diff --git a/analysis/new analysis Aug 2025/analayis11_2020_nooutliers.ipynb b/analysis/new analysis Aug 2025/analayis11_2020_nooutliers.ipynb
new file mode 100644
index 0000000..8c79ba9
--- /dev/null
+++ b/analysis/new analysis Aug 2025/analayis11_2020_nooutliers.ipynb	
@@ -0,0 +1,1799 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "8e73506c",
+   "metadata": {},
+   "source": [
+    "# Analysis of Colorado Spills Data\n",
+    "## Attempt 11: Using 2020 as the Year of Analysis for Change and Outlier Removal\n",
+    "### Focus on top 3 counties\n",
+    "#### [David P. Adams](https://dadams.io)\n",
+    "##### [GitHub Repository](https://github.com/dadams-AU/colorado_spills)\n",
+    "##### Date: 2025-08-14"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "f476d20e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import pandas as pd\n",
+    "import geopandas as gpd\n",
+    "from sqlalchemy import create_engine\n",
+    "from dotenv import load_dotenv\n",
+    "\n",
+    "# Load environment variables (e.g., DB_USER, DB_PASSWORD)\n",
+    "load_dotenv()\n",
+    "\n",
+    "# Database credentials from .env or shell\n",
+    "db_name = 'colorado_spills'\n",
+    "user = os.getenv('DB_USER')\n",
+    "password = os.getenv('DB_PASSWORD')\n",
+    "host = os.getenv('DB_HOST')\n",
+    "port = os.getenv('DB_PORT')\n",
+    "\n",
+    "# SQLAlchemy engine for PostgreSQL/PostGIS\n",
+    "engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')\n",
+    "\n",
+    "# --- Load data with geometry preserved ---\n",
+    "try:\n",
+    "    spills = gpd.read_postgis(\n",
+    "        sql='SELECT * FROM spills_with_ruca',\n",
+    "        con=engine,\n",
+    "        geom_col='geometry',\n",
+    "        crs='EPSG:4326'\n",
+    "    )\n",
+    "except Exception as e:\n",
+    "    print(\"GeoPandas failed to load geometry. Falling back to plain Pandas.\")\n",
+    "    spills = pd.read_sql('SELECT * FROM spills_with_ruca', engine)\n",
+    "    # Optional: spills.drop(columns='geometry', inplace=True, errors='ignore')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "03a508ea",
+   "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>Document #</th>\n",
+       "      <th>Report</th>\n",
+       "      <th>Operator</th>\n",
+       "      <th>Operator #</th>\n",
+       "      <th>Tracking #</th>\n",
+       "      <th>Initial Report Date</th>\n",
+       "      <th>Date of Discovery</th>\n",
+       "      <th>Spill Type</th>\n",
+       "      <th>Qtr Qtr</th>\n",
+       "      <th>Section</th>\n",
+       "      <th>...</th>\n",
+       "      <th>poverty_population</th>\n",
+       "      <th>unemployed_population</th>\n",
+       "      <th>percent_white</th>\n",
+       "      <th>percent_hispanic</th>\n",
+       "      <th>percent_poverty</th>\n",
+       "      <th>unemployment_rate</th>\n",
+       "      <th>geometry</th>\n",
+       "      <th>ruca_code</th>\n",
+       "      <th>ruca_description</th>\n",
+       "      <th>rurality</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>400811665</td>\n",
+       "      <td>I w/S</td>\n",
+       "      <td>CHEVRON USA INC</td>\n",
+       "      <td>16700</td>\n",
+       "      <td>400811665</td>\n",
+       "      <td>03/19/2015</td>\n",
+       "      <td>06/04/2012</td>\n",
+       "      <td>Historical</td>\n",
+       "      <td>nwne</td>\n",
+       "      <td>19</td>\n",
+       "      <td>...</td>\n",
+       "      <td>249.0</td>\n",
+       "      <td>86.0</td>\n",
+       "      <td>84.676145</td>\n",
+       "      <td>11.650869</td>\n",
+       "      <td>9.834123</td>\n",
+       "      <td>3.396524</td>\n",
+       "      <td>POINT (-108.88214 40.1324)</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Rural area</td>\n",
+       "      <td>Rural</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>400811979</td>\n",
+       "      <td>I</td>\n",
+       "      <td>WHITING OIL &amp; GAS CORPORATION</td>\n",
+       "      <td>96155</td>\n",
+       "      <td>400811979</td>\n",
+       "      <td>03/19/2015</td>\n",
+       "      <td>03/18/2015</td>\n",
+       "      <td>Recent</td>\n",
+       "      <td>SWSW</td>\n",
+       "      <td>33</td>\n",
+       "      <td>...</td>\n",
+       "      <td>1193.0</td>\n",
+       "      <td>253.0</td>\n",
+       "      <td>86.064736</td>\n",
+       "      <td>14.412266</td>\n",
+       "      <td>20.323680</td>\n",
+       "      <td>4.310051</td>\n",
+       "      <td>POINT (-103.87652 40.78919)</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Rural area</td>\n",
+       "      <td>Rural</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>400812102</td>\n",
+       "      <td>I w/S</td>\n",
+       "      <td>CHEVRON USA INC</td>\n",
+       "      <td>16700</td>\n",
+       "      <td>400812102</td>\n",
+       "      <td>03/19/2015</td>\n",
+       "      <td>07/11/2012</td>\n",
+       "      <td>Historical</td>\n",
+       "      <td>nese</td>\n",
+       "      <td>32</td>\n",
+       "      <td>...</td>\n",
+       "      <td>249.0</td>\n",
+       "      <td>86.0</td>\n",
+       "      <td>84.676145</td>\n",
+       "      <td>11.650869</td>\n",
+       "      <td>9.834123</td>\n",
+       "      <td>3.396524</td>\n",
+       "      <td>POINT (-108.86143 40.09959)</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Rural area</td>\n",
+       "      <td>Rural</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>400812258</td>\n",
+       "      <td>I</td>\n",
+       "      <td>KERR MCGEE OIL &amp; GAS ONSHORE LP</td>\n",
+       "      <td>47120</td>\n",
+       "      <td>400812258</td>\n",
+       "      <td>03/20/2015</td>\n",
+       "      <td>03/19/2015</td>\n",
+       "      <td>Historical</td>\n",
+       "      <td>NENW</td>\n",
+       "      <td>13</td>\n",
+       "      <td>...</td>\n",
+       "      <td>25.0</td>\n",
+       "      <td>10.0</td>\n",
+       "      <td>82.352941</td>\n",
+       "      <td>18.242740</td>\n",
+       "      <td>1.861504</td>\n",
+       "      <td>0.744602</td>\n",
+       "      <td>POINT (-104.9543 40.0574)</td>\n",
+       "      <td>2</td>\n",
+       "      <td>Metropolitan high commuting</td>\n",
+       "      <td>Urban</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>400812277</td>\n",
+       "      <td>I</td>\n",
+       "      <td>NOBLE ENERGY INC</td>\n",
+       "      <td>100322</td>\n",
+       "      <td>400812277</td>\n",
+       "      <td>03/20/2015</td>\n",
+       "      <td>03/18/2015</td>\n",
+       "      <td>Recent</td>\n",
+       "      <td>SENE</td>\n",
+       "      <td>28</td>\n",
+       "      <td>...</td>\n",
+       "      <td>268.0</td>\n",
+       "      <td>62.0</td>\n",
+       "      <td>89.013174</td>\n",
+       "      <td>15.212528</td>\n",
+       "      <td>6.661695</td>\n",
+       "      <td>1.541138</td>\n",
+       "      <td>POINT (-104.54757 40.2861)</td>\n",
+       "      <td>2</td>\n",
+       "      <td>Metropolitan high commuting</td>\n",
+       "      <td>Urban</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 122 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Document # Report                         Operator  Operator #  Tracking #  \\\n",
+       "0   400811665  I w/S                  CHEVRON USA INC       16700   400811665   \n",
+       "1   400811979      I    WHITING OIL & GAS CORPORATION       96155   400811979   \n",
+       "2   400812102  I w/S                  CHEVRON USA INC       16700   400812102   \n",
+       "3   400812258      I  KERR MCGEE OIL & GAS ONSHORE LP       47120   400812258   \n",
+       "4   400812277      I                 NOBLE ENERGY INC      100322   400812277   \n",
+       "\n",
+       "  Initial Report Date Date of Discovery  Spill Type Qtr Qtr  Section  ...  \\\n",
+       "0          03/19/2015        06/04/2012  Historical  nwne         19  ...   \n",
+       "1          03/19/2015        03/18/2015      Recent  SWSW         33  ...   \n",
+       "2          03/19/2015        07/11/2012  Historical  nese         32  ...   \n",
+       "3          03/20/2015        03/19/2015  Historical  NENW         13  ...   \n",
+       "4          03/20/2015        03/18/2015      Recent  SENE         28  ...   \n",
+       "\n",
+       "  poverty_population unemployed_population percent_white  percent_hispanic  \\\n",
+       "0              249.0                  86.0     84.676145         11.650869   \n",
+       "1             1193.0                 253.0     86.064736         14.412266   \n",
+       "2              249.0                  86.0     84.676145         11.650869   \n",
+       "3               25.0                  10.0     82.352941         18.242740   \n",
+       "4              268.0                  62.0     89.013174         15.212528   \n",
+       "\n",
+       "   percent_poverty unemployment_rate                     geometry ruca_code  \\\n",
+       "0         9.834123          3.396524   POINT (-108.88214 40.1324)        10   \n",
+       "1        20.323680          4.310051  POINT (-103.87652 40.78919)        10   \n",
+       "2         9.834123          3.396524  POINT (-108.86143 40.09959)        10   \n",
+       "3         1.861504          0.744602    POINT (-104.9543 40.0574)         2   \n",
+       "4         6.661695          1.541138   POINT (-104.54757 40.2861)         2   \n",
+       "\n",
+       "              ruca_description rurality  \n",
+       "0                   Rural area    Rural  \n",
+       "1                   Rural area    Rural  \n",
+       "2                   Rural area    Rural  \n",
+       "3  Metropolitan high commuting    Urban  \n",
+       "4  Metropolitan high commuting    Urban  \n",
+       "\n",
+       "[5 rows x 122 columns]"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# show head\n",
+    "spills.head()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "08bf0d17",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# use longitude and latitude to create a GeoDataFrame from spills data\n",
+    "spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])\n",
+    "spills = gpd.GeoDataFrame(spills, geometry='geometry')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "dc5dfab9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326') \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "ba3ca567",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "# Ensure date columns are in datetime format\n",
+    "spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])\n",
+    "spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])\n",
+    "\n",
+    "# create a report year column\n",
+    "spills_gdf['Report Year'] = spills_gdf['Initial Report Date'].dt.year\n",
+    "\n",
+    "# create a discovery year column\n",
+    "spills_gdf['Discovery Year'] = spills_gdf['Date of Discovery'].dt.year\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "1c537872",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(Timestamp('2015-03-19 00:00:00'), Timestamp('2025-03-17 00:00:00'))"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "spills_gdf['Initial Report Date'].min(), spills_gdf['Initial Report Date'].max()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "ed6768d0",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(Timestamp('1994-11-14 00:00:00'), Timestamp('2025-03-16 00:00:00'))"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# show first and last dates\n",
+    "spills_gdf['Date of Discovery'].min(), spills_gdf['Date of Discovery'].max()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "faeebec8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def classify_rurality(ruca_code):\n",
+    "    if pd.isna(ruca_code):\n",
+    "        return 'Unknown'\n",
+    "    elif ruca_code <= 3:\n",
+    "        return 'Urban'\n",
+    "    elif ruca_code <= 6:\n",
+    "        return 'Suburban'\n",
+    "    else:\n",
+    "        return 'Rural'\n",
+    "\n",
+    "spills_gdf['ruca_code'] = pd.to_numeric(spills_gdf['ruca_code'], errors='coerce')\n",
+    "spills_gdf['rurality'] = spills_gdf['ruca_code'].apply(classify_rurality)\n",
+    "spills_gdf = spills_gdf[spills_gdf['rurality'] != 'Unknown']  # drop unknowns\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "c53017a5",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
+       "columns": [
+        {
+         "name": "Discovery Year",
+         "rawType": "int32",
+         "type": "integer"
+        },
+        {
+         "name": "count",
+         "rawType": "int64",
+         "type": "integer"
+        }
+       ],
+       "ref": "83b9d48b-aa05-4949-9f63-5e684b192323",
+       "rows": [
+        [
+         "1994",
+         "1"
+        ],
+        [
+         "2004",
+         "1"
+        ],
+        [
+         "2009",
+         "1"
+        ],
+        [
+         "2011",
+         "4"
+        ],
+        [
+         "2012",
+         "7"
+        ],
+        [
+         "2013",
+         "9"
+        ],
+        [
+         "2014",
+         "27"
+        ],
+        [
+         "2015",
+         "1192"
+        ],
+        [
+         "2016",
+         "1376"
+        ],
+        [
+         "2017",
+         "1603"
+        ],
+        [
+         "2018",
+         "1584"
+        ],
+        [
+         "2019",
+         "1561"
+        ],
+        [
+         "2020",
+         "1171"
+        ],
+        [
+         "2021",
+         "1794"
+        ],
+        [
+         "2022",
+         "2170"
+        ],
+        [
+         "2023",
+         "2172"
+        ],
+        [
+         "2024",
+         "2700"
+        ],
+        [
+         "2025",
+         "447"
+        ]
+       ],
+       "shape": {
+        "columns": 1,
+        "rows": 18
+       }
+      },
+      "text/plain": [
+       "Discovery Year\n",
+       "1994       1\n",
+       "2004       1\n",
+       "2009       1\n",
+       "2011       4\n",
+       "2012       7\n",
+       "2013       9\n",
+       "2014      27\n",
+       "2015    1192\n",
+       "2016    1376\n",
+       "2017    1603\n",
+       "2018    1584\n",
+       "2019    1561\n",
+       "2020    1171\n",
+       "2021    1794\n",
+       "2022    2170\n",
+       "2023    2172\n",
+       "2024    2700\n",
+       "2025     447\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# show frequency of discovery by year\n",
+    "spills_gdf['Discovery Year'].value_counts().sort_index()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "1ce3eefd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from datetime import datetime\n",
+    "\n",
+    "\n",
+    "start_date = datetime(2015, 1, 1)\n",
+    "end_date = datetime(2024, 12, 31)  # exclusive end date\n",
+    "\n",
+    "spills_gdf = spills_gdf[\n",
+    "    (spills_gdf['Date of Discovery'] >= start_date) &\n",
+    "    (spills_gdf['Date of Discovery'] < end_date)\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "21324e82",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot discovery year frequency\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "spills_gdf['Discovery Year'].value_counts().sort_index().plot(kind='bar')\n",
+    "plt.title('Frequency of Discovery by Year')\n",
+    "plt.xlabel('Year')\n",
+    "plt.ylabel('Frequency')\n",
+    "plt.xticks(rotation=45)\n",
+    "plt.grid(axis='y')\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "fda5768c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Separate Historical vs. Recent Spills\n",
+    "historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']\n",
+    "recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "78d1b001",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# only use the top 3 counties by number of spills\n",
+    "top_counties = spills_gdf['county'].value_counts().nlargest(3).index.tolist()\n",
+    "# filter spills_gdf to only include top counties\n",
+    "spills_gdf = spills_gdf[spills_gdf['county'].isin(top_counties)]\n",
+    "# filter historical spills to only include top counties\n",
+    "historical_spills = historical_spills[historical_spills['county'].isin(top_counties)]\n",
+    "# filter recent spills to only include top counties\n",
+    "recent_spills = recent_spills[recent_spills['county'].isin(top_counties)]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "9d110ce8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKUAAAJOCAYAAABm7rQwAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjUsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvWftoOwAAAAlwSFlzAAAPYQAAD2EBqD+naQAAZIJJREFUeJzt3XlYVVX//vH7MCoiICagOaA54ixOqDnkgIqVaaVlzmYDmrPmk6ZY5lDmUNpgKdqgpablkErmmJiK4TynaSJooeDIuH9/+OX8QhygYB+E9+u6uB73Wmvv89nIc+LcrrW2xTAMQwAAAAAAAICJ7GxdAAAAAAAAAPIfQikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAJAtfH191atXL1uXkY7FYtH48eNz5NqnT5+WxWJRaGhojlz/XjZt2iSLxaJNmzZZ23r16iVfX99047Lr/sePHy+LxaK//vrrP18LAAAgDaEUAADIIDQ0VBaLRbt3775jf/PmzVWtWrX//Dpr1qzJsdAoNzl9+rR69+6tRx55RAUKFJCPj4+aNm2qcePG2bq0XGn58uVq166dHnroITk5OalEiRJ69tln9fPPP9u6NElSVFSUxo8fr8jISFuXAgDAA83B1gUAAIC84ejRo7Kzy9q/d61Zs0azZ8/OsWDqxo0bcnCw7a87J06cUL169VSwYEH16dNHvr6+On/+vPbs2aMpU6YoJCQky9ds2rSpbty4IScnpxyo2HYMw1CfPn0UGhqq2rVra+jQofLx8dH58+e1fPlytWzZUr/88osaNWpk0zqjoqIUEhIiX19f1apVy6a1AADwICOUAgAA2cLZ2dnWJUiSUlNTlZiYqAIFCqhAgQK2LkfTp0/X1atXFRkZqTJlyqTru3Dhwr+6pp2dXa64t+w2bdo0hYaGavDgwXr//fdlsVisfW+88Ya++OILm4eMAAAg+7B8DwAAZIvb95RKSkpSSEiIKlSooAIFCqho0aJq0qSJwsLCJN3aA2n27NmSbu19lPaV5tq1axo2bJhKlSolZ2dnVapUSe+9954Mw0j3uhaLRQMGDNBXX32lqlWrytnZWWvXrrX23T4L69y5c+rbt69KlCghZ2dnlS1bVq+88ooSExMlSbGxsRo+fLiqV68uV1dXubm5qV27dtq7d++/+r6cPHlSJUuWzBBISZKXl1eG72GHDh20fv161apVSwUKFJCfn5++++67dOPutKdUZly5ckWDBw+Wr6+vnJ2d5eXlpdatW2vPnj2ZOv+vv/7Ss88+Kzc3NxUtWlSDBg3SzZs3rf3NmjVTzZo173hupUqVFBgYeNdr37hxQ5MmTVLlypX13nvvpftZSNO9e3fVr1/fevz777/rmWeekaenp1xcXNSwYUOtXr063TlpS1FPnz6drv1O38O0ZamHDh1SixYt5OLioocfflhTp05Nd169evUkSb1797b+3IaGhmrcuHFydHTUxYsXM9Tev39/eXh4pPt+AQCQ3xFKAQCAu4qLi9Nff/2V4SspKem+544fP14hISFq0aKFPvzwQ73xxhsqXbq0NQB56aWX1Lp1a0nSF198Yf2Sbi3jeuKJJzR9+nS1bdtW77//vipVqqQRI0Zo6NChGV7r559/1pAhQ9SlSxfNnDkzw4bfaaKiolS/fn0tXrxYXbp00axZs9S9e3dt3rxZ169fl3Qr6FixYoU6dOig999/XyNGjND+/fvVrFkzRUVFZfl7WKZMGZ09ezbT+yEdP35cXbp0Ubt27TRp0iQ5ODjomWeesYZ5/8XLL7+sjz76SJ07d9acOXM0fPhwFSxYUIcPH87U+c8++6xu3rypSZMmqX379po1a5b69+9v7e/evbv27dunAwcOpDtv165dOnbsmF544YW7Xnvbtm2KjY3V888/L3t7+/vWEhMTo0aNGmndunV69dVXNXHiRN28eVNPPPGEli9fnqn7uZNLly6pbdu2qlmzpqZNm6bKlStr1KhR+vHHHyVJVapU0YQJEyTdCprSfm6bNm2q7t27Kzk5Wd988026ayYmJmrp0qXq3LlznpzhBgDAv2YAAADcZv78+Yake35VrVo13TllypQxevbsaT2uWbOmERQUdM/XCQ4ONu7068iKFSsMScbbb7+drv3pp582LBaLceLECWubJMPOzs44ePBghutIMsaNG2c97tGjh2FnZ2fs2rUrw9jU1FTDMAzj5s2bRkpKSrq+U6dOGc7OzsaECRPStUky5s+ff897PHDggFGwYEFDklGrVi1j0KBBxooVK4xr165lGFumTBlDkrFs2TJrW1xcnFG8eHGjdu3a1raNGzcakoyNGzda23r27GmUKVPmnvfv7u5uBAcH37PeOxk3bpwhyXjiiSfStb/66quGJGPv3r2GYRjG5cuXjQIFChijRo1KN+61114zChUqZFy9evWurzFz5kxDkrF8+fJM1TR48GBDkrF161Zr25UrV4yyZcsavr6+1r/DtJ/lU6dOpTv/Tt/DZs2aGZKMhQsXWtsSEhIMHx8fo3Pnzta2Xbt23fXvPiAgwGjQoEG6tu+++y7DawEAAMNgphQAALir2bNnKywsLMNXjRo17nuuh4eHDh48qOPHj2f5ddesWSN7e3u99tpr6dqHDRsmwzCss1bSNGvWTH5+fve8ZmpqqlasWKHHH39cdevWzdCftlzM2dnZumF7SkqK/v77b7m6uqpSpUqZXub2T1WrVlVkZKReeOEFnT59WjNnzlTHjh3l7e2tuXPnZhhfokQJPfXUU9ZjNzc39ejRQ7/99puio6Oz/Pr/5OHhoV9//fVfzfiSpODg4HTHAwcOlHTr70uS3N3d9eSTT2rRokXWZZYpKSn65ptv1LFjRxUqVOiu146Pj5ckFS5cOFO1rFmzRvXr11eTJk2sba6ururfv79Onz6tQ4cOZf7G/sHV1TXdjC4nJyfVr19fv//+e6bO79Gjh3799VedPHnS2vbVV1+pVKlSatas2b+qCQCAvIpQCgAA3FX9+vXVqlWrDF9FihS577kTJkzQ5cuXVbFiRVWvXl0jRozQvn37MvW6f/zxh0qUKJEhoKhSpYq1/5/Kli1732tevHhR8fHxqlat2j3Hpaamavr06apQoYKcnZ310EMPqVixYtq3b5/i4uIyVf/tKlasqC+++EJ//fWX9u3bp3feeUcODg7q37+/fvrpp3Rjy5cvn2E/pYoVK0pShn2Rsmrq1Kk6cOCASpUqpfr162v8+PGZDlskqUKFCumOH3nkEdnZ2aWrq0ePHjpz5oy2bt0qSfrpp58UExOj7t273/Pabm5ukm7te5UZf/zxhypVqpSh/W4/I5lVsmTJDN//IkWK6NKlS5k6v0uXLnJ2dtZXX30l6dYS2FWrVqlbt2533CcLAID8jFAKAADkiKZNm+rkyZOaN2+eqlWrps8++0x16tTRZ599lu2vVbBgwWy71jvvvKOhQ4eqadOm+vLLL7Vu3TqFhYWpatWqSk1N/U/Xtre3V/Xq1TV69Gjrvkdp4YUZnn32Wf3+++/64IMPVKJECb377ruqWrVqhplnmXWnkCUwMFDe3t768ssvJUlffvmlfHx81KpVq3teq3LlypKk/fv3/6taslKjdGsG153cbT8r47YN9u+mSJEi6tChg/XvdenSpUpISLjnfloAAORXhFIAACDHeHp6qnfv3lq0aJHOnj2rGjVqpHsa3t0CgzJlyigqKirDrJkjR45Y+7OqWLFicnNzy7AJ9+2WLl2qFi1a6PPPP1fXrl3Vpk0btWrVSpcvX87ya95L2hLC8+fPp2s/ceJEhgDk2LFjknTXDdyzonjx4nr11Ve1YsUKnTp1SkWLFtXEiRMzde7tSzFPnDih1NTUdHXZ29vr+eef19KlS3Xp0iWtWLFCzz333H03L2/SpImKFCmiRYsW3TUw+qcyZcro6NGjGdpv/xlJm9V3+9/fv51JJd395zZNjx49dOzYMe3atUtfffWVateurapVq/7r1wMAIK8ilAIAADni77//Tnfs6uqq8uXLKyEhwdqWtsfQ7YFB+/btlZKSog8//DBd+/Tp02WxWNSuXbss12NnZ6eOHTtq5cqV2r17d4b+tCDI3t4+Qyi0ZMkSnTt3LsuvKUlbt26949MK0/Zhun0JWlRUVLqnx8XHx2vhwoWqVauWfHx8/lUN0q2ZQbcvP/Ty8lKJEiXS/Z3cy+zZs9Mdf/DBB5KU4e+je/fuunTpkl566SVdvXo1U7OEXFxcNGrUKB0+fFijRo2648ykL7/8Ujt37pR062dk586dCg8Pt/Zfu3ZNn376qXx9fa17jD3yyCOSpC1btljHpaSk6NNPP83MLd/R3X5u07Rr104PPfSQpkyZos2bNzNLCgCAu3CwdQEAACBv8vPzU/PmzeXv7y9PT0/t3r1bS5cu1YABA6xj/P39JUmvvfaaAgMDZW9vr65du+rxxx9XixYt9MYbb+j06dOqWbOm1q9fr++//16DBw+2Bg1Z9c4772j9+vVq1qyZ+vfvrypVquj8+fNasmSJtm3bJg8PD3Xo0EETJkxQ79691ahRI+3fv19fffWVypUr969ec8qUKYqIiFCnTp2sG8Tv2bNHCxculKenpwYPHpxufMWKFdW3b1/t2rVL3t7emjdvnmJiYjR//vx/9fpprly5opIlS+rpp59WzZo15erqqp9++km7du3StGnTMnWNU6dO6YknnlDbtm0VHh6uL7/8Us8//7xq1qyZblzt2rVVrVo1LVmyRFWqVFGdOnUydf0RI0bo4MGDmjZtmjZu3Kinn35aPj4+io6O1ooVK7Rz505t375dkvT6669r0aJFateunV577TV5enpqwYIFOnXqlJYtW2bdrL5q1apq2LChRo8erdjYWHl6emrx4sVKTk7OwncvvUceeUQeHh76+OOPVbhwYRUqVEgNGjSw7m3m6Oiorl276sMPP5S9vb2ee+65f/1aAADkaTZ88h8AAMil5s+fb0gydu3adcf+Zs2aGVWrVk3XVqZMGaNnz57W47ffftuoX7++4eHhYRQsWNCoXLmyMXHiRCMxMdE6Jjk52Rg4cKBRrFgxw2KxGP/81eTKlSvGkCFDjBIlShiOjo5GhQoVjHfffddITU1N97qSjODg4DvWKckYN25curY//vjD6NGjh1GsWDHD2dnZKFeunBEcHGwkJCQYhmEYN2/eNIYNG2YUL17cKFiwoNG4cWMjPDzcaNasmdGsWTPrdU6dOmVIMubPn3+3b6NhGIbxyy+/GMHBwUa1atUMd3d3w9HR0ShdurTRq1cv4+TJkxm+h0FBQca6deuMGjVqGM7OzkblypWNJUuWpBu3ceNGQ5KxceNGa1vPnj2NMmXK3PX+ExISjBEjRhg1a9Y0ChcubBQqVMioWbOmMWfOnHvWbxiGMW7cOEOScejQIePpp582ChcubBQpUsQYMGCAcePGjTueM3XqVEOS8c4779z3+rdbunSp0aZNG8PT09NwcHAwihcvbnTp0sXYtGlTunEnT540nn76acPDw8MoUKCAUb9+fWPVqlUZrnfy5EmjVatWhrOzs+Ht7W3873//M8LCwjJ8D+/0c20Yd/7efv/994afn5/h4OBwx5+DnTt3GpKMNm3aZPn+AQDILyyGkcldGwEAAJCjfH19Va1aNa1atcrWpfxnM2fO1JAhQ3T69GmVLl3a1uWYbu/evapVq5YWLlx43ycPAgCQX7GnFAAAALKVYRj6/PPP1axZs3wZSEnS3Llz5erqqk6dOtm6FAAAci32lAIAAEC2uHbtmn744Qdt3LhR+/fv1/fff2/rkky3cuVKHTp0SJ9++qkGDBhg3RQdAABkRCgFAACAbHHx4kU9//zz8vDw0P/+9z898cQTti7JdAMHDlRMTIzat2+vkJAQW5cDAECuxp5SAAAAAAAAMB17SgEAAAAAAMB0hFIAAAAAAAAwHXtKZUJqaqqioqJUuHBhWSwWW5cDAAAAAACQaxmGoStXrqhEiRKys7v7fChCqUyIiopSqVKlbF0GAAAAAADAA+Ps2bMqWbLkXfsJpTKhcOHCkm59M93c3GxcDQAAAAAAQO4VHx+vUqVKWfOUuyGUyoS0JXtubm6EUgAAAAAAAJlwvy2Q2OgcAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkIpQAAAAAAAGA6QikAAAAAAACYjlAKAAAAAAAApiOUAgAAAAAAgOkcbF0AcDe+r6+2dQlAjjk9OcjWJQAAAACATTFTCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOpuGUuPHj5fFYkn3VblyZWv/zZs3FRwcrKJFi8rV1VWdO3dWTExMumucOXNGQUFBcnFxkZeXl0aMGKHk5OR0YzZt2qQ6derI2dlZ5cuXV2hoqBm3BwAAAAAAgLuw+UypqlWr6vz589avbdu2WfuGDBmilStXasmSJdq8ebOioqLUqVMna39KSoqCgoKUmJio7du3a8GCBQoNDdWbb75pHXPq1CkFBQWpRYsWioyM1ODBg9WvXz+tW7fO1PsEAAAAAADA/+dg8wIcHOTj45OhPS4uTp9//rm+/vprPfbYY5Kk+fPnq0qVKtqxY4caNmyo9evX69ChQ/rpp5/k7e2tWrVq6a233tKoUaM0fvx4OTk56eOPP1bZsmU1bdo0SVKVKlW0bds2TZ8+XYGBgabeKwAAAAAAAG6x+Uyp48ePq0SJEipXrpy6deumM2fOSJIiIiKUlJSkVq1aWcdWrlxZpUuXVnh4uCQpPDxc1atXl7e3t3VMYGCg4uPjdfDgQeuYf14jbUzaNQAAAAAAAGA+m86UatCggUJDQ1WpUiWdP39eISEhevTRR3XgwAFFR0fLyclJHh4e6c7x9vZWdHS0JCk6OjpdIJXWn9Z3rzHx8fG6ceOGChYsmKGuhIQEJSQkWI/j4+MlSUlJSUpKSvpvN41Mc7Y3bF0CkGN4LwEAAACQV2X2845NQ6l27dpZ/1yjRg01aNBAZcqU0bfffnvHsMgskyZNUkhISIb29evXy8XFxQYV5U9T69u6AiDnrFmzxtYlAAAAAECOuH79eqbG2XxPqX/y8PBQxYoVdeLECbVu3VqJiYm6fPlyutlSMTEx1j2ofHx8tHPnznTXSHs63z/H3P7EvpiYGLm5ud01+Bo9erSGDh1qPY6Pj1epUqXUpk0bubm5/ef7ROZUG89m9Mi7DoxnTzsAAAAAeVPairP7yVWh1NWrV3Xy5El1795d/v7+cnR01IYNG9S5c2dJ0tGjR3XmzBkFBARIkgICAjRx4kRduHBBXl5ekqSwsDC5ubnJz8/POub2GQlhYWHWa9yJs7OznJ2dM7Q7OjrK0dExW+4V95eQYrF1CUCO4b0EAAAAQF6V2c87Nt3ofPjw4dq8ebNOnz6t7du366mnnpK9vb2ee+45ubu7q2/fvho6dKg2btyoiIgI9e7dWwEBAWrYsKEkqU2bNvLz81P37t21d+9erVu3TmPGjFFwcLA1VHr55Zf1+++/a+TIkTpy5IjmzJmjb7/9VkOGDLHlrQMAAAAAAORrNp0p9eeff+q5557T33//rWLFiqlJkybasWOHihUrJkmaPn267Ozs1LlzZyUkJCgwMFBz5syxnm9vb69Vq1bplVdeUUBAgAoVKqSePXtqwoQJ1jFly5bV6tWrNWTIEM2cOVMlS5bUZ599psBAls4AAAAAAADYisUwDB5xdh/x8fFyd3dXXFwce0qZyPf11bYuAcgxpycH2boEAAAAAMgRmc1RbLp8DwAAAAAAAPkToRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMl2tCqcmTJ8tisWjw4MHWtps3byo4OFhFixaVq6urOnfurJiYmHTnnTlzRkFBQXJxcZGXl5dGjBih5OTkdGM2bdqkOnXqyNnZWeXLl1doaKgJdwQAAAAAAIC7yRWh1K5du/TJJ5+oRo0a6dqHDBmilStXasmSJdq8ebOioqLUqVMna39KSoqCgoKUmJio7du3a8GCBQoNDdWbb75pHXPq1CkFBQWpRYsWioyM1ODBg9WvXz+tW7fOtPsDAAAAAABAejYPpa5evapu3bpp7ty5KlKkiLU9Li5On3/+ud5//3099thj8vf31/z587V9+3bt2LFDkrR+/XodOnRIX375pWrVqqV27drprbfe0uzZs5WYmChJ+vjjj1W2bFlNmzZNVapU0YABA/T0009r+vTpNrlfAAAAAAAASA62LiA4OFhBQUFq1aqV3n77bWt7RESEkpKS1KpVK2tb5cqVVbp0aYWHh6thw4YKDw9X9erV5e3tbR0TGBioV155RQcPHlTt2rUVHh6e7hppY/65TPB2CQkJSkhIsB7Hx8dLkpKSkpSUlPRfbxmZ5Gxv2LoEIMfwXgIAAAAgr8rs5x2bhlKLFy/Wnj17tGvXrgx90dHRcnJykoeHR7p2b29vRUdHW8f8M5BK60/ru9eY+Ph43bhxQwULFszw2pMmTVJISEiG9vXr18vFxSXzN4j/ZGp9W1cA5Jw1a9bYugQAAAAAyBHXr1/P1DibhVJnz57VoEGDFBYWpgIFCtiqjDsaPXq0hg4daj2Oj49XqVKl1KZNG7m5udmwsvyl2nj2/ULedWB8oK1LAAAAAIAckbbi7H5sFkpFRETowoULqlOnjrUtJSVFW7Zs0Ycffqh169YpMTFRly9fTjdbKiYmRj4+PpIkHx8f7dy5M911057O988xtz+xLyYmRm5ubnecJSVJzs7OcnZ2ztDu6OgoR0fHrN8s/pWEFIutSwByDO8lAAAAAPKqzH7esdlG5y1bttT+/fsVGRlp/apbt666detm/bOjo6M2bNhgPefo0aM6c+aMAgICJEkBAQHav3+/Lly4YB0TFhYmNzc3+fn5Wcf88xppY9KuAQAAAAAAAPPZbKZU4cKFVa1atXRthQoVUtGiRa3tffv21dChQ+Xp6Sk3NzcNHDhQAQEBatiwoSSpTZs28vPzU/fu3TV16lRFR0drzJgxCg4Ots50evnll/Xhhx9q5MiR6tOnj37++Wd9++23Wr16tbk3DAAAAAAAACubP33vXqZPny47Ozt17txZCQkJCgwM1Jw5c6z99vb2WrVqlV555RUFBASoUKFC6tmzpyZMmGAdU7ZsWa1evVpDhgzRzJkzVbJkSX322WcKDGQ/FwAAAAAAAFuxGIZh2LqI3C4+Pl7u7u6Ki4tjo3MT+b7ObDbkXacnB9m6BAAAAADIEZnNUWy2pxQAAAAAAADyL0IpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKbLcii1Z88e7d+/33r8/fffq2PHjvrf//6nxMTEbC0OAAAAAAAAeVOWQ6mXXnpJx44dkyT9/vvv6tq1q1xcXLRkyRKNHDky2wsEAAAAAABA3pPlUOrYsWOqVauWJGnJkiVq2rSpvv76a4WGhmrZsmXZXR8AAAAAAADyoCyHUoZhKDU1VZL0008/qX379pKkUqVK6a+//sre6gAAAAAAAJAnZTmUqlu3rt5++2198cUX2rx5s4KCgiRJp06dkre3d7YXCAAAAAAAgLwny6HUjBkztGfPHg0YMEBvvPGGypcvL0launSpGjVqlO0FAgAAAAAAIO9xyOoJNWrUSPf0vTTvvvuu7O3ts6UoAAAAAAAA5G1ZDqXupkCBAtl1KQAAAAAAAORxmQqlihQpIovFkqkLxsbG/qeCAAAAAAAAkPdlKpSaMWNGDpcBAAAAAACA/CRToVTPnj1zug4AAAAAAADkI5kKpeLj4zN9QTc3t39dDAAAAAAAAPKHTIVSHh4e991TyjAMWSwWpaSkZEthAAAAAAAAyLsyFUpt3Lgxp+sAAAAAAABAPpKpUKpZs2Y5XQcAAAAAAADykUyFUvv27VO1atVkZ2enffv23XNsjRo1sqUwAAAAAAAA5F2ZCqVq1aql6OhoeXl5qVatWrJYLDIMI8M49pQCAAAAAABAZmQqlDp16pSKFStm/TMAAAAAAADwX2QqlCpTpswd/wwAAAAAAAD8G5kKpW539OhRffDBBzp8+LAkqUqVKho4cKAqVaqUrcUBAAAAAAAgb7LL6gnLli1TtWrVFBERoZo1a6pmzZras2ePqlWrpmXLluVEjQAAAAAAAMhjsjxTauTIkRo9erQmTJiQrn3cuHEaOXKkOnfunG3FAQAAAAAAIG/K8kyp8+fPq0ePHhnaX3jhBZ0/fz5bigIAAAAAAEDeluVQqnnz5tq6dWuG9m3btunRRx/NlqIAAAAAAACQt2V5+d4TTzyhUaNGKSIiQg0bNpQk7dixQ0uWLFFISIh++OGHdGMBAAAAAACA21kMwzCycoKdXeYmV1ksFqWkpPyronKb+Ph4ubu7Ky4uTm5ubrYuJ9/wfX21rUsAcszpyUG2LgEAAAAAckRmc5Qsz5RKTU39T4UBAAAAAAAAWd5TCgAAAAAAAPivMh1KhYeHa9WqVenaFi5cqLJly8rLy0v9+/dXQkJCthcIAAAAAACAvCfTodSECRN08OBB6/H+/fvVt29ftWrVSq+//rpWrlypSZMm5UiRAAAAAAAAyFsyHUpFRkaqZcuW1uPFixerQYMGmjt3roYOHapZs2bp22+/zZEiAQAAAAAAkLdkOpS6dOmSvL29rcebN29Wu3btrMf16tXT2bNns7c6AAAAAAAA5EmZDqW8vb116tQpSVJiYqL27Nmjhg0bWvuvXLkiR0fH7K8QAAAAAAAAeU6mQ6n27dvr9ddf19atWzV69Gi5uLjo0Ucftfbv27dPjzzySI4UCQAAAAAAgLzFIbMD33rrLXXq1EnNmjWTq6urFixYICcnJ2v/vHnz1KZNmxwpEgAAAAAAAHlLpkOphx56SFu2bFFcXJxcXV1lb2+frn/JkiVydXXN9gIBAAAAAACQ92Q6lErj7u5+x3ZPT8//XAwAAAAAAADyh0zvKQUAAAAAAABkF5uGUh999JFq1KghNzc3ubm5KSAgQD/++KO1/+bNmwoODlbRokXl6uqqzp07KyYmJt01zpw5o6CgILm4uMjLy0sjRoxQcnJyujGbNm1SnTp15OzsrPLlyys0NNSM2wMAAAAAAMBd2DSUKlmypCZPnqyIiAjt3r1bjz32mJ588kkdPHhQkjRkyBCtXLlSS5Ys0ebNmxUVFaVOnTpZz09JSVFQUJASExO1fft2LViwQKGhoXrzzTetY06dOqWgoCC1aNFCkZGRGjx4sPr166d169aZfr8AAAAAAAC4xWIYhnG/QXXq1NGGDRtUpEgRTZgwQcOHD5eLi0uOFOTp6al3331XTz/9tIoVK6avv/5aTz/9tCTpyJEjqlKlisLDw9WwYUP9+OOP6tChg6KiouTt7S1J+vjjjzVq1ChdvHhRTk5OGjVqlFavXq0DBw5YX6Nr1666fPmy1q5dm6ma4uPj5e7urri4OLm5uWX/TeOOfF9fbesSgBxzenKQrUsAAAAAgByR2RwlUzOlDh8+rGvXrkmSQkJCdPXq1eyp8h9SUlK0ePFiXbt2TQEBAYqIiFBSUpJatWplHVO5cmWVLl1a4eHhkqTw8HBVr17dGkhJUmBgoOLj462zrcLDw9NdI21M2jUAAAAAAABgvkw9fa9WrVrq3bu3mjRpIsMw9N5778nV1fWOY/+5dC4z9u/fr4CAAN28eVOurq5avny5/Pz8FBkZKScnJ3l4eKQb7+3trejoaElSdHR0ukAqrT+t715j4uPjdePGDRUsWDBDTQkJCUpISLAex8fHS5KSkpKUlJSUpfvDv+dsf99JfMADi/cSAAAAAHlVZj/vZCqUCg0N1bhx47Rq1SpZLBb9+OOPcnDIeKrFYslyKFWpUiVFRkYqLi5OS5cuVc+ePbV58+YsXSO7TZo0SSEhIRna169fn2PLFpHR1Pq2rgDIOWvWrLF1CQAAAACQI65fv56pcZkKpSpVqqTFixdLkuzs7LRhwwZ5eXn9++r+wcnJSeXLl5ck+fv7a9euXZo5c6a6dOmixMREXb58Od1sqZiYGPn4+EiSfHx8tHPnznTXS3s63z/H3P7EvpiYGLm5ud1xlpQkjR49WkOHDrUex8fHq1SpUmrTpg17Spmo2ng2o0fedWB8oK1LAAAAAIAckbbi7H4yFUr9U2pqapaLyer1ExIS5O/vL0dHR23YsEGdO3eWJB09elRnzpxRQECAJCkgIEATJ07UhQsXrCFZWFiY3Nzc5OfnZx1z+4yEsLAw6zXuxNnZWc7OzhnaHR0d5ejomC33iftLSLHYugQgx/BeAgAAACCvyuznnSyHUpJ08uRJzZgxQ4cPH5Yk+fn5adCgQXrkkUeydJ3Ro0erXbt2Kl26tK5cuaKvv/5amzZt0rp16+Tu7q6+fftq6NCh8vT0lJubmwYOHKiAgAA1bNhQktSmTRv5+fmpe/fumjp1qqKjozVmzBgFBwdbQ6WXX35ZH374oUaOHKk+ffro559/1rfffqvVq3myGwAAAAAAgK1kOZRat26dnnjiCdWqVUuNGzeWJP3yyy+qWrWqVq5cqdatW2f6WhcuXFCPHj10/vx5ubu7q0aNGlq3bp31GtOnT5ednZ06d+6shIQEBQYGas6cOdbz7e3ttWrVKr3yyisKCAhQoUKF1LNnT02YMME6pmzZslq9erWGDBmimTNnqmTJkvrss88UGMjSGQAAAAAAAFuxGIaRpUec1a5dW4GBgZo8eXK69tdff13r16/Xnj17srXA3CA+Pl7u7u6Ki4tjTykT+b7ObDbkXacnB9m6BAAAAADIEZnNUeyyeuHDhw+rb9++Gdr79OmjQ4cOZfVyAAAAAAAAyIeyHEoVK1ZMkZGRGdojIyOz7Yl8AAAAAAAAyNuyvKfUiy++qP79++v3339Xo0aNJN3aU2rKlCkaOnRothcIAAAAAACAvCfLodTYsWNVuHBhTZs2TaNHj5YklShRQuPHj9drr72W7QUCAAAAAAAg78lyKGWxWDRkyBANGTJEV65ckSQVLlw42wsDAAAAAABA3pXlUOqfCKMAAAAAAADwb2R5o3MAAAAAAADgvyKUAgAAAAAAgOkIpQAAAAAAAGC6LIVSSUlJatmypY4fP55T9QAAAAAAACAfyFIo5ejoqH379uVULQAAAAAAAMgnsrx874UXXtDnn3+eE7UAAAAAAAAgn3DI6gnJycmaN2+efvrpJ/n7+6tQoULp+t9///1sKw4AAAAAAAB5U5ZDqQMHDqhOnTqSpGPHjqXrs1gs2VMVAAAAAAAA8rQsh1IbN27MiToAAAAAAACQj2R5T6k0J06c0Lp163Tjxg1JkmEY2VYUAAAAAAAA8rYsh1J///23WrZsqYoVK6p9+/Y6f/68JKlv374aNmxYthcIAAAAAACAvCfLodSQIUPk6OioM2fOyMXFxdrepUsXrV27NluLAwAAAAAAQN6U5T2l1q9fr3Xr1qlkyZLp2itUqKA//vgj2woDAAAAAABA3pXlmVLXrl1LN0MqTWxsrJydnbOlKAAAAAAAAORtWQ6lHn30US1cuNB6bLFYlJqaqqlTp6pFixbZWhwAAAAAAADypiwv35s6dapatmyp3bt3KzExUSNHjtTBgwcVGxurX375JSdqBAAAAAAAQB6T5ZlS1apV07Fjx9SkSRM9+eSTunbtmjp16qTffvtNjzzySE7UCAAAAAAAgDwmyzOlJMnd3V1vvPFGdtcCAAAAAACAfOJfhVKXLl3S559/rsOHD0uS/Pz81Lt3b3l6emZrcQAAAAAAAMibsrx8b8uWLfL19dWsWbN06dIlXbp0SbNmzVLZsmW1ZcuWnKgRAAAAAAAAeUyWZ0oFBwerS5cu+uijj2Rvby9JSklJ0auvvqrg4GDt378/24sEAAAAAABA3pLlmVInTpzQsGHDrIGUJNnb22vo0KE6ceJEthYHAAAAAACAvCnLoVSdOnWse0n90+HDh1WzZs1sKQoAAAAAAAB5W6aW7+3bt8/659dee02DBg3SiRMn1LBhQ0nSjh07NHv2bE2ePDlnqgQAAAAAAECeYjEMw7jfIDs7O1ksFt1vqMViUUpKSrYVl1vEx8fL3d1dcXFxcnNzs3U5+Ybv66ttXQKQY05PDrJ1CQAAAACQIzKbo2RqptSpU6eyrTAAAAAAAAAgU6FUmTJlcroOAAAAAAAA5COZCqVuFxUVpW3btunChQtKTU1N1/faa69lS2EAAAAAAADIu7IcSoWGhuqll16Sk5OTihYtKovFYu2zWCyEUgAAAAAAALivLIdSY8eO1ZtvvqnRo0fLzs4uJ2oCAAAAAABAHpflVOn69evq2rUrgRQAAAAAAAD+tSwnS3379tWSJUtyohYAAAAAAADkE1levjdp0iR16NBBa9euVfXq1eXo6Jiu//3338+24gAAAAAAAJA3/atQat26dapUqZIkZdjoHAAAAAAAALifLIdS06ZN07x589SrV68cKAcAAAAAAAD5QZb3lHJ2dlbjxo1zohYAAAAAAADkE1kOpQYNGqQPPvggJ2oBAAAAAABAPpHl5Xs7d+7Uzz//rFWrVqlq1aoZNjr/7rvvsq04AAAAAAAA5E1ZDqU8PDzUqVOnnKgFAAAAAAAA+USWQ6n58+fnRB0AAAAAAADIR7K8pxQAAAAAAADwX2V5plTZsmVlsVju2v/777//p4IAAAAAAACQ92U5lBo8eHC646SkJP32229au3atRowYkV11AQAAAAAAIA/Lcig1aNCgO7bPnj1bu3fv/s8FAQAAAAAAIO/Ltj2l2rVrp2XLlmXX5QAAAAAAAJCHZVsotXTpUnl6embX5QAAAAAAAJCHZXn5Xu3atdNtdG4YhqKjo3Xx4kXNmTMnW4sDAAAAAABA3pTlUKpjx47pju3s7FSsWDE1b95clStXzq66AAAAAAAAkIdlOZQaN25cTtQBAAAAAACAfCTb9pQCAAAAAAAAMivTM6Xs7OzS7SV1JxaLRcnJyf+5KAAAAAAAAORtmQ6lli9ffte+8PBwzZo1S6mpqdlSFAAAAAAAAPK2TIdSTz75ZIa2o0eP6vXXX9fKlSvVrVs3TZgwIVuLAwAAAAAAQN70r/aUioqK0osvvqjq1asrOTlZkZGRWrBggcqUKZPd9QEAAAAAACAPylIoFRcXp1GjRql8+fI6ePCgNmzYoJUrV6patWo5VR8AAAAAAADyoEwv35s6daqmTJkiHx8fLVq06I7L+QAAAAAAAIDMsBiGYWRmoJ2dnQoWLKhWrVrJ3t7+ruO+++67bCsut4iPj5e7u7vi4uLk5uZm63LyDd/XV9u6BCDHnJ4cZOsSAAAAACBHZDZHyfRMqR49eshisWRLcQAAAAAAAMjfMh1KhYaG5mAZAAAAAAAAyE/+1dP3ssukSZNUr149FS5cWF5eXurYsaOOHj2abszNmzcVHBysokWLytXVVZ07d1ZMTEy6MWfOnFFQUJBcXFzk5eWlESNGKDk5Od2YTZs2qU6dOnJ2dlb58uUJ2QAAAAAAAGzIpqHU5s2bFRwcrB07digsLExJSUlq06aNrl27Zh0zZMgQrVy5UkuWLNHmzZsVFRWlTp06WftTUlIUFBSkxMREbd++XQsWLFBoaKjefPNN65hTp04pKChILVq0UGRkpAYPHqx+/fpp3bp1pt4vAAAAAAAAbsn0RudmuHjxory8vLR582Y1bdpUcXFxKlasmL7++ms9/fTTkqQjR46oSpUqCg8PV8OGDfXjjz+qQ4cOioqKkre3tyTp448/1qhRo3Tx4kU5OTlp1KhRWr16tQ4cOGB9ra5du+ry5ctau3btfetio3PbYKNz5GVsdA4AAAAgr8psjmLTmVK3i4uLkyR5enpKkiIiIpSUlKRWrVpZx1SuXFmlS5dWeHi4JCk8PFzVq1e3BlKSFBgYqPj4eB08eNA65p/XSBuTdg0AAAAAAACYK9Mbnee01NRUDR48WI0bN1a1atUkSdHR0XJycpKHh0e6sd7e3oqOjraO+Wcgldaf1nevMfHx8bpx44YKFiyYri8hIUEJCQnW4/j4eElSUlKSkpKS/uOdIrOc7XPNJD4g2/FeAgAAACCvyuznnVwTSgUHB+vAgQPatm2brUvRpEmTFBISkqF9/fr1cnFxsUFF+dPU+rauAMg5a9assXUJAAAAAJAjrl+/nqlxuSKUGjBggFatWqUtW7aoZMmS1nYfHx8lJibq8uXL6WZLxcTEyMfHxzpm586d6a6X9nS+f465/Yl9MTExcnNzyzBLSpJGjx6toUOHWo/j4+NVqlQptWnThj2lTFRtPBvRI+86MD7Q1iUAAAAAQI5IW3F2PzYNpQzD0MCBA7V8+XJt2rRJZcuWTdfv7+8vR0dHbdiwQZ07d5YkHT16VGfOnFFAQIAkKSAgQBMnTtSFCxfk5eUlSQoLC5Obm5v8/PysY26flRAWFma9xu2cnZ3l7Oycod3R0VGOjo7/7aaRaQkpFluXAOQY3ksAAAAA5FWZ/bxj01AqODhYX3/9tb7//nsVLlzYugeUu7u7ChYsKHd3d/Xt21dDhw6Vp6en3NzcNHDgQAUEBKhhw4aSpDZt2sjPz0/du3fX1KlTFR0drTFjxig4ONgaLL388sv68MMPNXLkSPXp00c///yzvv32W61ezdPdAAAAAAAAbMGmT9/76KOPFBcXp+bNm6t48eLWr2+++cY6Zvr06erQoYM6d+6spk2bysfHR9999521397eXqtWrZK9vb0CAgL0wgsvqEePHpowYYJ1TNmyZbV69WqFhYWpZs2amjZtmj777DMFBrJ8BgAAAAAAwBYshmHwiLP7iI+Pl7u7u+Li4thTykS+rzOTDXnX6clBti4BAAAAAHJEZnMUm86UAgAAAAAAQP5EKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADT2TSU2rJlix5//HGVKFFCFotFK1asSNdvGIbefPNNFS9eXAULFlSrVq10/PjxdGNiY2PVrVs3ubm5ycPDQ3379tXVq1fTjdm3b58effRRFShQQKVKldLUqVNz+tYAAAAAAABwDzYNpa5du6aaNWtq9uzZd+yfOnWqZs2apY8//li//vqrChUqpMDAQN28edM6plu3bjp48KDCwsK0atUqbdmyRf3797f2x8fHq02bNipTpowiIiL07rvvavz48fr0009z/P4AAAAAAABwZxbDMAxbFyFJFotFy5cvV8eOHSXdmiVVokQJDRs2TMOHD5ckxcXFydvbW6GhoeratasOHz4sPz8/7dq1S3Xr1pUkrV27Vu3bt9eff/6pEiVK6KOPPtIbb7yh6OhoOTk5SZJef/11rVixQkeOHMlUbfHx8XJ3d1dcXJzc3Nyy/+ZxR76vr7Z1CUCOOT05yNYlAAAAAECOyGyO4mBiTVly6tQpRUdHq1WrVtY2d3d3NWjQQOHh4eratavCw8Pl4eFhDaQkqVWrVrKzs9Ovv/6qp556SuHh4WratKk1kJKkwMBATZkyRZcuXVKRIkUyvHZCQoISEhKsx/Hx8ZKkpKQkJSUl5cTt4g6c7XNFXgrkCN5LAAAAAORVmf28k2tDqejoaEmSt7d3unZvb29rX3R0tLy8vNL1Ozg4yNPTM92YsmXLZrhGWt+dQqlJkyYpJCQkQ/v69evl4uLyL+8IWTW1vq0rAHLOmjVrbF0CAAAAAOSI69evZ2pcrg2lbGn06NEaOnSo9Tg+Pl6lSpVSmzZtWL5nomrj19m6BCDHHBgfaOsSAAAAACBHpK04u59cG0r5+PhIkmJiYlS8eHFre0xMjGrVqmUdc+HChXTnJScnKzY21nq+j4+PYmJi0o1JO04bcztnZ2c5OztnaHd0dJSjo+O/uyFkWUKKxdYlADmG9xIAAAAAeVVmP+/Y9Ol791K2bFn5+Phow4YN1rb4+Hj9+uuvCggIkCQFBATo8uXLioiIsI75+eeflZqaqgYNGljHbNmyJd16xrCwMFWqVOmOS/cAAAAAAACQ82waSl29elWRkZGKjIyUdGtz88jISJ05c0YWi0WDBw/W22+/rR9++EH79+9Xjx49VKJECesT+qpUqaK2bdvqxRdf1M6dO/XLL79owIAB6tq1q0qUKCFJev755+Xk5KS+ffvq4MGD+uabbzRz5sx0y/MAAAAAAABgLpsu39u9e7datGhhPU4Linr27KnQ0FCNHDlS165dU//+/XX58mU1adJEa9euVYECBaznfPXVVxowYIBatmwpOzs7de7cWbNmzbL2u7u7a/369QoODpa/v78eeughvfnmm+rfv795NwoAAAAAAIB0LIZhGLYuIreLj4+Xu7u74uLi2OjcRL6vr7Z1CUCOOT05yNYlAAAAAECOyGyOkmv3lAIAAAAAAEDeRSgFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABM52DrAgAAQN7i+/pqW5cA5JjTk4NsXQIAAHkGM6UAAAAAAABgOmZKAQAAAAAkMdsVeRuzXXMfZkoBAAAAAADAdIRSAAAAAAAAMB2hFAAAAAAAAExHKAUAAAAAAADTEUoBAAAAAADAdPkqlJo9e7Z8fX1VoEABNWjQQDt37rR1SQAAAAAAAPlSvgmlvvnmGw0dOlTjxo3Tnj17VLNmTQUGBurChQu2Lg0AAAAAACDfyTeh1Pvvv68XX3xRvXv3lp+fnz7++GO5uLho3rx5ti4NAAAAAAAg38kXoVRiYqIiIiLUqlUra5udnZ1atWql8PBwG1YGAAAAAACQPznYugAz/PXXX0pJSZG3t3e6dm9vbx05ciTD+ISEBCUkJFiP4+LiJEmxsbFKSkrK2WJh5ZB8zdYlADnm77//tnUJQI7h/Rt5Ge/fyOt4D0dexnu4ea5cuSJJMgzjnuPyRSiVVZMmTVJISEiG9rJly9qgGgB50UPTbF0BAODf4P0bAB5cvIeb78qVK3J3d79rf74IpR566CHZ29srJiYmXXtMTIx8fHwyjB89erSGDh1qPU5NTVVsbKyKFi0qi8WS4/UCZouPj1epUqV09uxZubm52bocAEAm8f4NAA8u3sORlxmGoStXrqhEiRL3HJcvQiknJyf5+/trw4YN6tixo6RbQdOGDRs0YMCADOOdnZ3l7Oycrs3Dw8OESgHbcnNz4z+IAPAA4v0bAB5cvIcjr7rXDKk0+SKUkqShQ4eqZ8+eqlu3rurXr68ZM2bo2rVr6t27t61LAwAAAAAAyHfyTSjVpUsXXbx4UW+++aaio6NVq1YtrV27NsPm5wAAAAAAAMh5+SaUkqQBAwbccbkekN85Oztr3LhxGZatAgByN96/AeDBxXs4IFmM+z2fDwAAAAAAAMhmdrYuAAAAAAAAAPkPoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHQOti4AAAAAAABkTXJysgzDUEJCghwcHFSgQAFblwRkGTOlAAAAAAB4gBw9elR9+/ZVo0aN5OfnpyZNmmjZsmW6evWqrUsDsoRQCgAAAACAB8T+/fsVEBAgOzs7devWTX369JGbm5ueffZZvfXWWzp//rytSwQyjeV7AAAAAAA8AKKjo9W1a1f1799fkydPtrbHxsZqxowZmjhxojw9PTVq1CilpqbKzo55KMjdCKUAAAAAAHgA7N27V87Oznr11VclyRo8eXp6auzYsYqPj9f48eP1xBNPqEqVKjauFrg/YlMAAAAAAB4A+/bt099//63SpUtLkiwWi7XPwcFB3bp1k5OTk3bv3m2rEoEsIZQCAADIJwzDsHUJAID/4OGHH9a5c+e0a9cuSelDKYvFonr16snZ2Vnnzp2zVYlAlhBKAQAA5EE3btzQ1atXFRERod9//11S+g8vAIDc7+LFi9q+fbt++uknJSQkqFKlSipZsqRmz55tDZ5SU1MlScnJyfrzzz/l6+urOnXq2LJsINMIpQAAAPKY48eP6+WXX1bDhg1Vr1491a9fX88884wuXrxo69IAAJl08OBBdezYUW+99Za2bt2q1NRU+fv76+mnn9bChQs1bdo0nTp1yrqZuYODg+bOnavLly/Lz8/PxtUDmWMxmMcNAACQZ+zbt0/t27fX448/rgYNGqhq1ar64Ycf9Mknn8jX11dffPGFKlWqJMMwmDkFALnUoUOH1KRJE7366qvq27evypYtm66/f//++uqrr1SpUiX17t1bV69e1R9//KHFixdr48aNql27to0qB7KGUAoAACCP2Lt3rxo1aqTXXntNb731lhwcbj1o+caNG/r111/Vu3dvlSxZUlu3brVxpQCAu7l06ZL16XmffvqptT01NVWGYcje3l6SNGfOHK1evVo7duxQ6dKlVaNGDY0aNYpZUnigEEoBAADkAefOnVOpUqU0bNgwvfvuu9bHhKfNiEpJSdHKlSvVpUsXvf322xoxYoStSwYA3MGBAwfUtWtXzZ49W82aNcvQn5SUJEdHR+vxxYsXVaxYMSUmJsrJycnMUoH/jD2lAAAA8oDY2FhVqFBB27dv17Vr12RnZ6fU1FTrEj17e3s99thj8vPz06lTp2xcLQDgbo4dO6Y//vhDZcqUkZTxyamOjo5KTEzUqlWrlJiYqKJFi1rbgQcNoRQAAEAeUL16dS1dulSXL19W06ZN0wVTadzc3FS4cGFduXLFhpUCAO7Fw8ND165ds/4Dwp0WN82bN09r1qyRk5OTdaNz9gnEg4hQCgAAII+oXr26Fi1apJs3b6pZs2a6fv267OzslJKSIknWpzS1a9dO0p0/6AAAzHX8+HGNGjXKeuzv76+6detq8ODB1vftxMTEdOccPXpUrq6uSkpKMrtcIFsRSgEAAOQhNWrU0KJFi3Tjxg01bdpUV69etW6KO3fuXF25ckVNmzaVxL+qA0BucOLECb377rsaNGiQJMnd3V09evRQbGys+vbtqz/++MO6V9SlS5f0xhtv6Ntvv1W/fv1YsocHHhudAwAAPGDOnj2rmzdvqkKFCncds3//fj333HNydnZWRESEJk2apHfeeUfbtm1TzZo1TawWAHA/P/zwg7p27apevXppzpw5kqS3335bn376qS5fvqxevXrp77//1rVr1/Trr79qzZo1ql27to2rBv47QikAAIAHhGEYSk1NVfXq1eXs7KxFixapcuXKdx2/f/9+vfDCC9q/f7+cnZ21bds2+fv7m1gxAOBOUlJSrLNY06xYsULPPfecevbsqY8//liSFBYWph9++EG7d++Ws7Ozmjdvrm7dut3zHyWABwmhFJDPpD0iHADw4El73PfZs2fVuHFjPfLII5ozZ46qVKly13PSZkmNHTuWGVIAkAscP35cM2bMULNmzdSgQQPrU/Yk6bvvvtMLL7ygbt26ae7cudb2q1evytXVVYZhsPQaeQqhFJCHXblyRfHx8dq1a5fKlSunEiVK6KGHHiKYAoAH0IEDB/Tpp59qzJgx8vLyUlRUlPz9/VW5cuU7BlMJCQn69ttv1bhxYz388MNydna2UeUAgDRXrlxRhw4dtHXrVklS8+bNFR8fr/79+6tOnTqqU6eOtmzZoieffFK9e/fW+++/b32Sqp2dHaEU8hw+lQJ51NGjR9W7d2+1bt1anTp1UuPGjdW8eXPt2rXL+h80AMCDYe/evapRo4a8vLzk5eWl5ORklShRQhERETpy5IheffVVHT582Do+ISFBQ4cOVc+ePWVnZ0cgBQC5hIuLi3r06KG2bduqVq1aGjhwoBo3bqyFCxeqUaNGatu2rZYtW6Z+/fpp1qxZGj16tAzDsP6DMoEU8hpmSgF50N69e9WmTRt16dJFLVq00GOPPaavvvpKX3zxhQ4cOKAffvhBLVq0YMYUADwAIiMj1bhxYw0ePFgTJ060tl+7dk2FChXKMGPqkUce0fDhwzV//nxt3rxZderUsWH1AIA0ab97Jycn66uvvtL8+fNVrFgxffXVV3JyclJYWJgOHDighQsXyt7eXnv27JEkXbhwQQ899JCNqwdyBqEUkMfs3btXjRs31sCBAzVp0qR0fRERERo1apQiIiK0e/duPfLIIzaqEgCQGQcPHlS9evU0evRojR071tr+ySefyN3dXR07dlSBAgWswZSfn598fHy0fPlybdu2jUAKAHKZ5ORkOTg4KDk5WYsXL9YHH3wgb29vhYaGytPT0zru+PHj2rlzp+rUqXPPfQOBBx2hFJCHXLhwQZUrV1bLli21ZMkS67/G/PPpHj///LN69Oihzp07a+bMmaxLB4Bc6tKlS2rcuLEcHR21efNmeXh4SJKmTJmiMWPGaNOmTWrcuLH1A8758+dVsWJFXbt2Tb/99hubmgNALnDkyBFt2rRJNWrUUKNGjdL1paSkaNGiRZozZ46KFi2qL7/8Uu7u7vx+jnyFdTtAHhIbG6uWLVvql19+0bZt26ybItrb21v3kHrsscdUt25d7du3TxLr0gEgtypSpIieeOIJFShQQFOmTJEkzZgxQ++++67WrFmjxo0bS5IcHBxkGIaKFy+ukydP6vjx4wRSAJALXL9+Xd27d9fcuXM1aNAgvfjii9q2bZv193J7e3t17dpVr7zyiv7++2/16tVLly5d4vdz5CuEUkAeUrlyZU2cOFGPPfaYOnbsqK1bt1qDqX9ycXFRoUKFbFQlAOBezp8/r+3bt0uSJk+erLZt2+qnn35S8+bNNX78eH3//fdq3bp1ugdWfPnll9q3b5+8vLxYmg0AuYSLi4sqV64sLy8vff/994qNjVVISIiaNm2qLVu26Ny5c3JwcFDXrl312muv6dChQ3r11Vd5IBHyFQdbFwAge6RN861YsaJ135GnnnpKK1asUJMmTaxL+C5cuKCrV6+qdevW6c4DANiWYRi6fv26evXqJYvFov/9739q2rSpQkJC5ODgoI8//lht27aVn59fuvPGjh2rWbNmKSIiwkaVAwBul/Y79qRJk/TMM88oKipKCxYs0MWLFzV79mwFBgaqTp066tWrl7p166auXbvK1dVVVatW5Xdz5CvMlALyiH/+x6tSpUoaO3as2rZta50xlban1PTp03Xy5El17Ngxw3kAANuxWCwqVKiQXn75ZV29elWzZ8/W5s2bJd0Knvr376/jx49rypQpio6OlsVi0Ztvvqn33ntPGzZsUPny5W18BwCANGm/YxcuXFi+vr5avny5XF1dVbZsWV2/fl1eXl5q1KiRhg8frtq1a2vixInq0KGDypYta+PKAXMxUwp4QJ08eVJxcXF3fbJSWjAl3ZoxFRYWps2bN2vWrFn65ZdfVKZMGTPLBQDcw+7duzVy5EiFhYXpqaeeUsGCBTV+/HjNnj1bktSsWTONGzdOqampWrlypVxcXBQbG6tPPvlE27Ztk7+/v43vAABwJ+7u7urWrZteeOEF9evXT++8845Wr16tNWvWqFatWurXr58WL16sZ5991talAjbB0/eAB9Tzzz+vJUuWaPv27apXr95dxx09elTvvPOOvvjiC1ksFu3cuZMPLwCQi+zdu1dNmjRRr1699MEHH1jb16xZowkTJqh06dIKDg5Ws2bNJEkhISH68MMPdfPmTW3evPmu/zgBALA9wzBkGIZefPFFrVq1SoUKFdLixYtVv35965Oy0/4XyI8IpYAHzI0bN1SwYEElJiaqa9euCg8P14oVK9SgQYO7nnPkyBF98skn6tevn6pWrWpitQCAe9m7d68CAgI0ZMgQTZw4MUP/jz/+qJCQkAzB1IwZM9S+fXtVrFjR7JIBALfJzB6ts2fP1sCBA7Vr1y75+/uzryvwf4hjgQfI3r179corr+js2bNycnLSokWLVK9ePXXs2FG//vprhvE3b97UxIkTZbFY9O677xJIAUAucvLkSTVr1ky9e/dOF0jNmzdPCxYskCS1a9dO48eP15kzZ/TJJ58oLCxMkjR48GACKQDIBU6cOKE+ffro8uXLd+xPmwMSHBysevXqKTQ0VBL7ugJpCKWAB8TevXtVp04dPfzwwypVqpQkydnZWUuWLFHdunUzBFOJiYkaNWqUxo4dq5SUFDk4sIUcAOQm69evl6Ojozw9PRUVFSVJeueddzR48GBVqlTJOq5t27YKCQnRnj17tGjRIt24ccNWJQMAbnPlyhUtWLBAL7/8suLi4jL0WywW6xK+zp07a+vWrfr9999tUCmQO7F8D3gAREZGqlGjRhmWd9y8eVMFChRQQkKCnnnmGe3atUvff/+96tatq0GDBmnevHnatm2bateubcPqAQB3M2nSJC1ZskQdO3bUjRs3rLOk2rZtKyn9kpANGzbokUceka+vrw0rBgDcbvfu3Wrbtq2aN2+uzz//XO7u7nccd+zYMbVv316bNm1SyZIlTa4SyJ0IpYBc7uDBg2rYsKGGDh2qkJAQa/uUKVNUsGBBvfLKK3J0dFRiYqKeeeYZRUREyN/fXz///DMb4ALAA+Dtt9/WF198obNnz+qzzz7T888/ny6MYt8RAMhdEhIS5ODgIHt7e2vbrl271K5duzsGU4mJierTp48qV66s4cOHq0CBArYoG8iVWL4H5GI3btzQM888o2LFiqlnz57W9ilTpmjcuHGqUqWKHB0dJUlOTk769ttvVbNmTa1cuVJbt24lkAKAXOSPP/7Q559/rilTpujAgQPW9jFjxqh///4qV66cDhw4oHPnzlmXe0jsOwIAucmRI0fUvXt3BQUF6amnnlJqaqokqV69evrxxx+1adMm9e3bV/Hx8ZKkpKQkDR8+XEuWLFHr1q0JpIDbEEoBudTNmzdVsGBBTZgwQZI0efJk/fXXX3rvvfc0depU/fDDD2rdunW6c5ydnbVixQqdO3dOtWrVskHVAIA72bdvn1q0aKGPPvpIH3zwgRo0aKBt27ZZ+4cNG6bnnntOa9eu1cyZMxUVFUUYBQC5TGRkpBo2bChXV1dVq1ZNx44d0zPPPGPtTwumNm7cqD59+ujixYsaNmyYPvvsM+3YseOeT8sG8it2PgZyoT///FOtW7fW+vXr9fTTT8vOzk6DBw/Wnj17dPLkSS1fvlzNmzdXamqq7OxuZcvz5s1TsWLF9Pjjj6t48eI2vgMAQJq9e/eqUaNGGjRokEaNGqVjx46pY8eOGjRokDZt2qQCBQrI0dFRb7zxhgzD0A8//KBr165pzJgxvJ8DQC6xf/9+NWrUSCNGjFBISIhSU1NVrFgxHT58WFeuXJGzs7McHByswdSTTz6pMmXKyMHBgT1egXtgphSQCxmGocTERI0ePVqJiYnq1KmT5syZo3Pnzqlhw4bWjRHTAqkxY8YoODg43dOaAAC2d/78eTVq1EjPP/+83nnnHbm7u6tevXoqU6aMbt68KRcXl3QzosaMGaNWrVpp//796fYqAQDYTlRUlAIDA9WkSRPrHq92dnY6e/asfvzxR9WpU0flypXT+++/r9jYWNWvX18rVqxQlSpVtGXLFrbUAO6BUArIhUqUKKGXXnpJ+/fv1/r16yVJHTp00Ny5c3Xw4EFNmTJFBw8elCSNHz9e06dP19atW1WxYkVblg0AuE1UVJTq1q2rrVu36vTp05JuLcfesWOHkpKS1Lt3b7Vs2VILFy7Url27JEnvvPOOli9fLi8vLxtWDgD4p4oVK8pisejrr7+WJL333ntasGCBxo8fr48//lht2rTRmDFjFBkZKUlq0KCBtm/fzpYawH3w9D0gl7h06ZKKFCliPY6Li1PTpk318MMPa82aNdb2H374QQMGDFDHjh2VkJCghQsXatu2bfL397dF2QCA+9izZ49Gjx6tU6dO6YknntCXX36pGTNm6NFHH9WxY8e0detWffPNNzp79qxat26tb7/9lllSAJBLpG2X8ccff2jAgAFKTEyUu7u7NmzYoGXLlql58+bWsV5eXurevbumTZtmu4KBBwyhFJALnDx5Ug0bNlTjxo316aefytXVVS4uLtq5c6eaNWumt956S8OHD7eOX7lypZ599llZLBb98ssvrFEHgFwoOTlZDg63tu+MiIhQSEiIVq1apXnz5qlXr17pxp49e1anTp2St7c3S7EBIJdJC6ZOnz6tQYMGaePGjXrllVc0ZcoUSbfe769cuaK2bduqV69eeuWVV2xcMfDgIJQCcoHjx4+rfv36iouLU+vWrdW+fXu1bNlS1apV07Bhw7RlyxbNnj1b9evXt56zadMmlShRgiV7AJCLXLx4UZcvX1aFChUy9O3cuVMTJkzQoUOH9NNPP6lcuXJKTEyUk5OTDSoFAGRFWjD1559/6tVXX9W1a9fUu3dvvfDCC5Jubakxb948bdq0SeXKlbNxtcCDg1AKsBHDMGSxWKz/kj5r1iydPn1aLi4u+vvvvxUREaEJEyaoaNGi6t69u55//nm9+eab6f7lHQCQe0RFRal69eoqU6aMmjRpoqFDh8rHx0cFChSwjtm1a5fGjh2ro0ePasOGDSpXrpxSUlJYrgcAuUja7+m3++eMqYEDB+rq1at67bXXtHfvXk2ePFnbt29nU3Mgi9joHLCRa9euSZI1YKpZs6YOHz6sxo0b6/3331ePHj303HPPadu2bSpbtqymT5+uAwcOEEgBQC517tw5FStWTGPGjNGxY8cUHByszp076+DBg4qNjZUk1atXTxMnTlTlypVVu3ZtnT59mkAKAHKZ2wOptHkcdnZ2Sk1Nla+vrz744AN5eHjopZde0tSpU/XLL78QSAH/AqEUYAPR0dHy8/PTG2+8oTNnzkiSmjVrpsaNG6tHjx6KjY3VgAEDtHLlSh04cED29vaKi4vT//73P6WkpIgJjgCQ+9SrV08VKlTQhg0btHbtWo0cOVIlS5bUs88+qxdffNH6xCZ/f3+99957atGihZKTk21cNQBAkk6dOqUZM2aoe/fu6t69uxYtWqQTJ05IuhVS3SmYmj59ugIDA7V7924eOgT8SyzfA2zg8uXLmjVrlt5//335+/vr8ccf1+DBgyXJuvntzJkz5e7urpiYGB06dEjTpk3TpEmTVL16ddsVDgC4o7Sl1bt27dL//vc/TZs2TTVq1JAk1alTR7GxsTp//rzatm2r8uXL691331VKSoocHR1tXDkAYN++fWrXrp3q1aunS5cuyTAMbdu2TY8++qhGjRql9u3bZzgnbSkfS7CB/4aZUoANeHh46M0339T27dvl6emp2bNnq0WLFjp69KiCgoIk3dp3RJK8vb3VokULrVy5kkAKAHKptKXVvr6+unz5stasWSNJ6tmzp6KiorRhwwbt3btXPj4++umnn3T27FkCKQDIBX7//Xe1bdtWvXv31jfffKPNmzdry5Yt+uKLL3Tu3DmNHTtWmzZtynCend2tj9IEUsB/w0wpwMZiY2MVHh6usWPHKi4uTs8++6zCwsLk7++vTz75xNblAQDu4M8//9TmzZt1/fp1tW3bVqVKlbJujLtq1SqNGjVKxYoV05EjR/Tjjz+qdu3akqQbN24oJSVFrq6uNr4DAIAkvffee/rll1+0ePFiOTg4yN7e3vp+/t1336lfv37q1KmTPvnkE9nZ2d1xA3QA/x4zpQAb8/T0VFBQkPbs2aMnnnhCkZGRio6O1ty5c/XZZ5/ZujwAwG0OHjyoDh06aO3atTpx4oRKlSol6f9vjFujRg15enrq3Llz2rp1qzWQMgxDBQsWJJACgFxk+/btSkhIkLOzc4ZZT506ddKgQYO0aNEi/fnnnwRSQA4glAJygbQJi9OnT9eoUaP0wgsvyNXVVU2aNLFxZQCAfzp48KAeffRRPf7445o9e7amTJkiSVq5cqVWrlwpSSpdurTatm2ry5cvy9vbW9LdHy8OALCdpKQkOTk5WX8XT0pKknTrHxlSUlIkSW3atJGDg4POnTtnszqBvIxQCsgF/vlEj+bNm2vy5Mn6888/VblyZRtXBgBIExsbq1dffVXdunXTW2+9JTc3N0nSlClT9OSTT2rGjBlasWKFJGnIkCEqV66c3nvvPaWmphJIAUAucfbsWR09elSS5OjoqIYNG2rdunXasWOHHB0dlZqaKsMwrHtG3bhxQz4+PvLx8bFl2UCeRSgF5BK3f2BJ+7ADAMgdYmJidO7cOXXq1EmpqamSpI8//lhjx47Vhx9+KHt7e82dO1c//PCDXFxcVLp0aW3ZskUJCQk2rhwAIN16Yl779u31zDPP6ODBg5KkJ598UrVr11bnzp3122+/WfeNSvvdfPXq1SpevLiKFi1qy9KBPIuNzgEAADLhyy+/VK9evZSUlGT9sPLnn3/q1KlTevTRR3XgwAENHjxYcXFxWrFihSwWi65du6YKFSrYuHIAQJrz588rICBApUuX1ty5c1WpUiUtW7ZMY8aMUUxMjKZMmaJatWopJSVFS5cu1SeffKJt27apZs2ati4dyJMcbF0AAADAg8DX11cODg5avny5OnXqJMMwVLJkSZUsWVKpqamqVq2aunTporlz58owDD388MO2LhkA8A+GYah48eL69ddf5e/vr759+yo0NFSdO3dW4cKF9cknn+ill16Sk5OTypcvL3d3dwIpIIcRSgEAAGSCr6+v3N3dtWDBAvn7+6tMmTLWvrS9R44ePWodBwCwvb///lsxMTFKSUlR9erVJUne3t6KiIiQv7+/evbsqdDQULVp00Zt2rRRZGSkLl++rBIlSsjLy0seHh62vQEgj2P5HgAAQCYtW7ZMzz//vLp06aLXX39dfn5+kqT4+Hi9/fbb+uyzz7R161ZVrVrVxpUCAA4cOKA+ffro4sWLMgxDbdq00aeffmrtj4mJkb+/v3x9ffXJJ5/w3g3YAKEUAABAJqWkpOizzz7TgAEDVL58eTVq1EiOjo46d+6cdu/erTVr1qh27dq2LhMA8r29e/eqcePGevnll9WhQwctXbpUc+fO1YwZM/TKK68oMTFRTk5O1mCqYsWKmjFjhmrUqGHr0oF8hVAKAAAgi3799VdNnTpVJ0+eVOHChdWkSRP17dtX5cuXt3VpAJDvnThxQtWrV9fw4cP11ltvSZJOnTqlypUra+DAgXrvvffSjY+JiVGZMmXUsmVLLV++XE5OTrYoG8iX2FMKAAAgixo0aKBvv/1W9vb2ti4FAPAPqampmjdvngoXLqyiRYta2xcvXqykpCQdP35cM2bMUNGiRfXMM8/IYrHI29tbf/zxh+Lj4wmkAJMxUwoAAOBfMAxDFoslw58BALYVFRWlqVOnaseOHerZs6euXLmiyZMnKzg4WLVq1dJXX32ls2fPKiYmRhUqVNDAgQPVsWNHW5cN5EuEUgAAAACAPCU6OloTJ05UWFiYTp48qXXr1umxxx6TJCUnJ8vBwUEffvih9uzZo+HDh1sfXAHAXIRSAAAAAIA8JyYmRu+88442bdqkHj16aNiwYZJk3eRc+v8BFQDb4P99AAAAAIA8x9vbW6NHj1ZqaqqWLFmi5ORkjRo1Sk5OTtYwikAKsC1mSgEAAAAA8qy0pXy//fabWrZsqZCQEFuXBOD/2Nm6AAAAAAAAcoqPj4/eeOMNVahQQdu3b9fff/9t65IA/B9mSgEAAAAA8ryYmBhJt5b1AcgdCKUAAAAAAABgOpbvAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAAAAAwHSEUgAAAAAAADAdoRQAAAAAAABMRygFAAAAAAAA0xFKAQAAmCA6OloDBw5UuXLl5OzsrFKlSunxxx/Xhg0bTK3DYrFoxYoVpr4mAADAnTjYugAAAIC87vTp02rcuLE8PDz07rvvqnr16kpKStK6desUHBysI0eO2LpEAAAA0zFTCgAAIIe9+uqrslgs2rlzpzp37qyKFSuqatWqGjp0qHbs2CFJOnPmjJ588km5urrKzc1Nzz77rGJiYqzX6NWrlzp27JjuuoMHD1bz5s2tx82bN9drr72mkSNHytPTUz4+Pho/fry139fXV5L01FNPyWKxyNfXV6dPn5adnZ12796d7tozZsxQmTJllJqamq3fCwAAgDSEUgAAADkoNjZWa9euVXBwsAoVKpSh38PDQ6mpqXryyScVGxurzZs3KywsTL///ru6dOmS5ddbsGCBChUqpF9//VVTp07VhAkTFBYWJknatWuXJGn+/Pk6f/68du3aJV9fX7Vq1Urz589Pd5358+erV69esrPj10UAAJAz+C0DAAAgB504cUKGYahy5cp3HbNhwwbt379fX3/9tfz9/dWgQQMtXLhQmzdvtgZJmVWjRg2NGzdOFSpUUI8ePVS3bl3rvlXFihWTdCsI8/HxsR7369dPixYtUkJCgiRpz5492r9/v3r37v1vbhkAACBTCKUAAABykGEY9x1z+PBhlSpVSqVKlbK2+fn5ycPDQ4cPH87S69WoUSPdcfHixXXhwoV7ntOxY0fZ29tr+fLlkqTQ0FC1aNHCutwPAAAgJxBKAQAA5KAKFSrIYrH8583M7ezsMgRcSUlJGcY5OjqmO7ZYLPfdF8rJyUk9evTQ/PnzlZiYqK+//lp9+vT5T/UCAADcD6EUAABADvL09FRgYKBmz56ta9euZei/fPmyqlSporNnz+rs2bPW9kOHDuny5cvy8/OTdGvp3fnz59OdGxkZmeV6HB0dlZKSkqG9X79++umnnzRnzhwlJyerU6dOWb42AABAVhBKAQAA5LDZs2crJSVF9evX17Jly3T8+HEdPnxYs2bNUkBAgFq1aqXq1aurW7du2rNnj3bu3KkePXqoWbNmqlu3riTpscce0+7du7Vw4UIdP35c48aN04EDB7Jci6+vrzZs2KDo6GhdunTJ2l6lShU1bNhQo0aN0nPPPaeCBQtm2/0DAADcCaEUAABADitXrpz27NmjFi1aaNiwYapWrZpat26tDRs26KOPPpLFYtH333+vIkWKqGnTpmrVqpXKlSunb775xnqNwMBAjR07ViNHjlS9evV05coV9ejRI8u1TJs2TWFhYSpVqpRq166drq9v375KTExk6R4AADCFxcjM7psAAADI89566y0tWbJE+/bts3UpAAAgH2CmFAAAQD539epVHThwQB9++KEGDhxo63IAAEA+QSgFAACQzw0YMED+/v5q3rw5S/cAAIBpWL4HAAAAAAAA0zFTCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKYjlAIAAAAAAIDpCKUAAAAAAABgOkIpAAAAAAAAmI5QCgAAAAAAAKb7fynfVmxOucTmAAAAAElFTkSuQmCC",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot historical spills by county\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "historical_spills['county'].value_counts().plot(kind='bar')\n",
+    "plt.title('Historical Spills by County')\n",
+    "plt.xlabel('County')\n",
+    "plt.ylabel('Number of Spills')\n",
+    "plt.xticks(rotation=45)\n",
+    "plt.grid(axis='y')\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n",
+    "\n",
+    "# plot recent spills by county\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "recent_spills['county'].value_counts().plot(kind='bar')\n",
+    "plt.title('Recent Spills by County')\n",
+    "plt.xlabel('County')\n",
+    "plt.ylabel('Number of Spills')\n",
+    "plt.xticks(rotation=45)\n",
+    "plt.grid(axis='y')\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "5fdfc7fc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create period column for 2020\n",
+    "import statsmodels.formula.api as smf\n",
+    "\n",
+    "# Create 'Period' column\n",
+    "spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2020' if x < 2020 else '2020 and After')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "71ce1288",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spills_gdf['report_delay'] = (spills_gdf['Initial Report Date'] - spills_gdf['Date of Discovery']).dt.days\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "3940b2ae",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spills_gdf = spills_gdf[spills_gdf['report_delay'] >= 0]\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "d69294c0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spills_gdf = spills_gdf.rename(columns={\n",
+    "    'Report Delay (Days)': 'report_delay',\n",
+    "    'Spill Type': 'spill_type'\n",
+    "})\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "fb84dc23",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "IQR bounds: Q1=0.0, Q3=1.0, IQR=1.0\n",
+      "Keeping report_delay in [0.0, 2.5]\n",
+      "Rows before outlier filter: 13227, after: 11376, removed: 1851\n",
+      "count    11376.000000\n",
+      "mean         0.663326\n",
+      "std          0.691330\n",
+      "min          0.000000\n",
+      "25%          0.000000\n",
+      "50%          1.000000\n",
+      "75%          1.000000\n",
+      "max          2.000000\n",
+      "Name: report_delay, dtype: float64\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Outlier removal using IQR on report_delay (run after report_delay is created)\n",
+    "if 'report_delay' not in spills_gdf.columns:\n",
+    "    print('report_delay column not found — skipping outlier removal in this cell.')\n",
+    "else:\n",
+    "    q1 = spills_gdf['report_delay'].quantile(0.25)\n",
+    "    q3 = spills_gdf['report_delay'].quantile(0.75)\n",
+    "    iqr = q3 - q1\n",
+    "    lower = max(0, q1 - 1.5 * iqr)\n",
+    "    upper = q3 + 1.5 * iqr\n",
+    "    print(f'IQR bounds: Q1={q1:.1f}, Q3={q3:.1f}, IQR={iqr:.1f}')\n",
+    "    print(f'Keeping report_delay in [{lower:.1f}, {upper:.1f}]')\n",
+    "    before_n = len(spills_gdf)\n",
+    "    # keep only rows with report_delay inside bounds\n",
+    "    spills_gdf = spills_gdf[(spills_gdf['report_delay'] >= lower) & (spills_gdf['report_delay'] <= upper)].copy()\n",
+    "    after_n = len(spills_gdf)\n",
+    "    print(f'Rows before outlier filter: {before_n}, after: {after_n}, removed: {before_n - after_n}')\n",
+    "    # show a short summary of report_delay after filtering\n",
+    "    print(spills_gdf['report_delay'].describe())\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "d03aed63",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# spill type by rurality\n",
+    "spill_type_rurality = spills_gdf.groupby(['rurality', 'spill_type']).size().unstack(fill_value=0)\n",
+    "spill_type_rurality = spill_type_rurality.reindex(['Urban', 'Suburban', 'Rural'], fill_value=0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "9cecca31",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Axes: xlabel='rurality'>"
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot spill type by rurality\n",
+    "spill_type_rurality.plot(kind='bar', stacked=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "29261e71",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "                 Generalized Linear Model Regression Results                  \n",
+      "==============================================================================\n",
+      "Dep. Variable:           report_delay   No. Observations:                11376\n",
+      "Model:                            GLM   Df Residuals:                    11364\n",
+      "Model Family:        NegativeBinomial   Df Model:                           11\n",
+      "Link Function:                    Log   Scale:                          1.0000\n",
+      "Method:                          IRLS   Log-Likelihood:                -12475.\n",
+      "Date:                Thu, 14 Aug 2025   Deviance:                       6528.1\n",
+      "Time:                        16:30:34   Pearson chi2:                 4.87e+03\n",
+      "No. Iterations:                     5   Pseudo R-squ. (CS):            0.04302\n",
+      "Covariance Type:            nonrobust                                         \n",
+      "============================================================================================================================================\n",
+      "                                                                               coef    std err          z      P>|z|      [0.025      0.975]\n",
+      "--------------------------------------------------------------------------------------------------------------------------------------------\n",
+      "Intercept                                                                   -0.4954      0.069     -7.201      0.000      -0.630      -0.361\n",
+      "C(spill_type)[T.Recent]                                                      0.1155      0.077      1.497      0.135      -0.036       0.267\n",
+      "C(Period)[T.Before 2020]                                                     0.3768      0.110      3.416      0.001       0.161       0.593\n",
+      "C(rurality)[T.Suburban]                                                     -0.3718      0.157     -2.364      0.018      -0.680      -0.064\n",
+      "C(rurality)[T.Urban]                                                        -0.5702      0.079     -7.231      0.000      -0.725      -0.416\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]                            -0.0985      0.121     -0.812      0.417      -0.336       0.139\n",
+      "C(spill_type)[T.Recent]:C(rurality)[T.Suburban]                              0.1757      0.184      0.955      0.339      -0.185       0.536\n",
+      "C(spill_type)[T.Recent]:C(rurality)[T.Urban]                                 0.4987      0.098      5.115      0.000       0.308       0.690\n",
+      "C(Period)[T.Before 2020]:C(rurality)[T.Suburban]                             0.2468      0.251      0.984      0.325      -0.245       0.738\n",
+      "C(Period)[T.Before 2020]:C(rurality)[T.Urban]                                0.4714      0.124      3.790      0.000       0.228       0.715\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]:C(rurality)[T.Suburban]    -0.1710      0.290     -0.589      0.556      -0.740       0.398\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]:C(rurality)[T.Urban]       -0.3650      0.149     -2.455      0.014      -0.656      -0.074\n",
+      "============================================================================================================================================\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/dadams/CSU Fullerton Dropbox/David Adams/Research Projects/colorado_spills_ejproject/github/colorado_spills/.venv/lib/python3.13/site-packages/statsmodels/genmod/families/family.py:1367: ValueWarning: Negative binomial dispersion parameter alpha not set. Using default value alpha=1.0.\n",
+      "  warnings.warn(\"Negative binomial dispersion parameter alpha not \"\n"
+     ]
+    }
+   ],
+   "source": [
+    "import statsmodels.formula.api as smf\n",
+    "import statsmodels.api as sm\n",
+    "\n",
+    "# Make sure spill_type and Period are clean, and report_delay is numeric\n",
+    "spills_gdf = spills_gdf.rename(columns={\n",
+    "    'Report Delay (Days)': 'report_delay',\n",
+    "    'Spill Type': 'spill_type',\n",
+    "    'rurality': 'rurality'\n",
+    "})\n",
+    "\n",
+    "# Fit the Negative Binomial model with interaction\n",
+    "nb_model = smf.glm(\n",
+    "    formula=\"report_delay ~ C(spill_type) * C(Period) * C(rurality)\",\n",
+    "    data=spills_gdf,\n",
+    "    family=sm.families.NegativeBinomial()\n",
+    ").fit()\n",
+    "\n",
+    "# View the results\n",
+    "print(nb_model.summary())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "id": "be634f4e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Pearson chi2: 7807.6, dispersion (chi2/df): 0.687\n"
+     ]
+    }
+   ],
+   "source": [
+    "# after nb_model = smf.glm(...).fit()\n",
+    "# 1) Pearson dispersion\n",
+    "pred = nb_model.predict(spills_gdf)\n",
+    "pearson_chi2 = ((spills_gdf['report_delay'] - pred)**2 / pred).sum()\n",
+    "dispersion = pearson_chi2 / nb_model.df_resid\n",
+    "print(f'Pearson chi2: {pearson_chi2:.1f}, dispersion (chi2/df): {dispersion:.3f}')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "id": "258159b7",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "spill_type         Period rurality    n\n",
+      "Historical    Before 2020 Suburban   74\n",
+      "Historical 2020 and After Suburban  169\n",
+      "    Recent    Before 2020 Suburban  206\n",
+      "Historical    Before 2020    Rural  286\n",
+      "    Recent 2020 and After Suburban  354\n",
+      "Historical 2020 and After    Rural  558\n",
+      "    Recent    Before 2020    Urban 1022\n",
+      "Historical    Before 2020    Urban 1233\n",
+      "    Recent 2020 and After    Urban 1247\n",
+      "    Recent    Before 2020    Rural 1594\n",
+      "    Recent 2020 and After    Rural 2006\n",
+      "Historical 2020 and After    Urban 2627\n",
+      "Summary of cell counts:\n",
+      "count      12.00000\n",
+      "mean      948.00000\n",
+      "std       819.99357\n",
+      "min        74.00000\n",
+      "25%       266.00000\n",
+      "50%       790.00000\n",
+      "75%      1333.75000\n",
+      "max      2627.00000\n",
+      "Name: n, dtype: float64\n",
+      "\n",
+      "Min count per cell: 74\n",
+      "\n",
+      "Counts look sufficient for the 3-way interaction (min cell >= 20).\n"
+     ]
+    }
+   ],
+   "source": [
+    "# (Inserted) Check counts for spill_type × Period × rurality — run this BEFORE relying on the 3-way interaction\n",
+    "import pandas as pd\n",
+    "\n",
+    "group_counts = spills_gdf.groupby(['spill_type','Period','rurality']).size().reset_index(name='n').sort_values('n')\n",
+    "# Print smallest groups to inspect sparsity\n",
+    "print(group_counts.head(50).to_string(index=False))\n",
+    "\n",
+    "print('Summary of cell counts:')\n",
+    "print(group_counts['n'].describe())\n",
+    "min_n = int(group_counts['n'].min())\n",
+    "print(f'\\nMin count per cell: {min_n}')\n",
+    "if min_n < 20:\n",
+    "    print('\\nWarning: many cells have small counts (<20). Consider removing the 3-way interaction or stratifying by rurality.')\n",
+    "else:\n",
+    "    print('\\nCounts look sufficient for the 3-way interaction (min cell >= 20).')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "id": "97b74c6c",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Fitting Poisson GLM:  report_delay ~ C(spill_type) * C(Period) * C(rurality)\n",
+      "                 Generalized Linear Model Regression Results                  \n",
+      "==============================================================================\n",
+      "Dep. Variable:           report_delay   No. Observations:                11376\n",
+      "Model:                            GLM   Df Residuals:                    11364\n",
+      "Model Family:                 Poisson   Df Model:                           11\n",
+      "Link Function:                    Log   Scale:                          1.0000\n",
+      "Method:                          IRLS   Log-Likelihood:                -11245.\n",
+      "Date:                Thu, 14 Aug 2025   Deviance:                       9405.6\n",
+      "Time:                        16:47:44   Pearson chi2:                 7.81e+03\n",
+      "No. Iterations:                     5   Pseudo R-squ. (CS):            0.06825\n",
+      "Covariance Type:            nonrobust                                         \n",
+      "============================================================================================================================================\n",
+      "                                                                               coef    std err          z      P>|z|      [0.025      0.975]\n",
+      "--------------------------------------------------------------------------------------------------------------------------------------------\n",
+      "Intercept                                                                   -0.4954      0.054     -9.135      0.000      -0.602      -0.389\n",
+      "C(spill_type)[T.Recent]                                                      0.1155      0.061      1.907      0.056      -0.003       0.234\n",
+      "C(Period)[T.Before 2020]                                                     0.3768      0.083      4.543      0.000       0.214       0.539\n",
+      "C(rurality)[T.Suburban]                                                     -0.3718      0.130     -2.849      0.004      -0.628      -0.116\n",
+      "C(rurality)[T.Urban]                                                        -0.5702      0.064     -8.965      0.000      -0.695      -0.446\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]                            -0.0985      0.091     -1.081      0.280      -0.277       0.080\n",
+      "C(spill_type)[T.Recent]:C(rurality)[T.Suburban]                              0.1757      0.151      1.164      0.244      -0.120       0.472\n",
+      "C(spill_type)[T.Recent]:C(rurality)[T.Urban]                                 0.4987      0.078      6.420      0.000       0.346       0.651\n",
+      "C(Period)[T.Before 2020]:C(rurality)[T.Suburban]                             0.2468      0.195      1.263      0.207      -0.136       0.630\n",
+      "C(Period)[T.Before 2020]:C(rurality)[T.Urban]                                0.4714      0.095      4.972      0.000       0.286       0.657\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]:C(rurality)[T.Suburban]    -0.1710      0.225     -0.760      0.448      -0.612       0.270\n",
+      "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]:C(rurality)[T.Urban]       -0.3650      0.113     -3.237      0.001      -0.586      -0.144\n",
+      "============================================================================================================================================\n",
+      "Pearson chi2: 7807.6, dispersion (chi2/df): 0.687\n",
+      "Failed to compute HC3 robust covariance: 'GLMResults' object has no attribute 'resid'\n",
+      "\n",
+      "Predicted group means (Poisson, HC3 CI where available):\n",
+      "spill_type         Period rurality  pred_mean  ci_lower  ci_upper    n\n",
+      "Historical 2020 and After    Rural   0.609319  0.547874  0.677655  558\n",
+      "Historical 2020 and After Suburban   0.420118  0.332928  0.530142  169\n",
+      "Historical 2020 and After    Urban   0.344499  0.322770  0.367692 2627\n",
+      "Historical    Before 2020    Rural   0.888112  0.785340  1.004333  286\n",
+      "Historical    Before 2020 Suburban   0.783784  0.605935  1.013833   74\n",
+      "Historical    Before 2020    Urban   0.804542  0.756001  0.856199 1233\n",
+      "    Recent 2020 and After    Rural   0.683948  0.648698  0.721114 2006\n",
+      "    Recent 2020 and After Suburban   0.562147  0.489225  0.645938  354\n",
+      "    Recent 2020 and After    Urban   0.636728  0.593944  0.682594 1247\n",
+      "    Recent    Before 2020    Rural   0.903388  0.857912  0.951274 1594\n",
+      "    Recent    Before 2020 Suburban   0.800971  0.687622  0.933005  206\n",
+      "    Recent    Before 2020    Urban   0.935421  0.877964  0.996638 1022\n"
+     ]
+    }
+   ],
+   "source": [
+    "# (Inserted) Fit Poisson GLM, compute HC3 robust covariance (fallback), and produce delta-method CIs for group means\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import patsy\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "\n",
+    "formula = \"report_delay ~ C(spill_type) * C(Period) * C(rurality)\"\n",
+    "print('Fitting Poisson GLM: ', formula)\n",
+    "poisson = smf.glm(formula=formula, data=spills_gdf, family=sm.families.Poisson()).fit()\n",
+    "print(poisson.summary())\n",
+    "\n",
+    "# Pearson dispersion check\n",
+    "pred = poisson.predict(spills_gdf)\n",
+    "pearson_chi2 = ((spills_gdf['report_delay'] - pred)**2 / (pred + 1e-12)).sum()\n",
+    "dispersion = pearson_chi2 / poisson.df_resid\n",
+    "print(f'Pearson chi2: {pearson_chi2:.1f}, dispersion (chi2/df): {dispersion:.3f}')\n",
+    "\n",
+    "# Robust covariance (HC3) with fallback to sandwich routine if wrapper missing\n",
+    "try:\n",
+    "    poisson_robust = poisson.get_robustcov_results(cov_type='HC3')\n",
+    "    cov = poisson_robust.cov_params()\n",
+    "    print('Used get_robustcov_results(cov_type=\"HC3\")')\n",
+    "except Exception as e:\n",
+    "    from statsmodels.stats.sandwich_covariance import cov_hc3\n",
+    "    try:\n",
+    "        cov = cov_hc3(poisson)\n",
+    "        print('Used cov_hc3() fallback')\n",
+    "    except Exception as e2:\n",
+    "        print('Failed to compute HC3 robust covariance:', e2)\n",
+    "        cov = poisson.cov_params()  # last resort\n",
+    "\n",
+    "# Prepare group combinations for prediction\n",
+    "group_df = (spills_gdf[['spill_type','Period','rurality']]\n",
+    "            .drop_duplicates()\n",
+    "            .sort_values(['spill_type','Period','rurality'])\n",
+    "            .reset_index(drop=True))\n",
+    "# Build design matrix for the group_df using the same RHS as the model\n",
+    "rhs = '1 + C(spill_type) * C(Period) * C(rurality)'\n",
+    "exog = patsy.dmatrix(rhs, group_df, return_type='dataframe')\n",
+    "# Align parameter vector and cov matrix to exog columns\n",
+    "params = poisson.params.copy()\n",
+    "# Ensure cov is a DataFrame indexed by model params\n",
+    "if isinstance(cov, np.ndarray):\n",
+    "    cov_df = pd.DataFrame(cov, index=poisson.params.index, columns=poisson.params.index)\n",
+    "else:\n",
+    "    cov_df = cov.copy()\n",
+    "# Add missing params to params (zero) so alignment is safe\n",
+    "for c in exog.columns:\n",
+    "    if c not in params.index:\n",
+    "        params.loc[c] = 0.0\n",
+    "params = params.reindex(exog.columns)\n",
+    "# Ensure cov_df has rows/cols for params (fill missing with 0)\n",
+    "cov_df = cov_df.reindex(index=params.index, columns=params.index).fillna(0.0)\n",
+    "\n",
+    "# Linear predictor and delta-method CIs\n",
+    "linpred = exog.values.dot(params.values)\n",
+    "# Var(linpred) = row * cov * row.T for each row. Compute diagonal efficiently:\n",
+    "linvar = np.einsum('ij,jk,ik->i', exog.values, cov_df.values, exog.values)\n",
+    "linse = np.sqrt(np.maximum(0, linvar))\n",
+    "crit = 1.96\n",
+    "pred_mean = np.exp(linpred)  # inverse link for Poisson with log link\n",
+    "lower = np.exp(linpred - crit * linse)\n",
+    "upper = np.exp(linpred + crit * linse)\n",
+    "\n",
+    "poisson_predicted_groups = group_df.copy()\n",
+    "poisson_predicted_groups['pred_mean'] = pred_mean\n",
+    "poisson_predicted_groups['ci_lower'] = lower\n",
+    "poisson_predicted_groups['ci_upper'] = upper\n",
+    "\n",
+    "# Add counts for reference\n",
+    "counts = spills_gdf.groupby(['spill_type','Period','rurality']).size().reset_index(name='n')\n",
+    "poisson_predicted_groups = poisson_predicted_groups.merge(counts, on=['spill_type','Period','rurality'], how='left')\n",
+    "\n",
+    "# Sort and display\n",
+    "poisson_predicted_groups = poisson_predicted_groups.sort_values(['spill_type','Period','rurality']).reset_index(drop=True)\n",
+    "print('\\nPredicted group means (Poisson, HC3 CI where available):')\n",
+    "print(poisson_predicted_groups.to_string(index=False))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "id": "b4c26a24",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Parametric bootstrap results (median, 95% CI):\n",
+      "spill_type         Period rurality    n  pred_median_boot  ci_lower_boot  ci_upper_boot\n",
+      "Historical 2020 and After    Rural  558          0.609319       0.546595       0.673835\n",
+      "Historical 2020 and After Suburban  169          0.414201       0.316420       0.514793\n",
+      "Historical 2020 and After    Urban 2627          0.345261       0.322802       0.369262\n",
+      "Historical    Before 2020    Rural  286          0.888112       0.783217       0.993007\n",
+      "Historical    Before 2020 Suburban   74          0.783784       0.594595       0.986486\n",
+      "Historical    Before 2020    Urban 1233          0.806164       0.748966       0.855251\n",
+      "    Recent 2020 and After    Rural 2006          0.684696       0.648056       0.719105\n",
+      "    Recent 2020 and After Suburban  354          0.556497       0.490042       0.645551\n",
+      "    Recent 2020 and After    Urban 1247          0.636728       0.594186       0.680834\n",
+      "    Recent    Before 2020    Rural 1594          0.904642       0.854125       0.957042\n",
+      "    Recent    Before 2020 Suburban  206          0.800971       0.684466       0.924879\n",
+      "    Recent    Before 2020    Urban 1022          0.935421       0.881556       0.999022\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Parametric bootstrap (Poisson) for group mean CIs — Option B\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
+    "import patsy\n",
+    "\n",
+    "# Number of bootstrap replications (increase to 1000+ for final results)\n",
+    "B = 500\n",
+    "n_groups = len(group_df)\n",
+    "preds = np.full((n_groups, B), np.nan)\n",
+    "failures = 0\n",
+    "\n",
+    "full_formula = formula  # uses the same formula string defined earlier\n",
+    "# Pre-build group design matrix once\n",
+    "ex_group = patsy.dmatrix(rhs, group_df, return_type='dataframe')\n",
+    "\n",
+    "for b in range(B):\n",
+    "    # simulate outcome under fitted Poisson mean\n",
+    "    y_sim = np.random.poisson(lam=poisson.fittedvalues)\n",
+    "    df_sim = spills_gdf.copy()\n",
+    "    df_sim['report_delay'] = y_sim\n",
+    "    try:\n",
+    "        m_sim = smf.glm(formula=full_formula, data=df_sim, family=sm.families.Poisson()).fit(disp=False)\n",
+    "        # predict for group_df using analytic prediction from coef vector\n",
+    "        p = m_sim.params.reindex(ex_group.columns).fillna(0.0)\n",
+    "        lin = ex_group.values.dot(p.values)\n",
+    "        preds[:, b] = np.exp(lin)\n",
+    "    except Exception:\n",
+    "        failures += 1\n",
+    "        # leave column as NaN for this replication\n",
+    "\n",
+    "# Report failures\n",
+    "if failures:\n",
+    "    print(f'Bootstrap replications with errors: {failures} / {B}')\n",
+    "\n",
+    "# Compute bootstrap summaries (median and 95% CI)\n",
+    "lower = np.nanpercentile(preds, 2.5, axis=1)\n",
+    "upper = np.nanpercentile(preds, 97.5, axis=1)\n",
+    "median = np.nanmedian(preds, axis=1)\n",
+    "\n",
+    "poisson_predicted_groups['pred_median_boot'] = median\n",
+    "poisson_predicted_groups['ci_lower_boot'] = lower\n",
+    "poisson_predicted_groups['ci_upper_boot'] = upper\n",
+    "\n",
+    "# Show results alongside counts\n",
+    "print('\\nParametric bootstrap results (median, 95% CI):')\n",
+    "print(poisson_predicted_groups[['spill_type','Period','rurality','n','pred_median_boot','ci_lower_boot','ci_upper_boot']].to_string(index=False))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "id": "be3ce0f5",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Bootstrap finished — failures: 0 / 2000\n",
+      "Updated bootstrap summaries (B=2000):\n",
+      "spill_type         Period rurality    n  pred_median_boot  ci_lower_boot  ci_upper_boot\n",
+      "Historical 2020 and After    Rural  558          0.607527       0.544803       0.677419\n",
+      "Historical 2020 and After Suburban  169          0.420118       0.325296       0.520710\n",
+      "Historical 2020 and After    Urban 2627          0.345261       0.322792       0.368100\n",
+      "Historical    Before 2020    Rural  286          0.884615       0.779720       0.996503\n",
+      "Historical    Before 2020 Suburban   74          0.783784       0.594595       0.986486\n",
+      "Historical    Before 2020    Urban 1233          0.804947       0.753447       0.856448\n",
+      "    Recent 2020 and After    Rural 2006          0.683948       0.650050       0.720837\n",
+      "    Recent 2020 and After Suburban  354          0.562147       0.485876       0.638418\n",
+      "    Recent 2020 and After    Urban 1247          0.635525       0.594226       0.678448\n",
+      "    Recent    Before 2020    Rural 1594          0.904015       0.858830       0.949812\n",
+      "    Recent    Before 2020 Suburban  206          0.800971       0.679490       0.927184\n",
+      "    Recent    Before 2020    Urban 1022          0.934442       0.875709       0.997089\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Re-run parametric bootstrap with more reps (B=2000) and save draws as preds_boot\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
+    "import patsy\n",
+    "\n",
+    "B = 2000  # more stable CIs for reporting\n",
+    "n_groups = len(group_df)\n",
+    "preds_boot = np.full((n_groups, B), np.nan)\n",
+    "failures = 0\n",
+    "ex_group = patsy.dmatrix(rhs, group_df, return_type='dataframe')\n",
+    "full_formula = formula\n",
+    "\n",
+    "for b in range(B):\n",
+    "    y_sim = np.random.poisson(lam=poisson.fittedvalues)\n",
+    "    df_sim = spills_gdf.copy()\n",
+    "    df_sim['report_delay'] = y_sim\n",
+    "    try:\n",
+    "        m_sim = smf.glm(formula=full_formula, data=df_sim, family=sm.families.Poisson()).fit(disp=False)\n",
+    "        p = m_sim.params.reindex(ex_group.columns).fillna(0.0)\n",
+    "        lin = ex_group.values.dot(p.values)\n",
+    "        preds_boot[:, b] = np.exp(lin)\n",
+    "    except Exception:\n",
+    "        failures += 1\n",
+    "\n",
+    "print(f'Bootstrap finished — failures: {failures} / {B}')\n",
+    "\n",
+    "# Update poisson_predicted_groups with more precise bootstrap CIs\n",
+    "poisson_predicted_groups['pred_median_boot'] = np.nanmedian(preds_boot, axis=1)\n",
+    "poisson_predicted_groups['ci_lower_boot'] = np.nanpercentile(preds_boot, 2.5, axis=1)\n",
+    "poisson_predicted_groups['ci_upper_boot'] = np.nanpercentile(preds_boot, 97.5, axis=1)\n",
+    "\n",
+    "print('Updated bootstrap summaries (B=2000):')\n",
+    "print(poisson_predicted_groups[['spill_type','Period','rurality','n','pred_median_boot','ci_lower_boot','ci_upper_boot']].to_string(index=False))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "id": "4f5ca7db",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Bootstrap contrasts:\n",
+      "contrast_type spill_type         Period rurality          comparison   median  ci_lower  ci_upper  p_value\n",
+      "       Period Historical                   Rural      Before - After 0.274971  0.145211  0.402151    0.000\n",
+      "       Period Historical                Suburban      Before - After 0.358628  0.142310  0.579082    0.001\n",
+      "       Period Historical                   Urban      Before - After 0.460224  0.404396  0.516836    0.000\n",
+      "       Period     Recent                   Rural      Before - After 0.220333  0.159595  0.277098    0.000\n",
+      "       Period     Recent                Suburban      Before - After 0.238385  0.095303  0.387656    0.000\n",
+      "       Period     Recent                   Urban      Before - After 0.298987  0.221982  0.374905    0.000\n",
+      "    SpillType            2020 and After    Rural Recent - Historical 0.075513  0.001944  0.149510    0.045\n",
+      "    SpillType            2020 and After Suburban Recent - Historical 0.144319  0.014505  0.262164    0.029\n",
+      "    SpillType            2020 and After    Urban Recent - Historical 0.291087  0.243903  0.341355    0.000\n",
+      "    SpillType               Before 2020    Rural Recent - Historical 0.019446 -0.098070  0.133420    0.753\n",
+      "    SpillType               Before 2020 Suburban Recent - Historical 0.020467 -0.221622  0.243548    0.850\n",
+      "    SpillType               Before 2020    Urban Recent - Historical 0.129103  0.049869  0.211084    0.002\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Compute bootstrap contrasts using preds_boot\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "\n",
+    "# Helper: find row index in group_df matching values\n",
+    "def find_idx(spill_type, period, rurality):\n",
+    "    mask = (group_df['spill_type'] == spill_type) & (group_df['Period'] == period) & (group_df['rurality'] == rurality)\n",
+    "    matches = group_df.index[mask].tolist()\n",
+    "    return matches[0] if matches else None\n",
+    "\n",
+    "contrast_rows = []\n",
+    "# Period contrast: Before 2020 - 2020 and After for each spill_type × rurality\n",
+    "for st in group_df['spill_type'].unique():\n",
+    "    for r in group_df['rurality'].unique():\n",
+    "        idx_before = find_idx(st, 'Before 2020', r)\n",
+    "        idx_after = find_idx(st, '2020 and After', r)\n",
+    "        if idx_before is None or idx_after is None:\n",
+    "            continue\n",
+    "        diffs = preds_boot[idx_before, :] - preds_boot[idx_after, :]\n",
+    "        med = np.nanmedian(diffs)\n",
+    "        lo = np.nanpercentile(diffs, 2.5)\n",
+    "        hi = np.nanpercentile(diffs, 97.5)\n",
+    "        # two-sided p-value: proportion of draws where diff <= 0 (or >=0), convert to two-sided\n",
+    "        prop_le0 = np.nanmean(diffs <= 0)\n",
+    "        p_two = 2 * min(prop_le0, 1 - prop_le0)\n",
+    "        contrast_rows.append({'contrast_type':'Period','spill_type':st,'rurality':r,'comparison':'Before - After','median':med,'ci_lower':lo,'ci_upper':hi,'p_value':p_two})\n",
+    "\n",
+    "# Spill type contrast: Recent - Historical for each Period × rurality\n",
+    "for per in group_df['Period'].unique():\n",
+    "    for r in group_df['rurality'].unique():\n",
+    "        idx_recent = find_idx('Recent', per, r)\n",
+    "        idx_hist = find_idx('Historical', per, r)\n",
+    "        if idx_recent is None or idx_hist is None:\n",
+    "            continue\n",
+    "        diffs = preds_boot[idx_recent, :] - preds_boot[idx_hist, :]\n",
+    "        med = np.nanmedian(diffs)\n",
+    "        lo = np.nanpercentile(diffs, 2.5)\n",
+    "        hi = np.nanpercentile(diffs, 97.5)\n",
+    "        prop_le0 = np.nanmean(diffs <= 0)\n",
+    "        p_two = 2 * min(prop_le0, 1 - prop_le0)\n",
+    "        contrast_rows.append({'contrast_type':'SpillType','Period':per,'rurality':r,'comparison':'Recent - Historical','median':med,'ci_lower':lo,'ci_upper':hi,'p_value':p_two})\n",
+    "\n",
+    "contrast_df = pd.DataFrame(contrast_rows)\n",
+    "# tidy ordering\n",
+    "contrast_df = contrast_df[['contrast_type','spill_type','Period','rurality','comparison','median','ci_lower','ci_upper','p_value']].fillna('')\n",
+    "print('Bootstrap contrasts:')\n",
+    "print(contrast_df.to_string(index=False))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "id": "0ba6ac9e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved poisson_predicted_groups_boot_final.csv and poisson_boot_contrasts.csv\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Save results to CSV for reporting\n",
+    "poisson_predicted_groups.to_csv('poisson_predicted_groups_boot_final.csv', index=False)\n",
+    "contrast_df.to_csv('poisson_boot_contrasts.csv', index=False)\n",
+    "print('Saved poisson_predicted_groups_boot_final.csv and poisson_boot_contrasts.csv')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "71c9cf97",
+   "metadata": {},
+   "source": [
+    "## Short summary of results (for report)\n",
+    "The Poisson model with parametric bootstrap shows consistent reductions in predicted reporting delay after 2020 across spill types and ruralities (median reductions ≈ 0.22–0.46 days, ~5–11 hours).\n",
+    "Recent spills tend to have longer predicted delays than Historical spills, with the largest and most robust Recent>Historical gap in Urban areas (≈0.29 days post‑2020).\n",
+    "These conclusions are based on model‑based inference with parametric and nonparametric bootstrap checks (cells below compute percent changes, run a nonparametric bootstrap for comparison, and plot results)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "id": "d272a930",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Percent-change (Recent vs Historical) by Period × rurality:\n",
+      "        Period rurality  median_abs_days  ci_lower_days  ci_upper_days  median_hours  median_pct\n",
+      "2020 and After    Rural         0.075513       0.001944       0.149510      1.812305   12.423566\n",
+      "2020 and After Suburban         0.144319       0.014505       0.262164      3.463645   34.669871\n",
+      "2020 and After    Urban         0.291087       0.243903       0.341355      6.986082   84.405516\n",
+      "   Before 2020    Rural         0.019446      -0.098070       0.133420      0.466698    2.205239\n",
+      "   Before 2020 Suburban         0.020467      -0.221622       0.243548      0.491210    2.635229\n",
+      "   Before 2020    Urban         0.129103       0.049869       0.211084      3.098462   16.054024\n",
+      "Period change (Before - After) by Spill Type × rurality (days and hours):\n",
+      "spill_type rurality  median_days  ci_lower_days  ci_upper_days  median_hours\n",
+      "Historical    Rural     0.274971       0.145211       0.402151      6.599293\n",
+      "Historical Suburban     0.358628       0.142310       0.579082      8.607069\n",
+      "Historical    Urban     0.460224       0.404396       0.516836     11.045380\n",
+      "    Recent    Rural     0.220333       0.159595       0.277098      5.287993\n",
+      "    Recent Suburban     0.238385       0.095303       0.387656      5.721244\n",
+      "    Recent    Urban     0.298987       0.221982       0.374905      7.175694\n",
+      "Saved pct_change_spilltype_boot.csv and pct_change_period_boot.csv\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Add percent-change columns and an hours table for intuition\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "# Use preds_boot (parametric bootstrap draws) when available\n",
+    "if 'preds_boot' not in globals():\n",
+    "    print('preds_boot not found; run the B=2000 parametric bootstrap cell first.')\n",
+    "else:\n",
+    "    # Compute percent change for Recent vs Historical (within each Period × rurality)\n",
+    "    pct_rows = []\n",
+    "    def idx_for(st, per, r):\n",
+    "        mask = (group_df['spill_type'] == st) & (group_df['Period'] == per) & (group_df['rurality'] == r)\n",
+    "        res = group_df.index[mask].tolist()\n",
+    "        return res[0] if res else None\n",
+    "    for per in group_df['Period'].unique():\n",
+    "        for r in group_df['rurality'].unique():\n",
+    "            idx_rec = idx_for('Recent', per, r)\n",
+    "            idx_hist = idx_for('Historical', per, r)\n",
+    "            if idx_rec is None or idx_hist is None:\n",
+    "                continue\n",
+    "            base = preds_boot[idx_hist, :]\n",
+    "            diff = preds_boot[idx_rec, :] - base\n",
+    "            # relative change = diff / base (avoid divide-by-zero)\n",
+    "            rel = diff / np.where(base==0, np.nan, base)\n",
+    "            pct_rows.append({'Period': per, 'rurality': r, 'median_abs_days': np.nanmedian(diff), 'ci_lower_days': np.nanpercentile(diff,2.5), 'ci_upper_days': np.nanpercentile(diff,97.5), 'median_hours': 24*np.nanmedian(diff), 'median_pct': 100*np.nanmedian(rel) })\n",
+    "    pct_change_spilltype = pd.DataFrame(pct_rows)\n",
+    "    print('Percent-change (Recent vs Historical) by Period × rurality:')\n",
+    "    print(pct_change_spilltype.to_string(index=False))\n",
+    "    # Period change: Before - After (absolute days and hours)\n",
+    "    per_rows = []\n",
+    "    for st in group_df['spill_type'].unique():\n",
+    "        for r in group_df['rurality'].unique():\n",
+    "            idx_b = idx_for(st, 'Before 2020', r)\n",
+    "            idx_a = idx_for(st, '2020 and After', r)\n",
+    "            if idx_b is None or idx_a is None:\n",
+    "                continue\n",
+    "            diffs = preds_boot[idx_b, :] - preds_boot[idx_a, :]\n",
+    "            per_rows.append({'spill_type': st, 'rurality': r, 'median_days': np.nanmedian(diffs), 'ci_lower_days': np.nanpercentile(diffs,2.5), 'ci_upper_days': np.nanpercentile(diffs,97.5), 'median_hours': 24*np.nanmedian(diffs) })\n",
+    "    pct_change_period = pd.DataFrame(per_rows)\n",
+    "    print('Period change (Before - After) by Spill Type × rurality (days and hours):')\n",
+    "    print(pct_change_period.to_string(index=False))\n",
+    "    # Save supplemental tables\n",
+    "    pct_change_spilltype.to_csv('pct_change_spilltype_boot.csv', index=False)\n",
+    "    pct_change_period.to_csv('pct_change_period_boot.csv', index=False)\n",
+    "    print('Saved pct_change_spilltype_boot.csv and pct_change_period_boot.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "id": "2ed3ea41",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Nonparametric bootstrap summaries (median, 95% CI):\n",
+      "spill_type         Period rurality    n  pred_median_npboot  ci_lower_npboot  ci_upper_npboot\n",
+      "Historical 2020 and After    Rural  558            0.607517         0.552628         0.663669\n",
+      "Historical 2020 and After Suburban  169            0.416146         0.331361         0.510265\n",
+      "Historical 2020 and After    Urban 2627            0.344247         0.323949         0.363912\n",
+      "Historical    Before 2020    Rural  286            0.887597         0.811366         0.960140\n",
+      "Historical    Before 2020 Suburban   74            0.786776         0.597217         1.000294\n",
+      "Historical    Before 2020    Urban 1233            0.803498         0.766579         0.842416\n",
+      "    Recent 2020 and After    Rural 2006            0.682993         0.653382         0.712348\n",
+      "    Recent 2020 and After Suburban  354            0.560345         0.491066         0.633436\n",
+      "    Recent 2020 and After    Urban 1247            0.636031         0.598396         0.674217\n",
+      "    Recent    Before 2020    Rural 1594            0.904938         0.864069         0.939892\n",
+      "    Recent    Before 2020 Suburban  206            0.803289         0.705556         0.896866\n",
+      "    Recent    Before 2020    Urban 1022            0.936089         0.889444         0.979473\n",
+      "Saved preds_npboot.npy\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Nonparametric (case) bootstrap: resample rows with replacement and refit Poisson to get group CIs\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
+    "import patsy\n",
+    "B_np = 1000\n",
+    "n_groups = len(group_df)\n",
+    "preds_np = np.full((n_groups, B_np), np.nan)\n",
+    "failures = 0\n",
+    "ex_group = patsy.dmatrix(rhs, group_df, return_type='dataframe')\n",
+    "full_formula = formula\n",
+    "for b in range(B_np):\n",
+    "    try:\n",
+    "        sample = spills_gdf.sample(frac=1.0, replace=True)\n",
+    "        m_sim = smf.glm(formula=full_formula, data=sample, family=sm.families.Poisson()).fit(disp=False)\n",
+    "        p = m_sim.params.reindex(ex_group.columns).fillna(0.0)\n",
+    "        lin = ex_group.values.dot(p.values)\n",
+    "        preds_np[:, b] = np.exp(lin)\n",
+    "    except Exception:\n",
+    "        failures += 1\n",
+    "if failures:\n",
+    "    print(f'Nonparametric bootstrap failures: {failures} / {B_np}')\n",
+    "# Summaries and attach to poisson_predicted_groups\n",
+    "poisson_predicted_groups['pred_median_npboot'] = np.nanmedian(preds_np, axis=1)\n",
+    "poisson_predicted_groups['ci_lower_npboot'] = np.nanpercentile(preds_np, 2.5, axis=1)\n",
+    "poisson_predicted_groups['ci_upper_npboot'] = np.nanpercentile(preds_np, 97.5, axis=1)\n",
+    "print('Nonparametric bootstrap summaries (median, 95% CI):')\n",
+    "print(poisson_predicted_groups[['spill_type','Period','rurality','n','pred_median_npboot','ci_lower_npboot','ci_upper_npboot']].to_string(index=False))\n",
+    "# Save draws for reproducibility\n",
+    "np.save('preds_npboot.npy', preds_np)\n",
+    "print('Saved preds_npboot.npy')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "id": "72a5f649",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved predicted_means_boot_CI.png\n",
+      "Saved poisson_predicted_groups_boot_all.csv\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Plot predicted means with parametric bootstrap CIs and save figure\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "# Use poisson_predicted_groups (must exist from previous cells)\n",
+    "dfp = poisson_predicted_groups.copy()\n",
+    "# Create a label for plotting\n",
+    "dfp['label'] = dfp['spill_type'] + ' | ' + dfp['Period'] + ' | ' + dfp['rurality']\n",
+    "x = np.arange(len(dfp))\n",
+    "y = dfp['pred_median_boot'].values\n",
+    "yerr_low = y - dfp['ci_lower_boot'].values\n",
+    "yerr_high = dfp['ci_upper_boot'].values - y\n",
+    "plt.figure(figsize=(12,6))\n",
+    "plt.errorbar(x, y, yerr=[yerr_low, yerr_high], fmt='o', ecolor='gray', capsize=3)\n",
+    "plt.xticks(x, dfp['label'], rotation=90)\n",
+    "plt.ylabel('Predicted report_delay (days)')\n",
+    "plt.title('Predicted group means with parametric bootstrap 95% CI')\n",
+    "plt.tight_layout()\n",
+    "plt.savefig('predicted_means_boot_CI.png', dpi=150)\n",
+    "plt.show()\n",
+    "print('Saved predicted_means_boot_CI.png')\n",
+    "# Save updated poisson_predicted_groups with all bootstrap summaries\n",
+    "poisson_predicted_groups.to_csv('poisson_predicted_groups_boot_all.csv', index=False)\n",
+    "print('Saved poisson_predicted_groups_boot_all.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "d3b44df3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved bar_by_rurality_spilltype_Historical.png\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved bar_by_rurality_spilltype_Recent.png\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved bar_by_spilltype_rurality_Urban.png\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved bar_by_spilltype_rurality_Suburban.png\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Saved bar_by_spilltype_rurality_Rural.png\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Create grouped bar charts: (A) for each spill_type show rurality grouped bars (Before vs After),\n",
+    "# and (B) for each rurality show spill_type grouped bars (Before vs After).\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "# Verify required data\n",
+    "if 'poisson_predicted_groups' not in globals():\n",
+    "    print('poisson_predicted_groups not found - run the prediction/bootstrap cells first')\n",
+    "else:\n",
+    "    df = poisson_predicted_groups.copy()\n",
+    "    # Ensure required columns exist\n",
+    "    required = {'spill_type','Period','rurality','pred_median_boot','ci_lower_boot','ci_upper_boot'}\n",
+    "    if not required.issubset(set(df.columns)):\n",
+    "        print('Missing required columns in poisson_predicted_groups:', required - set(df.columns))\n",
+    "    else:\n",
+    "        period_order = ['Before 2020','2020 and After']\n",
+    "        rural_order = ['Urban','Suburban','Rural']\n",
+    "        # (A) by spill_type across rurality\n",
+    "        for st in sorted(df['spill_type'].unique()):\n",
+    "            sub = df[df['spill_type']==st].copy()\n",
+    "            # reindex to ensure consistent ordering\n",
+    "            idx = pd.MultiIndex.from_product([rural_order, period_order], names=['rurality','Period'])\n",
+    "            sub = sub.set_index(['rurality','Period']).reindex(idx).reset_index()\n",
+    "            means = sub['pred_median_boot'].values.reshape(len(rural_order), 2)\n",
+    "            lower = sub['ci_lower_boot'].values.reshape(len(rural_order), 2)\n",
+    "            upper = sub['ci_upper_boot'].values.reshape(len(rural_order), 2)\n",
+    "            err_low = means - lower\n",
+    "            err_high = upper - means\n",
+    "            x = np.arange(len(rural_order))\n",
+    "            width = 0.35\n",
+    "            fig, ax = plt.subplots(figsize=(8,4))\n",
+    "            ax.bar(x - width/2, means[:,0], width, yerr=[err_low[:,0], err_high[:,0]], label='Before 2020', capsize=5, color='#4C72B0')\n",
+    "            ax.bar(x + width/2, means[:,1], width, yerr=[err_low[:,1], err_high[:,1]], label='2020 and After', capsize=5, color='#55A868')\n",
+    "            ax.set_xticks(x)\n",
+    "            ax.set_xticklabels(rural_order)\n",
+    "            ax.set_ylabel('Predicted report_delay (days)')\n",
+    "            ax.set_title(f'Predicted mean report_delay by rurality – Spill type: {st}')\n",
+    "            ax.legend()\n",
+    "            plt.tight_layout()\n",
+    "            fname = f'bar_by_rurality_spilltype_{st.replace(' ','_')}.png'\n",
+    "            plt.savefig(fname, dpi=150)\n",
+    "            plt.show()\n",
+    "            print(f'Saved {fname}')\n",
+    "        # (B) by rurality across spill_type\n",
+    "        for r in rural_order:\n",
+    "            sub = df[df['rurality']==r].copy()\n",
+    "            spill_order = sorted(sub['spill_type'].unique())\n",
+    "            idx = pd.MultiIndex.from_product([spill_order, period_order], names=['spill_type','Period'])\n",
+    "            sub = sub.set_index(['spill_type','Period']).reindex(idx).reset_index()\n",
+    "            means = sub['pred_median_boot'].values.reshape(len(spill_order), 2)\n",
+    "            lower = sub['ci_lower_boot'].values.reshape(len(spill_order), 2)\n",
+    "            upper = sub['ci_upper_boot'].values.reshape(len(spill_order), 2)\n",
+    "            err_low = means - lower\n",
+    "            err_high = upper - means\n",
+    "            x = np.arange(len(spill_order))\n",
+    "            width = 0.35\n",
+    "            fig, ax = plt.subplots(figsize=(8,4))\n",
+    "            ax.bar(x - width/2, means[:,0], width, yerr=[err_low[:,0], err_high[:,0]], label='Before 2020', capsize=5, color='#4C72B0')\n",
+    "            ax.bar(x + width/2, means[:,1], width, yerr=[err_low[:,1], err_high[:,1]], label='2020 and After', capsize=5, color='#55A868')\n",
+    "            ax.set_xticks(x)\n",
+    "            ax.set_xticklabels(spill_order)\n",
+    "            ax.set_ylabel('Predicted report_delay (days)')\n",
+    "            ax.set_title(f'Predicted mean report_delay by spill type – Rurality: {r}')\n",
+    "            ax.legend()\n",
+    "            plt.tight_layout()\n",
+    "            fname = f'bar_by_spilltype_rurality_{r}.png'\n",
+    "            plt.savefig(fname, dpi=150)\n",
+    "            plt.show()\n",
+    "            print(f'Saved {fname}')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bcd6ba5b",
+   "metadata": {},
+   "source": [
+    "Methods\n",
+    "Data and sample\n",
+    "\n",
+    "Source: incident-level Colorado spills dataset (spatial fields preserved); analysis restricted to discovery dates from 2015–2024 and to the three counties with the largest number of spills (top‑3 counties), consistent with the study focus.\n",
+    "Outcome: report_delay, defined as days between Date of Discovery and Initial Report Date (non‑negative). Outliers were removed using the interquartile range (IQR) rule with the lower bound truncated to zero (lower = max(0, Q1 − 1.5·IQR); upper = Q3 + 1.5·IQR).\n",
+    "Key covariates: Spill Type (Historical vs Recent), Period (Before 2020 vs 2020 and After), and rurality classified from RUCA codes (Urban, Suburban, Rural). Analyses drop observations with Unknown RUCA.\n",
+    "Modeling approach\n",
+    "\n",
+    "Primary model: Poisson generalized linear model (log link) with full three‑way interaction: report_delay ~ C(spill_type) * C(Period) * C(rurality).\n",
+    "The three‑way interaction allows separate predicted means for each SpillType × Period × rurality cell.\n",
+    "Inference and contrasts\n",
+    "\n",
+    "Point predictions: predicted group means on the original scale (days) were obtained from the fitted Poisson model.\n",
+    "Primary contrasts:\n",
+    "Period contrasts = predicted mean (Before 2020) − predicted mean (2020 and After) within each SpillType × rurality cell (positive → larger delays before 2020).\n",
+    "SpillType contrasts = predicted mean (Recent) − predicted mean (Historical) within each Period × rurality cell (positive → Recent > Historical).\n",
+    "Uncertainty: parametric (model‑based) bootstrap was used as the main uncertainty estimator. Procedure: simulate outcomes Y_sim ∼ Poisson(μ̂) under the fitted model, refit the Poisson to each simulated dataset, compute predicted group means, and take percentiles of the resulting distribution. Final parametric bootstrap used B = 2,000 replications. Two‑sided bootstrap p‑values were computed as 2·min{Pr(diff ≤ 0), Pr(diff ≥ 0)} across draws.\n",
+    "Sensitivity: (1) attempted Negative Binomial — MLE collapsed toward Poisson (α ≈ 0) and provided unstable estimates; (2) a nonparametric (case) bootstrap (resample rows with replacement) with B = 1,000 was run to check robustness of group CIs.\n",
+    "Diagnostics and reporting: Pearson chi‑square dispersion reported. All model code used Python (pandas/geopandas, patsy, statsmodels, numpy); results, bootstrap draws, and figures were saved to CSV/PNG for reproducibility.\n",
+    "Results\n",
+    "Sample and diagnostics\n",
+    "\n",
+    "Final analytic sample: 11,376 observations (after filtering and outlier removal).\n",
+    "Model diagnostics: Poisson fit produced evidence of underdispersion (Pearson χ2/df ≈ 0.69). Negative Binomial MLE attempts did not stabilize (α→0), so the Poisson specification with bootstrap‑based inference was retained.\n",
+    "Primary contrasts (bootstrap medians, 95% bootstrap CI, two‑sided bootstrap p)\n",
+    "\n",
+    "Period contrasts (Before 2020 − 2020 and After). Positive numbers indicate larger predicted delays before 2020 (i.e., delays decreased after 2020).\n",
+    "\n",
+    "Historical, Urban: median = 0.460 days (95% CI 0.404, 0.517), p < 0.001 — ≈ 11.0 hours (0.460×24).\n",
+    "Historical, Suburban: median = 0.359 days (95% CI 0.142, 0.579), p = 0.001 — ≈ 8.6 hours.\n",
+    "Historical, Rural: median = 0.275 days (95% CI 0.145, 0.402), p < 0.001 — ≈ 6.6 hours.\n",
+    "Recent, Urban: median = 0.299 days (95% CI 0.222, 0.375), p < 0.001 — ≈ 7.2 hours.\n",
+    "Recent, Suburban: median = 0.238 days (95% CI 0.095, 0.388), p < 0.001 — ≈ 5.7 hours.\n",
+    "Recent, Rural: median = 0.220 days (95% CI 0.160, 0.277), p < 0.001 — ≈ 5.3 hours.\n",
+    "Interpretation: Across spill types and ruralities, predicted reporting delay fell after 2020 by roughly 0.22–0.46 days (≈5–11 hours). All Period contrasts are statistically robust by bootstrap p‑values.\n",
+    "\n",
+    "SpillType contrasts (Recent − Historical). Positive numbers indicate Recent spills have longer predicted delays than Historical spills.\n",
+    "\n",
+    "2020 and After:\n",
+    "Urban: median = 0.291 days (95% CI 0.244, 0.341), p < 0.001 — ≈ 7.0 hours.\n",
+    "Suburban: median = 0.144 days (95% CI 0.015, 0.262), p = 0.029 — ≈ 3.5 hours.\n",
+    "Rural: median = 0.0755 days (95% CI 0.0019, 0.1495), p = 0.045 — ≈ 1.8 hours.\n",
+    "Before 2020:\n",
+    "Urban: median = 0.129 days (95% CI 0.050, 0.211), p = 0.002 — ≈ 3.1 hours.\n",
+    "Suburban: median = 0.0205 days (95% CI −0.222, 0.244), p = 0.85 — not significant.\n",
+    "Rural: median = 0.0194 days (95% CI −0.0981, 0.1334), p = 0.753 — not significant.\n",
+    "Interpretation: The Recent vs Historical difference is largest and most consistent in Urban areas (≈3–7 hours depending on period), particularly after 2020. In Suburban and Rural areas the Recent > Historical gap is smaller and often not statistically distinguishable before 2020.\n",
+    "\n",
+    "Sensitivity and robustness\n",
+    "\n",
+    "Nonparametric (case) bootstrap (B = 1,000) produced broadly similar group CIs to the parametric bootstrap, supporting the direction and relative magnitude of the reported contrasts (detailed tables and saved draws are available in the repository: parametric and nonparametric bootstrap outputs and CSVs).\n",
+    "The estimated effects are small in absolute units. Although many contrasts are statistically significant under bootstrap inference, practical significance should be judged relative to policy thresholds (e.g., hours of delay important for response operations).\n",
+    "Limitations\n",
+    "\n",
+    "Model dependence: inference relies on the fitted model and the bootstrap procedure; the dataset showed underdispersion relative to Poisson assumptions. While bootstrap CIs mitigate reliance on analytic SEs, model misspecification could still bias point estimates and uncertainty.\n",
+    "Small cell counts: several SpillType × Period × rurality cells have small n; estimates and CIs for those cells are less stable and should be interpreted cautiously.\n",
+    "Multiple comparisons: many contrasts were evaluated and reported p‑values are unadjusted; emphasize effect sizes with CIs rather than lone p‑values in policy reporting.\n",
+    "Reproducibility\n",
+    "\n",
+    "All code used Python (pandas/geopandas, statsmodels, patsy, numpy, matplotlib). Key outputs (predicted means, bootstrap draws, contrast tables, and figure) were saved to CSV/PNG in the analysis folder for inclusion in supplementary materials."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1f786261",
+   "metadata": {},
+   "source": [
+    "## narrative\n",
+    "\n",
+    "In the analytic sample (N = 11,376; after IQR outlier trimming), predicted reporting delays declined after 2020 across spill types and rurality. Using a Poisson GLM with a full C(spill_type) × C(Period) × C(rurality) interaction and parametric bootstrap (B = 2,000) for uncertainty, the estimated reduction in predicted delay (Before 2020 − 2020 and After) ranged from about 0.22–0.46 days (≈5–11 hours). For example, Historical spills in urban areas show a median reduction of 0.460 days (95% bootstrap CI 0.404–0.517, p < 0.001), and Recent spills in urban areas show a median reduction of 0.299 days (95% CI 0.222–0.375, p < 0.001).\n",
+    "\n",
+    "Comparing spill types, Recent spills tend to have longer predicted delays than Historical spills, with the largest and most robust gap in urban settings. After 2020, the Recent − Historical difference in Urban areas is 0.291 days (95% CI 0.244–0.341, p < 0.001), equivalent to roughly 7.0 hours. Differences by spill type are smaller and often not statistically distinguishable in some Suburban and Rural cells before 2020 (e.g., Before 2020 Suburban: 0.020 days, 95% CI −0.222–0.244, p ≈ 0.85).\n",
+    "\n",
+    "Limitations: the primary inference is model‑dependent (Poisson fit showed underdispersion, Pearson χ2/df ≈ 0.69), so results rely on the bootstrap procedures used to quantify uncertainty; a nonparametric case bootstrap (B = 1,000) produced broadly similar group CIs. Small sample sizes in some SpillType×Period×rurality cells limit precision for those specific comparisons, and reported p‑values are unadjusted for multiple contrasts; therefore emphasize effect sizes and bootstrap CIs in policy interpretation.\n",
+    "\n",
+    "Figure caption Figure X. Predicted reporting delays (days) for each Spill Type × Period × rurality cell with parametric bootstrap (B = 2,000) 95% confidence intervals. Points show bootstrap medians of the model‑predicted group means; vertical bars are 2.5th–97.5th percentiles from the bootstrap distribution."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8817d09c",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "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": "3.13.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/analysis/new analysis Aug 2025/analysis10-2021.ipynb b/analysis/new analysis Aug 2025/analysis10-2021.ipynb
index 4ea297a..8cd8668 100644
--- a/analysis/new analysis Aug 2025/analysis10-2021.ipynb	
+++ b/analysis/new analysis Aug 2025/analysis10-2021.ipynb	
@@ -79,6 +79,14 @@
     "spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b8da71fc",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
   {
    "cell_type": "code",
    "execution_count": 32,
diff --git a/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Historical.png b/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Historical.png
new file mode 100644
index 0000000..cf539a1
Binary files /dev/null and b/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Historical.png differ
diff --git a/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Recent.png b/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Recent.png
new file mode 100644
index 0000000..adbd216
Binary files /dev/null and b/analysis/new analysis Aug 2025/bar_by_rurality_spilltype_Recent.png differ
diff --git a/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Rural.png b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Rural.png
new file mode 100644
index 0000000..0417e48
Binary files /dev/null and b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Rural.png differ
diff --git a/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Suburban.png b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Suburban.png
new file mode 100644
index 0000000..41f9e80
Binary files /dev/null and b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Suburban.png differ
diff --git a/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Urban.png b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Urban.png
new file mode 100644
index 0000000..f713d17
Binary files /dev/null and b/analysis/new analysis Aug 2025/bar_by_spilltype_rurality_Urban.png differ
diff --git a/analysis/new analysis Aug 2025/pct_change_period_boot.csv b/analysis/new analysis Aug 2025/pct_change_period_boot.csv
new file mode 100644
index 0000000..a0c7e2c
--- /dev/null
+++ b/analysis/new analysis Aug 2025/pct_change_period_boot.csv	
@@ -0,0 +1,7 @@
+spill_type,rurality,median_days,ci_lower_days,ci_upper_days,median_hours
+Historical,Rural,0.27497054916409897,0.14521079279143942,0.402150850941176,6.599293179938375
+Historical,Suburban,0.35862785862785695,0.14230969134815072,0.5790820406205027,8.607068607068566
+Historical,Urban,0.46022418017894584,0.4043955927758729,0.5168359271165858,11.045380324294701
+Recent,Rural,0.22033304102748014,0.1595946164017345,0.27709844118835425,5.287992984659523
+Recent,Suburban,0.23838516812023614,0.09530264384838895,0.38765632713509646,5.721244034885667
+Recent,Urban,0.29898723668702976,0.2219819347255294,0.37490452624458753,7.175693680488714
diff --git a/analysis/new analysis Aug 2025/pct_change_spilltype_boot.csv b/analysis/new analysis Aug 2025/pct_change_spilltype_boot.csv
new file mode 100644
index 0000000..39de4f2
--- /dev/null
+++ b/analysis/new analysis Aug 2025/pct_change_spilltype_boot.csv	
@@ -0,0 +1,7 @@
+Period,rurality,median_abs_days,ci_lower_days,ci_upper_days,median_hours,median_pct
+2020 and After,Rural,0.07551270918427522,0.0019442568352317586,0.1495096699149845,1.8123050204226052,12.423566035505765
+2020 and After,Suburban,0.1443185237187848,0.014505398990406747,0.2621640256744544,3.4636445692508353,34.6698709840626
+2020 and After,Urban,0.2910867314901773,0.24390264079546417,0.3413553243429466,6.986081555764255,84.40551641150363
+Before 2020,Rural,0.019445736196048558,-0.09807034245553696,0.13342012441761195,0.4666976687051654,2.20523886072146
+Before 2020,Suburban,0.020467069010756067,-0.2216216216216229,0.24354828129099937,0.4912096562581456,2.635228848820833
+Before 2020,Urban,0.12910256593388225,0.04986927894511647,0.21108430823584592,3.098461582413174,16.054023961309728
diff --git a/analysis/new analysis Aug 2025/poisson_boot_contrasts.csv b/analysis/new analysis Aug 2025/poisson_boot_contrasts.csv
new file mode 100644
index 0000000..fbddbff
--- /dev/null
+++ b/analysis/new analysis Aug 2025/poisson_boot_contrasts.csv	
@@ -0,0 +1,13 @@
+contrast_type,spill_type,Period,rurality,comparison,median,ci_lower,ci_upper,p_value
+Period,Historical,,Rural,Before - After,0.27497054916409897,0.14521079279143942,0.402150850941176,0.0
+Period,Historical,,Suburban,Before - After,0.35862785862785695,0.14230969134815072,0.5790820406205027,0.001
+Period,Historical,,Urban,Before - After,0.46022418017894584,0.4043955927758729,0.5168359271165858,0.0
+Period,Recent,,Rural,Before - After,0.22033304102748014,0.1595946164017345,0.27709844118835425,0.0
+Period,Recent,,Suburban,Before - After,0.23838516812023614,0.09530264384838895,0.38765632713509646,0.0
+Period,Recent,,Urban,Before - After,0.29898723668702976,0.2219819347255294,0.37490452624458753,0.0
+SpillType,,2020 and After,Rural,Recent - Historical,0.07551270918427522,0.0019442568352317586,0.1495096699149845,0.045
+SpillType,,2020 and After,Suburban,Recent - Historical,0.1443185237187848,0.014505398990406747,0.2621640256744544,0.029
+SpillType,,2020 and After,Urban,Recent - Historical,0.2910867314901773,0.24390264079546417,0.3413553243429466,0.0
+SpillType,,Before 2020,Rural,Recent - Historical,0.019445736196048558,-0.09807034245553696,0.13342012441761195,0.753
+SpillType,,Before 2020,Suburban,Recent - Historical,0.020467069010756067,-0.2216216216216229,0.24354828129099937,0.85
+SpillType,,Before 2020,Urban,Recent - Historical,0.12910256593388225,0.04986927894511647,0.21108430823584592,0.002
diff --git a/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_all.csv b/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_all.csv
new file mode 100644
index 0000000..83b7b2d
--- /dev/null
+++ b/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_all.csv	
@@ -0,0 +1,13 @@
+spill_type,Period,rurality,pred_mean,ci_lower,ci_upper,n,pred_median_boot,ci_lower_boot,ci_upper_boot,pred_median_npboot,ci_lower_npboot,ci_upper_npboot
+Historical,2020 and After,Rural,0.6093189964157697,0.5478743705834819,0.6776546948120998,558,0.6075268817204316,0.5448028673835136,0.6774193548387089,0.6075165926329025,0.5526279441698162,0.6636686256781194
+Historical,2020 and After,Suburban,0.42011834319526453,0.3329283283724966,0.5301423977705431,169,0.4201183431952644,0.32529585798816385,0.5207100591715994,0.41614617682429955,0.3313609467455611,0.5102654352880386
+Historical,2020 and After,Urban,0.3444994290064705,0.3227699462514921,0.36769178160507127,2627,0.34526075371145737,0.32279215835553743,0.3681004948610584,0.34424658365839533,0.32394868445108016,0.36391238591979386
+Historical,Before 2020,Rural,0.8881118881118876,0.7853397269730921,1.0043331550865073,286,0.8846153846153819,0.779720279720278,0.9965034965034927,0.8875968992248069,0.8113663607417112,0.9601403594732165
+Historical,Before 2020,Suburban,0.7837837837837845,0.6059349205795946,1.0138333323565736,74,0.7837837837837822,0.5945945945945937,0.9864864864864882,0.7867759562841541,0.5972170398009939,1.0002941176470606
+Historical,Before 2020,Urban,0.8045417680454156,0.7560010921988318,0.8561991023677035,1233,0.8049472830494726,0.7534468775344682,0.8564476885644735,0.8034984820307399,0.7665794550369249,0.842415697211154
+Recent,2020 and After,Rural,0.6839481555333987,0.6486978802574822,0.7211139325318346,2006,0.683948155533399,0.6500498504486558,0.7208374875373882,0.6829934642365323,0.6533815453602346,0.7123480478360265
+Recent,2020 and After,Suburban,0.5621468926553657,0.489224984417794,0.6459382472016476,354,0.5621468926553669,0.4858757062146898,0.6384180790960449,0.560344827586207,0.49106618332711593,0.6334355283035621
+Recent,2020 and After,Urban,0.636728147554129,0.5939439753667028,0.6825942356556498,1247,0.6355252606255006,0.5942261427425811,0.6784482758620706,0.6360308612737718,0.5983955375909462,0.6742169727961493
+Recent,Before 2020,Rural,0.9033877038895861,0.8579117829215046,0.9512741983327778,1594,0.9040150564617306,0.8588299874529501,0.9498117942283538,0.9049383902302881,0.8640685429903382,0.9398919791087684
+Recent,Before 2020,Suburban,0.800970873786409,0.687621671804964,0.9330048294873011,206,0.8009708737864045,0.6794902912621323,0.9271844660194164,0.8032885081913201,0.7055557852948506,0.8968659324716435
+Recent,Before 2020,Urban,0.9354207436399179,0.8779639126262511,0.9966377376655905,1022,0.9344422700587136,0.8757093933463747,0.9970890410958914,0.9360885621808561,0.8894442568574094,0.9794730421614359
diff --git a/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_final.csv b/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_final.csv
new file mode 100644
index 0000000..ea88106
--- /dev/null
+++ b/analysis/new analysis Aug 2025/poisson_predicted_groups_boot_final.csv	
@@ -0,0 +1,13 @@
+spill_type,Period,rurality,pred_mean,ci_lower,ci_upper,n,pred_median_boot,ci_lower_boot,ci_upper_boot
+Historical,2020 and After,Rural,0.6093189964157697,0.5478743705834819,0.6776546948120998,558,0.6075268817204316,0.5448028673835136,0.6774193548387089
+Historical,2020 and After,Suburban,0.42011834319526453,0.3329283283724966,0.5301423977705431,169,0.4201183431952644,0.32529585798816385,0.5207100591715994
+Historical,2020 and After,Urban,0.3444994290064705,0.3227699462514921,0.36769178160507127,2627,0.34526075371145737,0.32279215835553743,0.3681004948610584
+Historical,Before 2020,Rural,0.8881118881118876,0.7853397269730921,1.0043331550865073,286,0.8846153846153819,0.779720279720278,0.9965034965034927
+Historical,Before 2020,Suburban,0.7837837837837845,0.6059349205795946,1.0138333323565736,74,0.7837837837837822,0.5945945945945937,0.9864864864864882
+Historical,Before 2020,Urban,0.8045417680454156,0.7560010921988318,0.8561991023677035,1233,0.8049472830494726,0.7534468775344682,0.8564476885644735
+Recent,2020 and After,Rural,0.6839481555333987,0.6486978802574822,0.7211139325318346,2006,0.683948155533399,0.6500498504486558,0.7208374875373882
+Recent,2020 and After,Suburban,0.5621468926553657,0.489224984417794,0.6459382472016476,354,0.5621468926553669,0.4858757062146898,0.6384180790960449
+Recent,2020 and After,Urban,0.636728147554129,0.5939439753667028,0.6825942356556498,1247,0.6355252606255006,0.5942261427425811,0.6784482758620706
+Recent,Before 2020,Rural,0.9033877038895861,0.8579117829215046,0.9512741983327778,1594,0.9040150564617306,0.8588299874529501,0.9498117942283538
+Recent,Before 2020,Suburban,0.800970873786409,0.687621671804964,0.9330048294873011,206,0.8009708737864045,0.6794902912621323,0.9271844660194164
+Recent,Before 2020,Urban,0.9354207436399179,0.8779639126262511,0.9966377376655905,1022,0.9344422700587136,0.8757093933463747,0.9970890410958914
diff --git a/analysis/new analysis Aug 2025/predicted_means_boot_CI.png b/analysis/new analysis Aug 2025/predicted_means_boot_CI.png
new file mode 100644
index 0000000..e814cc2
Binary files /dev/null and b/analysis/new analysis Aug 2025/predicted_means_boot_CI.png differ
diff --git a/analysis/new analysis Aug 2025/preds_npboot.npy b/analysis/new analysis Aug 2025/preds_npboot.npy
new file mode 100644
index 0000000..982b421
Binary files /dev/null and b/analysis/new analysis Aug 2025/preds_npboot.npy differ