuAwTc3UQifpfmV6rkOHNlTOhPf8+ytEr8MCCkksu84yCVDx986SI3BANFdSyQsh5F/YNS/pJ8tVi82B35aUFK/4Aw6/mF+OrylYcrusIbr9qdDr+eMukfstALk26hse6JePwH9tOgzFgekEKSjLUuGp0BkQch3anF3qlFIUGFZQy+qpMVvUdLsgkInjPp91XSMVYMikbv1k0GgYVD47rTM8vE4sH1bWXo1fioWtGYU+lGa/8eCp+/0E5xmZrcaJx8MkHH47RkhqMMbTYPcjQyKP+WpeKY0m+Mp3VaoVOp4PFYoFWS0urk9h4fXcVnvzX8Us+TkmGGs/+cBKUMgnEIg4WlxdTRqQMq+bdeHP7AmAMNAZlCKtstmF3hRkun4DrJ2YhPy2y483sHj8sTi8cHj/OtDhwpsWO13ZXoWMQC52Oy9bio1/NicrSEKHYPX48/ekJ/HRWAUbFeGHUgVy/6XaBkAFyePx46YuKiByrstmO+g4XbphkjMjxyMBRU/vQV5KhieqMOjUvgZqX4IPD9RBxHBZNyMLITA1+8T8HB3ysdTeMjVmgAnSWfXpBKv742Sm8eNtkqBWJuZ4SBSuEDABjDB9+Vz+oO6bePPLP7zCrKA1p6qGZtXk4zVggw9uNpTnB/5dkaDAmS4OTJlvYz1fKxJgyIiUaRevTjaU5sLp8ONLQgXQ1j5GZiddLQZ3ihAxAfYcLT396KqLHXDm3cMgGKgCQ5Fm+CZI/VXu8XBPmmj4SEQeDmse0glTwcRrA/YMpuXjgH4exeVcV7B4/TBb3JU8iiCRqWSFkAAwKMdy+QGSOpeaxZGIWHrl2dESOl6hi2aRNooMaxgbnkWtH4cpR6fj8uAnVrU7sPNWMUZka3DOnEK/sOoPTTXbcNi0X624YB608vt0vbl8AFpcP7xw6h0/KGyERc1g9vwSzixMjSzMFK4QMgFzOY0ZhKr6qGPgdx4Zl4yCXiJGh5VGSrkFuioKyUxIyhIlEHGYUpmJGYSoAwGz3QCuXQiYRYcG4TDR2uDEyQwWRKL6dHE1WN9a+Wx5cjVwhE+O2aXm4Z25RXMt1IQpWCBmAZpsblc32AT+P44CZBakYa9RFoVSERBeNOYoMg5rH8QYrxhm10Mql0GZJwRjDgapWTC9Mi1u50tU8VswuQKvdg+/OWTDeqMWjCbZiNY1ZIWQAvjxtRqPFHda+Y7O1wZT8BjUPwxAelxKucJLtETKUjTNq8e+jjThlsoExhkaLG81Wd1zGBdV3uPDW/lrYPX7MLEzFhBwtUpRS/L8/mBjzsvSHWlYIGQBeIsK0EXoca7TB1c/Ylcm5Ojx180SYrG6kqmTgJTRFlhLdEQIca7ThF387hFSVDG0OL4rTVVgyOaf/J0aQLyDgjb3V+Ou+ary4rQJP/WAi0jVy/Ndd05GXmnjLfNA3ByEDsO2ECc/8cBJ0iv7j/P1Vbfj3MROON1gpUCGEBK2eXwK9Uoo2R+ciiFPzYz9d+eMjjfjLl2fh9gnI1stxeXEaHrpmFC6Lw9TpcFCwQkiYAgKDzelFk9UNk9XT7/5SsQi5KUpMztNHv3CEkKQhk4hw31XFkIg43FSajVunxW4dMH9AwOu7q/DYe+UAgGn5Kdj846kJn8GZuoEGwOL0QWACUlQ09mA48vj8WH/DOHx6vDms/e+bX4wJOTSglhDS08q5RVg62QitXAoVH/1LsdsXwIeHG/Bfe6pw0mSDiAMWjsvEzVNykalN/LWBKFgZAJ1SigAlRxqWAgKDqcOBryvMsHn8kIgAfx9jRdNUMpRSiwohpBccxyFbp4jJa9V3uLD6zUMoq+0IblsxuwDrl46PyetHAgUrA0QJroYnsYiD0yeASTh8dtTUa6CSpZXjlR9PwaQcLcTixG5WJYQMfW12N+56/RtUtDiC22RiEe6eUxjHUg0cBSuEhKGy2YbHPziBVJUUZ82OXve7bVpuwg5QI7EnCGzQif8YYxBYZ6ueVMxRrhMyKB5fADdeloMzLXZsPdYEvVKGF5eXIkcfm1adSInqANuCggJwHNft59FHH+22T21tLZYuXQqVSgWDwYAHHngAXq83msUiZMBkYhHWXj8GZ1rsYH30BP5/X1Ti0XeO4OMjjbErHElYds/gF7zkOA5iEQeZRESBChm07BQVfnFlEa6fkIXv1i/E7t/Ox/SC1KR7T0W9ZWXjxo1YuXJl8He1Wh38fyAQwJIlS5Ceno7du3ejtbUVK1asAGMML730UrSLRkjYvIEATjVaMCpTi4YOD7x9JDd760AdVLwESyZlx7CEJBFpFbJ4F4EMc4wxSCViXDt+cN9HgiDEfTkAIAbBikajQVZWVsjHtm7diuPHj6Ourg5GoxEA8Pzzz+Ouu+7Cpk2boNUm3jLVZHg6XNOB2lYnDtW09hmopGt4bFg6HtdPDP2eJyRRnGiwoDhdDZmUxlYNZZfagpIIgQoQgzwrzzzzDNLS0lBaWopNmzZ16+LZt28fJkyYEAxUAOC6666Dx+PBwYMHQx7P4/HAarV2+yEk2rQKKarbXFDI+l4Z9Yll47FkUnbSNbGS4efToya8+tVZHGuwxLsoJEEJAkuYGbBRbVl58MEHMWXKFKSkpGD//v1Yu3Ytqqqq8NprrwEATCYTMjMzuz0nJSUFMpkMJpMp5DGfeuopPPHEE9EsNiE95KTIsXB8BmpanSEfH5GqxAMLRuLqMRkxLhkhg/PIwtHxLgKJAa9fgEwyuHaJRFoVfsA12LBhQ49Bsxf/fPvttwCAhx9+GPPmzcOkSZNwzz33YPPmzXj99dfR2toaPF6oO1DGWK93pmvXroXFYgn+1NXVDbQKhAxYXqoavkAA442huyaXTs7GD6fmQt5Pk7rf74fF6UGbzQ2TxQ13P+sLEULIYPn8ATguYZB3Ihlwy8rq1auxfPnyPvcpKCgIuf3yyy8HAFRWViItLQ1ZWVn45ptvuu3T3t4On8/Xo8WlC8/z4HnKIEtiSyYRIVPJY3qBHla3H+X1FjTbvk+5/+9jTVh1ZTF0ir67iVpsXvx9fy3EIg7VrU7cMWMELi+K39LwhJChSyoRI2WIrEs24GDFYDDAYDAM6sXKysoAANnZnaOSZ82ahU2bNqGxsTG4bevWreB5HlOnTh3UaxASDXKpGHnpGvjggMniQkmGGgVpKgiCgLFGHUQccLzBilnFfQce2SlKrFk4Gk6PHwqZOGEGrxECdOZ0EXGXPiiTJCb/+ckBybj6edTGrOzbtw9ff/015s+fD51OhwMHDuDhhx/GsmXLMGLECADAwoULMW7cOPzkJz/Bs88+i7a2Nvz617/GypUraSYQSTgjMzUAY5g/NgOtdi+ONVixam4hdpwyY+txE+o73JiUq+t3nQ+O46CS990CM1iJMs2QJCcRB3Q4fUhR0ZTroSgZg5QuHGN9pbgavEOHDuG+++7DyZMn4fF4kJ+fj+XLl+M3v/kNlEplcL/a2lrcd999+OKLL6BQKHDnnXfiueeeC7urx2q1QqfTwWKxUIBDYuJMix1VzXZUtznQZvfiXLsTe8+2wWz3Ii9VgXvnFeOWKf2PX4kVu9uHw3Ud6HD5sHhCdtIvGeEPCEn9pUtIpCT7Z2Eg1++oBSuxQsEKiYfOgWt+7K9uQ0OHCyebbNDJZbC5/UhRSZGtU2BuiQFymRgGNZ8QAYLJ4oZOIYVMDFq3iJAhYDgFK7Q2ECGDIJWIoZeIsfB8VkiHx482hxdMCCAnVZ0QwcnFsnSJvww8ISR8yRyoDBQFK4REgIqXQCEVDYnxIl2NrYwlVp4FQsjwlfzfrIQkiKEQqAAI5ktijCHJe4kJIUPE0Ph2JYREHMcBHj8lrSOExB8FK4SQkBg4yKXUU0wIiT8KVgghPQQCAmi0CiEkUdBtEyFJyur2gQkMCplk0AuV9UY8jGYZEEISHwUrhCQIi8sHsYiDRMSFlVBOG6UsuIQQkmgoWCEkhvwBAQ6vH/6AgCarFzIJh5IMDQD0uwgiIYQMVxSsEBIDTq8fjDG4fAJc3gBy9AqkqeWwuX2oa3OittUJlVyCcdnaiHfpEEJIsqNghZAYCAjs/ErLEhxqbMe/jjTgdJMdqSopZhWlYXKeHsoojD0hhJChgIIVQqKozeHFa1+dxZgsDc61O7H1eBPmlKRjYq4Oiydkw+MPwB9gSFHKhlXqbEIIGQgKVgiJgnPtTuToFTBZ3Pj3MRNe+6oKq64swtu/mAVeIobT64dSRh8/Mry4fQFIxaJ+185qd3hxoLoNC8dnweMPwOryI13Dh/06jDFwHE2+H0ro25KQCGKMwWz3wuX14543vsWB6jbMHZmONDWPuSMNOFDVjin5egpUSL8YY/j3MRO+O2fB4doOWN0+tDu84KVi5OgV+PHlIzDeqENeqjLeRQXQ2dXZVxAiCKwzWBFxADg4vf7g86rMDozO1KDR6kKOToF2hxetDi9ueWUvOpxeqHgJ0tU8xmRrUJyuxsgMDTK1PNI1fDAoOVpvwdOfnkSDxYWGDhdGZWrwm+vGYFKebkAz5xhjOFDdjlSVDCUZ6uD2hg4X/n3MBK9fQIAxWF1+uLx+LJ6YjYpmOzS8BBkaHlKJCOONWsgl4pivrWV1+4bsLEGOJfniHwNZYpqQaHF6/Xj8/WO4ojgVbj/D05+egIqX4LeLxmD+mAxwHE01Jn1rsrqRruZR0+bAvjOtUMul+N+D5/Dl6ZaQ+y+bbESWTo4igwpjsrWYlKOLy8KTgsBwwmRFZZMNFpcfbn8AVWYHmqwe+AUGMQeY7V4cb7QiIDDIJCLoFFJYXT7IxCIIjMHpC4CXiMBLxFDzEtR3uMJ67ZIMNWYUpsLp8eObqjY0WtzBx5QyMa4oMUAmFuH26Xkw2z3gJWIoeTHGZGlgtnnRZHWhps2Fr8+2otnmQbXZAYNahjMtDmjkEqyaW4R0DQ8VL8HzW0+hutUZ9nmRiDiUZKih5iVQyyWYlKNDh8uHEalK3DotL2qz/7pabb1+IeHHwA3k+k3BCiERcLiuA+1OL8Qcwwv/PoVKswv/euAK5OqVNBaF9ODxB1Df7kKLzYPdlWa0O73415FGTM9PxeQ8HTbvOgu7xz+gYxYZVKhrdyIvVQm9QopUFY/6DhfUvPj8Nhm0Cgl4iRhpKhnc/gBmFqZhdJYm7Ndoc3hR2WxHtdmB0002NFrdqGtz4sg5y0BPQdRp5RL4AgwuX8/1rUQcIMTpynfd+ExkauWobnXi+glZuLE0BwpZ/3mVBsri9EGnTOwbJApWCEFnTpNoBQqMMTTbPDDbPSg0qILdOu0OD27789eoaLbjplIjrh6bCYmIw9VjMiCXisPusyfJw+MPYMfJFjDGsGBsZsi7WUFgOFTbjvJ6Cw5Ut+FgTTuarJ44lPZ7HAc8cs0o/GrByH73tbp9eOqTk/jH/toYlGz44Dhg6ogU6JUy8BIRAgJDgDHcVJoDBoYcvQJH6y043mjFQ9eMQrqaj0vrWbQM5PpNHedkyIpkoCIIDP7zTdgWlw/fnG3F+g+PodHixtVjMvD8rZPxydFGWFw+OL2dd3LvH27A+4cbkJuiwJlmG8Zm63DNuMyIlYkMnCAwOLx+aCLYJRcQGF796izq2114ZdcZvHh7KTiOQ6aWh8AANS/B8UYrfrh5X8ReMxLmlBiwrNTYY7svIEBy/oL4Xlk99lS24nBdO860OGJdxCGPMeDbmvYe2z8/3tRj23tl9cjUypGbokCGRg6pmMPkPD2m5acOqHUsWVHLCiFh+PcxEzx+AUsnZQcH9L28sxJHzlnwXV0HpoxIwfwx6Vj7bjl8gZ4fKV4iwl9/PgPF6SpoFbKE70uOpVa7B2nqzpkeDo8fFc12lObpu+1zrMGC8UYdmqxu6BRSnG1xYGy2Bi5vAOA6u+EK0lT49KgJU0boMTpTAyUvwdF6C76saMGYLA08PgEen4CyunasvLIIWoUUnx9rwjVjM6FTSuH2BbDzVAvmjUqHwBhsbj+ydHLUtTlR0WzD1WM6A02r24dvq9tgtnlhdftwpsWBT8o7A9ULyaWdd8q3TMnFu4fq4Q0IMTmf/SkyqPDQtaPQavfgZ1cUAgBsbh+cHj9e2FaBDw7Xw6DmIRWLUN3qQHJfIYY+o06O26bn4f6rSrD9ZDOuKE6DVNLZeitN8C5o6gYiJIL2V7WhocOFNocXOSkKXDZCjxONNrQ5PPjvPdXw+AWcNNmwen4JfIKAP+86G/I4hQYV7ptXhCtGpsOoV8S4FvHj9QvwBgTIxCJ8dsyEHL0CU/NT8Leva3Coph0cB/zhpokQizhYXF58XN6I6ydkQ8lLoJKJ4fAGUH7OgiPnOlDT6oDd40ejxY2GDjccXj8cHn+PADFFKcXIDA3K6y0hxyx0kUtFuGPGCADofI16Cy7L06PF7oHHJ+AHl+UgVSWDyxfAsQYLTjfZUWV2IBCvAQ8RouYl+O2i0fjJrAIAwNdnzHj4n991G6BKksuyyUYcrGlHu9MLnUKKRktnYM+fn51002U5mDIiBXmpyn5nbsVKQgQrO3fuxPz580M+tn//fkyfPr2zACHmwr/yyiu49957w3odClaGNkFg+LamHVaXDx6/AKNejhONNowzahEQBMilYhSnq8Na+O9S1LU58fLOMzjWYEFxuhqTcnXYdboFR+utMNs7xx78dFY+bpuWhwfeKsPZEE3m80en4+qxmVg+PS/h73gixRcQUG12YN/ZVuw704pPj5ogE4uwcHwmTplsqGi2d9tfK5fA6vZ3DgD1BbB4Yjb+daQBbl9itEoMBRNytPifn8+ERi6Bze3H0QYLNnx4jLp5honidBVunpKLu+cURv17sz8JEax4vV60tbV12/b4449j27ZtOHv2bDBI4TgOW7ZswaJFi4L76XQ6KBTh3XlSsDL0dCV0arK6caLRiikj9PAGGDqcXlQ02VHX7kSH04dUlQwijsPC8ZkQBCBNLYP4/IrFX1W0IFUlQ26KEhVNNsilYkzI0Q26TE1WNzgAaSoZznW4sPNUCxxeP/QKKSqbHZicp8M1YzNw5JwV9/39INqdvpDHmV2cig3LJmBkhnpIJq1qtLiQrev87FY02eD0BrDrdAsqm+348LuGOJeOAMA7v5yNUyYrvj7TimONVgpShjixiAu2BBakKTEiTYVNN01IiPw8CTHAViaTISsrK/i7z+fDhx9+iNWrV/f4ktbr9d32JYnlYE0bvjjZDKNegVum5IKXiOAXGCwuH9S8BLxEFNELb12bC0a9HBIRhx0nm3Goph0auRS8VIQdJ5txvNEKrVyKvFQlCtJUeH7r6WCTZrOts5VDLRNDxYtxvNGGujYnSkekQCuXoDhdjSqzAwvGZiBbp8CoTDVUvARSsQj/s68af91XA6lYhFVXFgXzIEzJ1+ONvdU43WTHPXMK0OHygQFwevyobXXC6Q3g71/X4vH3j8Lq7n26aUGaEhkaOU40WlGSrkYyxyoBgcHu8ffIFXHKZMMv/3YIHU4vBAbUtoWfl4JE33ijFm8fqMUXJ5thtnvjXRwSRZNzdTjRaMP0whTMH52BcUZtZ0sKYwkRqAxUzMasvPPOO7jttttQXV2NvLy87wvAccjJyYHb7UZhYSHuvvturFq1CiJReM3k1LISGW5fAK12D8x2L8YbtcGZNM02N6wuP4x6OQ7XdQTzKayYlY8NHx5Heb0Fk/P0uP+qYpTVdiBbL4dWIUWb3YMUlQxZOgVWv3kIY7O1WLNwFCQiEf73YB10ChmuHpPR70DTE41W1LU58dj7R+H1Cz0GMUaCTCyCQS1DQ4j+eo1ccn4GSe/jHsJVnK7ClaPS8dj1Y5M694rbF8C+s62wuny4sTQnuH1vpRlWtw9muxfbTzRhx6nQycwIIdGToeHR6vBi3qh0rFk4CqMyNQnb7ZwQ3UAXu/766wEAn3zySbftf/jDH7BgwQIoFAps374dv//977F27VqsW7cu5HE8Hg88nu/zE1itVuTl5VGw0ge3L4Aj5yxw+wJwegOwuX1459A5nG6yw6iXY1p+Kpptbri9AfAyMZRSMTRyKdocnRkeO1x+KKRiiDggwACVTAy5VAy/wMAYg8sbgMXlg83tR6vDE3I2zC1TcqFXSmG2e3DKZMPMwlTcOq0zaE1VyZCmlgHoTDpV2+pEQGDw+AWAA+QSMS4vSkV1qxOvfnUWuyvMSXPHnqqSYXZxGm6YlA25VIyrRmfEu0gR12hx4Q8fn8DHRxrjXRRChiWZWIR0DY/8NCVe+dFUtDk8yElRguOQsIEKEOVgZcOGDXjiiSf63OfAgQOYNm1a8Pdz584hPz8f//znP3HLLbf0+dznn38eGzduhMUSOiNib68/3IIVxjozMzLWOdvioyMN0MilOGmyovycBVa3D3kpndlTd51q7rN7IlFwHHqdJqmRS+D2BUIGQolELhUhRSlDjl6B8UYtxhm1mJijx9hsTdKNUXF6/TjeYEV9R2emVbvHD4vLB0FgYACOnLNAJhah1eHBWTNNcSUkXlTnM+DedFkO1i0ZFzIjriCwhEsoF9VgxWw2w2w297lPQUEB5HJ58Pcnn3wSL730Eurr6yGV9p2Mac+ePZgzZw5MJhMyM3sm0BruLSv1HU4wBuw82RxcA0Ii4vBNVRveO0wDGCOJ44AlE7MxIlUJm9sPvyCgzdE5FsMXEGBz+yEVc1gwJhNTC1KQqZUjJwmnJDPGYLK6setUC/ZXteF0sw0miztYV0JIcpiUq8ObKy+Hmu8+HDVRV6GO6gBbg8EAg8EQ9v6MMWzZsgU//elP+w1UAKCsrAxyuRx6vT7k4zzPg+fDXyo8WTR0uPDgW2WYmKPHVxUt0CmkEIs4FKerMKMwFQ0WNzI1cmTpeIg4DryEQ0WLG7yYg17Jw6Dh+2yZIAPHGLD9RDMWT8zC/NEZEBiDRi6BXinDZXn6hPzwh6PN4cXHRxpw5JwFDRYXjjdYe529RAhJHqebbPj5fx/An388FSmqzq713RVm7K9qxagsDVQyCWaXpIGXxHfK8mBEPd3+F198gaqqKtx99909Hvvoo49gMpkwa9YsKBQK7NixA4899hhWrVo1JAOSvnj9AhaMzcSZZjumF6YGx4D840Ad3txf121fvVKKDrq4RMSSidk402LH6SYbsnUKTMzRQS4VIVPbOVA4P62zVaXa7IDV7UO2ToGrRqtCBipuXwCVzXakqmQRT/pmtnuwZXcVDta0o6bNCbvHDzUvQX6aEmOztVh1ZVFwynBvTBY3/s/209h2ohkttviuS0MIiSwRB7h9AvZXteGml/fgkWtHweMXkJuiwIKxGZiQo0+IRHCDFfUBtnfeeSdqamqwZ8+eHo999tlnWLt2LSorKyEIAoqKinDPPffg/vvvh0QSXhyVTLOBtp9owj/218LtE1CSocYVJQbMKEztMf3zQHUb9lSaUdFsh93dmaGzw+VD5UUJtMilK0pXYcGYDBQa1MhLVaCy2Q4OgFouhU4hxYzCVGh4CbwBoUcCpeMNVmw9bsLpJhsEoXO2z2Uj9Jg/OgPiCA5qa7K6cc0Lu2ALMe4oR69Alk6OOSUGPHTNyF5bewSB4YuTzfj0aCOsLj/MDg/aHF5wAFJUMmjkUlSbHUkzcJkQ8r2rRqfj5im5cPsC0CukyNLJkaWTQy4Vo8XmgUIqTsis2Qk5GyhakilYcXj8KK+34FybEw2WzqXVfQEB88dkYNlkY/BC07XYmsAApUwMqbhzjZGGDhdqWp0wWd1otrmx81QLys/1nU6chC8vVQGHJwCPLwCDprO7TSrmMKckHb9fOq7H/luPmVDb5kSGVo6lk7LR6vDitMmGJpsbjAE3leZ0G9DWtfbM6SYbfjRzRHA9nFAYY3D7BBxvtGDr8SZs2V3d59oymVoer6+Y3m/iO8Y6F2QEOmcJNNvcONvsgFouwSmTDb5AAN4Ag8cnwO0LoMPlg9Prh9snoKLZhnPtLmrVIySBXDU6Hb+cV4zSEfqk695JiKRwpCcVL8HlRWlAUVqf+4lEXI9VYcUiDrkpCngDAvaeMeNEoxVmmwd+gdKQR0pdmyv4f0erEyqZGB6/gCarB8UZKiyZmA2nN4AMDY/KFjskYg4GNY9F47PAcZ3/N5TwqG11os3pxRcnm9Fkc8OoV+BAVRvOtjjwxclmiEWdq6XOG5XeowwBgeGPn53EJ0cb4Q8wiDgOEjGH4gw1SjLUSFVKkaKSIU0lQ4pKhlRl579ZWnmwj/pCXr8AqZgDx3FgjKG83oITjVYAQLvTh2arBy6fHw0dbtR3uNDY4YJfYAgI3wc1hJDEk67hMTlXjwk5WrQ6vJCGmZssWVHLSgy5fQEcqmmHT2BQycRQyyUYmaHp1o9Y2+rEkfoONFs9qG514Gi9BTa3H96AAN/5BeE8fgFevwBfQKDZGnEgFnEQGAsOZlZIO/+WSpm4c6GwFAVyU5TIT1NCp5RCxHHw+ju7/iTn/9bRSgr3XV0H3jpQi/oON06bbDBZ3ZBLRTDqFTjX5kqYlX8JIZEhEXH44dRc/GhmPoozVFDKkqcNgrqBkkzX1FG9QtZtfnxlsw1//6YWeytbcarJFscSkoGaXZyG5TNGQCUTwxdgEHGdd0JqXoIRacqQzbVOrx/l5yyoMjvQ5vTCH+h8X7TZvfALDA6PHx5/AClKGdRyCX5xZTHyUhWoaLaj2erGxo+Oh8zCSwgZ2nQKKdI1PHL0Ciwcn4nxRh0kIg5F6YkdvFCwEgNuXwCNFjd8AQF1rU60O70oSldBIeu8w1ZIxXD7BATOn169Qgq9UopDte1od/gwNT+lW7O92xdAu9OLmlYnOpxenGlx4NvqNjRZPWi2uWkdjyRnUPO4e04hphekYFpBKoDOGT47TjZjT6UZ350PUgaCpqoTQi524fdCuobHiFQlphekYmp+CsYbtQk10JaClTjxBwS0nQ84vjzdghabBwGBnU9NL6DF5sWXFS3I1PIw6hTBsQEqXozGDjesbj/MdppSOlSJOCBbp4BCJoYgMFS1UtZXQkhsyMQiKGRiSM+PtZOKRbhshB68RIRRmRpo5FIoZWJk6eRQysTQK2VQycRRzSdFA2wjZN+ZVryy6wx8fgEZWh4BgSFbJ4dIxMHm9oOXiCCXiiGXiMFLRZB3/S4V44oSA4rSVWAMEBhDu8OH2jYHxmVrUN/hhjcgwHV+loXLF4BMIoJc2rm+A2OgoGUIEhhQ3+Hqf0dCCIkwb0CA19U5Zq2rpb68PvSyNl3kUhG0cil+dXUJfjKrINpF7BMFK31osrrx5enorBwrk4ig5ju7jFQyCVJVMhSlq9Dh9KG+wwUzpVQhhBASR26fALfPg5rW+OdfomAlTrx+AW1+L9oGNkyBEEIIGXaG9sRsQgghhCQ9ClYIIYQQktAoWCGEEEJIQqNghRBCCCEJjQbY9qE0T49nbpkY72IQQgghcTMqUxPvIlCw0pcCgwoFBlW8i0EIIYQMa9QNRAghhJCERsEKIYQQQhIaBSuEEEIISWgUrBBCCCEkoVGwQgghhJCERsEKIYQQQhIaBSuEEEIISWgUrBBCCCEkoSV9UjjGGADAarXGuSSEEEIICVfXdbvrOt6XpA9WbDYbACAvLy/OJSGEEELIQNlsNuh0uj734Vg4IU0CEwQBDQ0N0Gg04Dgu3sWJOqvViry8PNTV1UGr1ca7ODE1XOs+XOsNUN2HY92Ha72B4Vd3xhhsNhuMRiNEor5HpSR9y4pIJEJubm68ixFzWq12WLyZQxmudR+u9Qao7sOx7sO13sDwqnt/LSpdaIAtIYQQQhIaBSuEEEIISWgUrCQZnuexfv168Dwf76LE3HCt+3CtN0B1H451H671BoZ33fuT9ANsCSGEEDK0UcsKIYQQQhIaBSuEEEIISWgUrBBCCCEkoVGwQgghhJCERsFKAtu0aRNmz54NpVIJvV4fch+O43r8bN68uds+5eXlmDdvHhQKBXJycrBx48aw1mKIl3DqXVtbi6VLl0KlUsFgMOCBBx6A1+vttk+y1TuUgoKCHn/fRx99tNs+4ZyLZPXyyy+jsLAQcrkcU6dOxVdffRXvIkXUhg0bevx9s7Kygo8zxrBhwwYYjUYoFApcddVVOHbsWBxLPHhffvklli5dCqPRCI7j8P7773d7PJy6ejwe/OpXv4LBYIBKpcKyZctw7ty5GNZi4Pqr91133dXjPXD55Zd32ycZ6x1pFKwkMK/Xi1tvvRW//OUv+9xvy5YtaGxsDP6sWLEi+JjVasW1114Lo9GIAwcO4KWXXsJzzz2HF154IdrFH7T+6h0IBLBkyRI4HA7s3r0bb731Ft555x2sWbMmuE8y1rs3Gzdu7Pb3XbduXfCxcM5Fsnr77bfx0EMP4bHHHkNZWRnmzp2LxYsXo7a2Nt5Fi6jx48d3+/uWl5cHH/vjH/+IF154AX/6059w4MABZGVl4dprrw2uiZZMHA4HJk+ejD/96U8hHw+nrg899BDee+89vPXWW9i9ezfsdjtuuOEGBAKBWFVjwPqrNwAsWrSo23vgk08+6fZ4MtY74hhJeFu2bGE6nS7kYwDYe++91+tzX375ZabT6Zjb7Q5ue+qpp5jRaGSCIES4pJHVW70/+eQTJhKJWH19fXDbP/7xD8bzPLNYLIyx5K73hfLz89l//Md/9Pp4OOciWc2YMYPde++93baNGTOGPfroo3EqUeStX7+eTZ48OeRjgiCwrKws9vTTTwe3ud1uptPp2ObNm2NUwui4+HsrnLp2dHQwqVTK3nrrreA+9fX1TCQSsc8++yxmZb8Uob6vV6xYwW688cZenzMU6h0J1LIyBKxevRoGgwHTp0/H5s2bIQhC8LF9+/Zh3rx53ZIMXXfddWhoaEB1dXUcSnvp9u3bhwkTJsBoNAa3XXfddfB4PDh48GBwn6FS72eeeQZpaWkoLS3Fpk2bunXxhHMukpHX68XBgwexcOHCbtsXLlyIvXv3xqlU0VFRUQGj0YjCwkIsX74cZ8+eBQBUVVXBZDJ1Owc8z2PevHlD7hyEU9eDBw/C5/N128doNGLChAlJfz527tyJjIwMjBo1CitXrkRzc3PwsaFc74FI+oUMh7snn3wSCxYsgEKhwPbt27FmzRqYzeZgV4HJZEJBQUG352RmZgYfKywsjHWRL5nJZArWoUtKSgpkMhlMJlNwn6FQ7wcffBBTpkxBSkoK9u/fj7Vr16KqqgqvvfYagPDORTIym80IBAI96paZmZnU9brYzJkz8de//hWjRo1CU1MT/vCHP2D27Nk4duxYsJ6hzkFNTU08ihs14dTVZDJBJpMhJSWlxz7J/J5YvHgxbr31VuTn56OqqgqPP/44rr76ahw8eBA8zw/Zeg8UtazEWKgBdRf/fPvtt2Efb926dZg1axZKS0uxZs0abNy4Ec8++2y3fTiO6/Y7Oz/I9OLt0RTpeocqO2Os2/ZEqHcoAzkXDz/8MObNm4dJkybhnnvuwebNm/H666+jtbU1eLxwzkWyCvU3HAr16rJ48WLccsstmDhxIq655hp8/PHHAIA33ngjuM9QPwcXGkxdk/183H777ViyZAkmTJiApUuX4tNPP8Xp06eD74XeJHu9B4paVmJs9erVWL58eZ/7XNwiMBCXX345rFYrmpqakJmZiaysrB7Rd1cT48V3MdEUyXpnZWXhm2++6batvb0dPp8vWKdEqXcol3IuumYJVFZWIi0tLaxzkYwMBgPEYnHIv2Ey16s/KpUKEydOREVFBW666SYAnS0K2dnZwX2G4jnomgHVV12zsrLg9XrR3t7erZWhubkZs2fPjm2Boyg7Oxv5+fmoqKgAMHzq3R9qWYkxg8GAMWPG9Pkjl8sHffyysjLI5fLglN9Zs2bhyy+/7DbOYevWrTAajZcUFA1UJOs9a9YsHD16FI2NjcFtW7duBc/zmDp1anCfRKh3KJdyLsrKygAg+IUezrlIRjKZDFOnTsXnn3/ebfvnn38+pL+gPR4PTpw4gezsbBQWFiIrK6vbOfB6vdi1a9eQOwfh1HXq1KmQSqXd9mlsbMTRo0eH1PlobW1FXV1d8DM+XOrdr7gN7SX9qqmpYWVlZeyJJ55garWalZWVsbKyMmaz2RhjjH344YfsL3/5CysvL2eVlZXs1VdfZVqtlj3wwAPBY3R0dLDMzEx2xx13sPLycvbuu+8yrVbLnnvuuXhVq1/91dvv97MJEyawBQsWsEOHDrFt27ax3Nxctnr16uAxkrHeF9u7dy974YUXWFlZGTt79ix7++23mdFoZMuWLQvuE865SFZvvfUWk0ql7PXXX2fHjx9nDz30EFOpVKy6ujreRYuYNWvWsJ07d7KzZ8+yr7/+mt1www1Mo9EE6/j0008znU7H3n33XVZeXs7uuOMOlp2dzaxWa5xLPnA2my34WQYQfG/X1NQwxsKr67333styc3PZtm3b2KFDh9jVV1/NJk+ezPx+f7yq1a++6m2z2diaNWvY3r17WVVVFduxYwebNWsWy8nJSfp6RxoFKwlsxYoVDECPnx07djDGGPv0009ZaWkpU6vVTKlUsgkTJrAXX3yR+Xy+bsc5cuQImzt3LuN5nmVlZbENGzYk9PTd/urNWGdAs2TJEqZQKFhqaipbvXp1t2nKjCVfvS928OBBNnPmTKbT6ZhcLmejR49m69evZw6Ho9t+4ZyLZPWf//mfLD8/n8lkMjZlyhS2a9eueBcpom6//XaWnZ3NpFIpMxqN7Oabb2bHjh0LPi4IAlu/fj3LyspiPM+zK6+8kpWXl8exxIO3Y8eOkJ/rFStWMMbCq6vL5WKrV69mqampTKFQsBtuuIHV1tbGoTbh66veTqeTLVy4kKWnpzOpVMpGjBjBVqxY0aNOyVjvSOMYS7KUnoQQQggZVmjMCiGEEEISGgUrhBBCCEloFKwQQgghJKFRsEIIIYSQhEbBCiGEEEISGgUrhBBCCEloFKwQQgghJKFRsEIIIYSQhEbBCiGEEEISGgUrhBBCCEloFKwQQgghJKFRsEIIIYSQhPb/A1dq46p7fPOHAAAAAElFTkSuQmCC"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 9
  },
  {
   "cell_type": "code",
   "id": "e0b9b262-7314-4c36-ab41-28b4484b9393",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:18:26.124042Z",
     "start_time": "2024-04-15T13:15:53.438594Z"
    }
   },
   "source": [
    "%%time \n",
    "# Load as dask-geopandas dataframe and intersect with land mask\n",
    "ddf = dask_geopandas.from_geopandas(gdf, npartitions=4)\n",
    "ind_land = ddf.intersects(land_highres.unary_union).compute()\n",
    "gdf_land = gdf[ind_land].reset_index(drop=True)\n",
    "\n",
    "# Add simple UTM bounds (left, down, right, up)\n",
    "gdf_land['utm_bounds'] = gdf_land.utm_wkt.apply(lambda x: wkt.loads(x).bounds).astype(str)\n",
    "gdf_land.head()"
   ],
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 7min 39s, sys: 3min 31s, total: 11min 10s\n",
      "Wall time: 2min 32s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "    tile                                           geometry   epsg  \\\n",
       "0  01CEH  POLYGON ((-177.00144 -82.85238, -169.14642 -82...  32701   \n",
       "1  01CEJ  POLYGON ((-177.00128 -81.95638, -170.00719 -81...  32701   \n",
       "2  01FBE  POLYGON ((180.00000 -50.59941, 178.76332 -50.5...  32701   \n",
       "3  01FBF  POLYGON ((180.00000 -49.70001, 178.84184 -49.6...  32701   \n",
       "4  01GBH  POLYGON ((180.00000 -47.89995, 178.98771 -47.8...  32701   \n",
       "\n",
       "                                             utm_wkt  \\\n",
       "0  POLYGON ((499980 800020,499980 690220,609780 6...   \n",
       "1  POLYGON ((499980 900040,499980 790240,609780 7...   \n",
       "2  POLYGON ((199980 4500040,199980 4390240,309780...   \n",
       "3  POLYGON ((199980 4600000,199980 4490200,309780...   \n",
       "4  POLYGON ((199980 4800040,199980 4690240,309780...   \n",
       "\n",
       "                                   utm_bounds  \n",
       "0    (499980.0, 690220.0, 609780.0, 800020.0)  \n",
       "1    (499980.0, 790240.0, 609780.0, 900040.0)  \n",
       "2  (199980.0, 4390240.0, 309780.0, 4500040.0)  \n",
       "3  (199980.0, 4490200.0, 309780.0, 4600000.0)  \n",
       "4  (199980.0, 4690240.0, 309780.0, 4800040.0)  "
      ],
      "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>tile</th>\n",
       "      <th>geometry</th>\n",
       "      <th>epsg</th>\n",
       "      <th>utm_wkt</th>\n",
       "      <th>utm_bounds</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01CEH</td>\n",
       "      <td>POLYGON ((-177.00144 -82.85238, -169.14642 -82...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((499980 800020,499980 690220,609780 6...</td>\n",
       "      <td>(499980.0, 690220.0, 609780.0, 800020.0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01CEJ</td>\n",
       "      <td>POLYGON ((-177.00128 -81.95638, -170.00719 -81...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((499980 900040,499980 790240,609780 7...</td>\n",
       "      <td>(499980.0, 790240.0, 609780.0, 900040.0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01FBE</td>\n",
       "      <td>POLYGON ((180.00000 -50.59941, 178.76332 -50.5...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((199980 4500040,199980 4390240,309780...</td>\n",
       "      <td>(199980.0, 4390240.0, 309780.0, 4500040.0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>01FBF</td>\n",
       "      <td>POLYGON ((180.00000 -49.70001, 178.84184 -49.6...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((199980 4600000,199980 4490200,309780...</td>\n",
       "      <td>(199980.0, 4490200.0, 309780.0, 4600000.0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01GBH</td>\n",
       "      <td>POLYGON ((180.00000 -47.89995, 178.98771 -47.8...</td>\n",
       "      <td>32701</td>\n",
       "      <td>POLYGON ((199980 4800040,199980 4690240,309780...</td>\n",
       "      <td>(199980.0, 4690240.0, 309780.0, 4800040.0)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 10
  },
  {
   "cell_type": "code",
   "id": "99616112-405d-4181-a8cb-d0acf23a3a3f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T13:18:26.157521Z",
     "start_time": "2024-04-15T13:18:26.132226Z"
    }
   },
   "source": [
    "print(f\"Number of tiles: {len(gdf_land)}\")\n",
    "gdf_land.to_parquet(\"sentinel-2-grid_LAND.parquet\")"
   ],
   "execution_count": 12,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of tiles: 20430\n"
     ]
    }
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}