snow_fall_insights.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyze\n",
    "\n",
    "Aim to determine whether early or late snow changes more year to year or place to place"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:05.007382Z",
     "start_time": "2020-03-25T14:08:04.488108Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:14.922484Z",
     "start_time": "2020-03-25T14:08:05.007382Z"
    }
   },
   "outputs": [],
   "source": [
    "from pyspark import SparkContext\n",
    "sc = SparkContext(master=\"local[2]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:15.058302Z",
     "start_time": "2020-03-25T14:08:14.922484Z"
    }
   },
   "outputs": [],
   "source": [
    "from pyspark.sql import SQLContext\n",
    "from pyspark.sql import *\n",
    "\n",
    "sqlContext = SQLContext(sc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:15.788748Z",
     "start_time": "2020-03-25T14:08:15.058302Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['SNOW_s20', 'TMIN_s20', 'TOBS', 'SNWD_s20', 'PRCP_s20', 'TOBS_s20', 'PRCP', 'TMAX', 'TMAX_s20', 'TMIN', 'SNOW', 'SNWD'])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "STAT, STAT_descr = pickle.load(open(\"../Data/STAT_NY.pickle\", \"rb\"))\n",
    "STAT.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:20.804111Z",
     "start_time": "2020-03-25T14:08:15.788748Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total count :  8174\n",
      "+-----------+-----------+--------------------+----+------------------+------------------+-------------------+------------------+------------------+------------------+----------------+------------------+------------------+---------+------------------+------------------+------------------+------------------+-------------------+------------------+---------+-----+\n",
      "|    Station|Measurement|              Values|Year|           coeff_1|           coeff_2|            coeff_3|           coeff_4|           coeff_5|        dist_coast|       elevation|          latitude|         longitude|     name|             res_1|             res_2|             res_3|             res_4|              res_5|          res_mean|total_var|State|\n",
      "+-----------+-----------+--------------------+----+------------------+------------------+-------------------+------------------+------------------+------------------+----------------+------------------+------------------+---------+------------------+------------------+------------------+------------------+-------------------+------------------+---------+-----+\n",
      "|USW00014735|       SNWD|[C0 54 60 56 00 0...|1939|-659.3190458468584|-169.3223185674074| 114.03551919851718|457.63475616405776|418.85486776821057|175.44400024414062|95.0999984741211|42.743099212646484|-73.80919647216797|ALBANY AP|0.6936982956503619|0.6717348260864184| 0.663298809472818|0.5188018093532125|0.39955117079145136|0.9805876688449685|1466850.0|   NY|\n",
      "|USW00014735|       SNWD|[40 4E 40 4E 40 4...|1943|-636.3811845052854| 272.3540924503061|-209.93059130386774|277.24524994684697| 5.927093765926917|175.44400024414062|95.0999984741211|42.743099212646484|-73.80919647216797|ALBANY AP|0.6164763435258529|0.5478257025034862|0.5066355807977382|0.4339928689629195|0.43395199595466527|0.7359716199890467|1441498.0|   NY|\n",
      "|USW00014735|       SNWD|[00 00 00 00 00 0...|1945|-33.31222533832221| 790.5793887906391|  347.2104137516154| 379.6550684921963| 329.8271722153101|175.44400024414062|95.0999984741211|42.743099212646484|-73.80919647216797|ALBANY AP|0.9993030174452017| 0.606768192055135|0.5333738807913538|0.4449806880212799|0.37717054459006166| 0.408684109819488|4001471.0|   NY|\n",
      "+-----------+-----------+--------------------+----+------------------+------------------+-------------------+------------------+------------------+------------------+----------------+------------------+------------------+---------+------------------+------------------+------------------+------------------+-------------------+------------------+---------+-----+\n",
      "only showing top 3 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "state, measure = \"NY\", \"SNWD\"\n",
    "\n",
    "df = sqlContext.read.parquet(\"../Data/decon_NY_SNWD.parquet/\")\n",
    "print(\"Total count : \", df.count())\n",
    "df.show(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:20.987831Z",
     "start_time": "2020-03-25T14:08:20.804111Z"
    }
   },
   "outputs": [],
   "source": [
    "feature = \"coeff_2\"\n",
    "sqlContext.registerDataFrameAsTable(df, \"weather\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:21.331372Z",
     "start_time": "2020-03-25T14:08:20.987831Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------+----+-------------------+\n",
      "|    station|year|            coeff_2|\n",
      "+-----------+----+-------------------+\n",
      "|USW00014735|1939| -169.3223185674074|\n",
      "|USW00014735|1943|  272.3540924503061|\n",
      "|USW00014735|1945|  790.5793887906391|\n",
      "|USW00014735|1947|-216.30283228614437|\n",
      "|USW00014735|1948|  1028.612178846539|\n",
      "+-----------+----+-------------------+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "query = \"\"\"\n",
    "SELECT station, year, coeff_2\n",
    "FROM weather\n",
    "\"\"\"\n",
    "\n",
    "df2 = sqlContext.sql(query)\n",
    "df2.show(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:21.666928Z",
     "start_time": "2020-03-25T14:08:21.335367Z"
    }
   },
   "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>station</th>\n",
       "      <th>year</th>\n",
       "      <th>coeff_2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>USW00014735</td>\n",
       "      <td>1939</td>\n",
       "      <td>-169.322319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>USW00014735</td>\n",
       "      <td>1943</td>\n",
       "      <td>272.354092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>USW00014735</td>\n",
       "      <td>1945</td>\n",
       "      <td>790.579389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>USW00014735</td>\n",
       "      <td>1947</td>\n",
       "      <td>-216.302832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>USW00014735</td>\n",
       "      <td>1948</td>\n",
       "      <td>1028.612179</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       station  year      coeff_2\n",
       "0  USW00014735  1939  -169.322319\n",
       "1  USW00014735  1943   272.354092\n",
       "2  USW00014735  1945   790.579389\n",
       "3  USW00014735  1947  -216.302832\n",
       "4  USW00014735  1948  1028.612179"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pdf = df2.toPandas()\n",
    "pdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:21.714859Z",
     "start_time": "2020-03-25T14:08:21.666928Z"
    }
   },
   "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>station</th>\n",
       "      <th>USC00300015</th>\n",
       "      <th>USC00300023</th>\n",
       "      <th>USC00300047</th>\n",
       "      <th>USC00300055</th>\n",
       "      <th>USC00300063</th>\n",
       "      <th>USC00300077</th>\n",
       "      <th>USC00300085</th>\n",
       "      <th>USC00300090</th>\n",
       "      <th>USC00300093</th>\n",
       "      <th>USC00300159</th>\n",
       "      <th>...</th>\n",
       "      <th>USW00014786</th>\n",
       "      <th>USW00014797</th>\n",
       "      <th>USW00014798</th>\n",
       "      <th>USW00094704</th>\n",
       "      <th>USW00094725</th>\n",
       "      <th>USW00094728</th>\n",
       "      <th>USW00094745</th>\n",
       "      <th>USW00094789</th>\n",
       "      <th>USW00094790</th>\n",
       "      <th>USW00094794</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>870.773867</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-31.805303</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-120.642691</td>\n",
       "      <td>472.534062</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>907.925667</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>806.258001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>NaN</td>\n",
       "      <td>149.013363</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.977231</td>\n",
       "      <td>NaN</td>\n",
       "      <td>97.643753</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 329 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "station  USC00300015  USC00300023  USC00300047  USC00300055  USC00300063  \\\n",
       "year                                                                       \n",
       "2009             NaN          NaN          NaN          NaN          NaN   \n",
       "2010             NaN          NaN          NaN          NaN          NaN   \n",
       "2011             NaN          NaN          NaN  -120.642691   472.534062   \n",
       "2012             NaN          NaN          NaN          NaN          NaN   \n",
       "2013             NaN   149.013363          NaN          NaN     0.977231   \n",
       "\n",
       "station  USC00300077  USC00300085  USC00300090  USC00300093  USC00300159  ...  \\\n",
       "year                                                                      ...   \n",
       "2009             NaN          NaN          NaN   870.773867          NaN  ...   \n",
       "2010             NaN          NaN          NaN   -31.805303          NaN  ...   \n",
       "2011             NaN          NaN          NaN   907.925667          NaN  ...   \n",
       "2012             NaN          NaN          NaN          NaN          NaN  ...   \n",
       "2013             NaN    97.643753          NaN          NaN          NaN  ...   \n",
       "\n",
       "station  USW00014786  USW00014797  USW00014798  USW00094704  USW00094725  \\\n",
       "year                                                                       \n",
       "2009             NaN          NaN          NaN          NaN          NaN   \n",
       "2010             NaN          NaN          NaN          NaN          NaN   \n",
       "2011             NaN          NaN          NaN          NaN          NaN   \n",
       "2012             NaN          NaN          NaN          NaN          NaN   \n",
       "2013             NaN          NaN          NaN          NaN          NaN   \n",
       "\n",
       "station  USW00094728  USW00094745  USW00094789  USW00094790  USW00094794  \n",
       "year                                                                      \n",
       "2009             NaN          NaN          NaN          NaN          NaN  \n",
       "2010             NaN          NaN          NaN          NaN          NaN  \n",
       "2011      806.258001          NaN          NaN          NaN          NaN  \n",
       "2012             NaN          NaN          NaN          NaN          NaN  \n",
       "2013             NaN          NaN          NaN          NaN          NaN  \n",
       "\n",
       "[5 rows x 329 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "year_station_df = pdf.pivot(index=\"year\", columns=\"station\", values=\"coeff_2\")\n",
    "year_station_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.042424Z",
     "start_time": "2020-03-25T14:08:21.718855Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'number of stations')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "null_stations = pd.isnull(year_station_df).mean()\n",
    "null_stations.hist()\n",
    "plt.xlabel(\"fraction of years that are undefined\")\n",
    "plt.ylabel(\"number of stations\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.186231Z",
     "start_time": "2020-03-25T14:08:22.042424Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "null_years = pd.isnull(year_station_df).mean(axis=1)\n",
    "null_years.plot()\n",
    "plt.xlabel(\"years\")\n",
    "plt.ylabel(\"fraction of stations that are undefined\")\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.218225Z",
     "start_time": "2020-03-25T14:08:22.186231Z"
    }
   },
   "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>station</th>\n",
       "      <th>USC00300023</th>\n",
       "      <th>USC00300055</th>\n",
       "      <th>USC00300063</th>\n",
       "      <th>USC00300077</th>\n",
       "      <th>USC00300085</th>\n",
       "      <th>USC00300093</th>\n",
       "      <th>USC00300183</th>\n",
       "      <th>USC00300220</th>\n",
       "      <th>USC00300321</th>\n",
       "      <th>USC00300331</th>\n",
       "      <th>...</th>\n",
       "      <th>USW00014768</th>\n",
       "      <th>USW00014771</th>\n",
       "      <th>USW00014786</th>\n",
       "      <th>USW00014798</th>\n",
       "      <th>USW00094725</th>\n",
       "      <th>USW00094728</th>\n",
       "      <th>USW00094745</th>\n",
       "      <th>USW00094789</th>\n",
       "      <th>USW00094790</th>\n",
       "      <th>USW00094794</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>870.773867</td>\n",
       "      <td>576.598705</td>\n",
       "      <td>NaN</td>\n",
       "      <td>580.459651</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>627.812099</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-31.805303</td>\n",
       "      <td>-25.393217</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-188.964850</td>\n",
       "      <td>-531.836350</td>\n",
       "      <td>...</td>\n",
       "      <td>-177.099611</td>\n",
       "      <td>156.548016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>NaN</td>\n",
       "      <td>-120.642691</td>\n",
       "      <td>472.534062</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>907.925667</td>\n",
       "      <td>305.092413</td>\n",
       "      <td>NaN</td>\n",
       "      <td>561.781691</td>\n",
       "      <td>566.875416</td>\n",
       "      <td>...</td>\n",
       "      <td>283.597976</td>\n",
       "      <td>584.134272</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>806.258001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>149.013363</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.977231</td>\n",
       "      <td>NaN</td>\n",
       "      <td>97.643753</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>170.083352</td>\n",
       "      <td>43.373130</td>\n",
       "      <td>...</td>\n",
       "      <td>15.194207</td>\n",
       "      <td>86.764841</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 268 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "station  USC00300023  USC00300055  USC00300063  USC00300077  USC00300085  \\\n",
       "year                                                                       \n",
       "2009             NaN          NaN          NaN          NaN          NaN   \n",
       "2010             NaN          NaN          NaN          NaN          NaN   \n",
       "2011             NaN  -120.642691   472.534062          NaN          NaN   \n",
       "2012             NaN          NaN          NaN          NaN          NaN   \n",
       "2013      149.013363          NaN     0.977231          NaN    97.643753   \n",
       "\n",
       "station  USC00300093  USC00300183  USC00300220  USC00300321  USC00300331  ...  \\\n",
       "year                                                                      ...   \n",
       "2009      870.773867   576.598705          NaN   580.459651          NaN  ...   \n",
       "2010      -31.805303   -25.393217          NaN  -188.964850  -531.836350  ...   \n",
       "2011      907.925667   305.092413          NaN   561.781691   566.875416  ...   \n",
       "2012             NaN          NaN          NaN          NaN          NaN  ...   \n",
       "2013             NaN          NaN          NaN   170.083352    43.373130  ...   \n",
       "\n",
       "station  USW00014768  USW00014771  USW00014786  USW00014798  USW00094725  \\\n",
       "year                                                                       \n",
       "2009             NaN   627.812099          NaN          NaN          NaN   \n",
       "2010     -177.099611   156.548016          NaN          NaN          NaN   \n",
       "2011      283.597976   584.134272          NaN          NaN          NaN   \n",
       "2012             NaN          NaN          NaN          NaN          NaN   \n",
       "2013       15.194207    86.764841          NaN          NaN          NaN   \n",
       "\n",
       "station  USW00094728  USW00094745  USW00094789  USW00094790  USW00094794  \n",
       "year                                                                      \n",
       "2009             NaN          NaN          NaN          NaN          NaN  \n",
       "2010             NaN          NaN          NaN          NaN          NaN  \n",
       "2011      806.258001          NaN          NaN          NaN          NaN  \n",
       "2012             NaN          NaN          NaN          NaN          NaN  \n",
       "2013             NaN          NaN          NaN          NaN          NaN  \n",
       "\n",
       "[5 rows x 268 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# based on above observations, range of dataset > 1960\n",
    "\n",
    "pdf2 = pdf[pdf[\"year\"] > 1960]\n",
    "year_station_df = pdf2.pivot(index=\"year\", columns=\"station\", values=\"coeff_2\")\n",
    "year_station_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.386101Z",
     "start_time": "2020-03-25T14:08:22.218225Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'number of stations')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAdvUlEQVR4nO3dfZQdVZnv8e8PCPLSmBACvTIBDQK+INFoehwcuNotyhIchTWCiqgJMmYtmQsKeMd4xxFm1LlhvKjjoGJGlOggDSJILoiCMQ0CBklDoHkRUYgIZBJBEmlAMPDcP2o3nnROd1efzj4np+v3Weusrtqnqvaz65w8qbOrapciAjMzq47tWh2AmZk1lxO/mVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxWRN/JJOlXSnpDskXShpJ0n7SrpJ0r2SLpK0Y84YzMxsc8p1Hb+kWcD1wIER8ZSki4EfAEcCl0ZEr6Rzgdsi4qujbWvGjBkxe/bshuJ44okn2HXXXRtat5253dXidldL2Xb39/c/EhF7Di/fIUtUm29/Z0l/AnYB1gJvAt6b3l8KnAmMmvhnz57NqlWrGgqgr6+P7u7uhtZtZ253tbjd1VK23ZJ+U688W1dPRDwE/F/gAYqEvxHoBzZExKa02IPArFwxmJnZlnJ29ewOfA94N7AB+G6aPyMi9k/L7AP8ICLm1Fl/IbAQoLOzc15vb29DcQwODtLR0dHQuu3M7a4Wt7tayra7p6enPyK6hpfn7Op5M3B/RPwOQNKlwF8D0yTtkI769wYerrdyRCwBlgB0dXVFoz/n/FOwWtzuanG7G5Pzqp4HgIMl7SJJwGHAXcAK4Ji0zHzg8owxmJnZMDn7+G8CLgFuAQZSXUuAjwOnSfoVsAdwXq4YzMxsS1mv6omIM4AzhhXfB7wuZ71mZjYy37lrZlYxTvxmZhXjxG9mVjG579w1M2t7sxdd2ZJ61yx+W5bt+ojfzKxinPjNzCrGid/MrGKc+M3MKsaJ38ysYpz4zcwqxonfzKxinPjNzCrGid/MrGKc+M3MKsaJ38ysYpz4zcwqxonfzKxinPjNzComW+KX9DJJq2tef5D0UUnTJV0j6d70d/dcMZiZ2ZZyPmz9noiYGxFzgXnAk8BlwCJgeUQcACxP82Zm1iTN6uo5DPh1RPwGOApYmsqXAkc3KQYzM6N5if89wIVpujMi1gKkv3s1KQYzMwMUEXkrkHYEHgZeGRHrJG2IiGk17z8WEVv080taCCwE6OzsnNfb29tQ/YODg3R0dDQWfBtzu6vF7c5r4KGN2euoZ86sqXXLy7a7p6enPyK6hpc345m7RwC3RMS6NL9O0syIWCtpJrC+3koRsQRYAtDV1RXd3d0NVd7X10ej67Yzt7ta3O68FrTqmbvHd9ctn2i7m9HVcxx/7uYBWAbMT9PzgcubEIOZmSVZE7+kXYC3AJfWFC8G3iLp3vTe4pwxmJnZ5rJ29UTEk8Aew8oepbjKx8zMWsB37pqZVYwTv5lZxTjxm5lVjBO/mVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxTjxm5lVjBO/mVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxTjxm5lVjBO/mVnFOPGbmVWME7+ZWcXkfubuNEmXSPqFpLslvV7SdEnXSLo3/d09ZwxmZra53Ef8/w78MCJeDrwauBtYBCyPiAOA5WnezMyaJFvil/RC4A3AeQAR8UxEbACOApamxZYCR+eKwczMtqSIyLNhaS6wBLiL4mi/H/gI8FBETKtZ7rGI2KK7R9JCYCFAZ2fnvN7e3obiGBwcpKOjo6F125nbXS1ud14DD23MXkc9c2ZNrVtett09PT39EdE1vDxn4u8CVgKHRMRNkv4d+ANwcpnEX6urqytWrVrVUBx9fX10d3c3tG47c7urxe3Oa/aiK7PXUc+axW+rW1623ZLqJv6cffwPAg9GxE1p/hLgtcA6STNTUDOB9RljMDOzYbIl/oj4b+C3kl6Wig6j6PZZBsxPZfOBy3PFYGZmW9oh8/ZPBi6QtCNwH3ACxX82F0s6EXgAODZzDGZmViNr4o+I1cAW/UsUR/9mZtYCvnPXzKxinPjNzCpmzMQv6SOSXqjCeZJukXR4M4IzM7Otr8wR/wcj4g/A4cCeFCdoF2eNyszMsimT+JX+Hgl8MyJuqykzM7M2Uybx90u6miLx/0jSbsBzecMyM7NcylzOeSIwF7gvIp6UtAdFd4+ZmbWhMRN/RDwnaR1woKTcN3yZmVlmYyZySWcB76YYbuHZVBzAdRnjMjOzTMocwR8NvCwins4djJmZ5Vfm5O59wJTcgZiZWXOUOeJ/ElgtaTnw/FF/RJySLSozM8umTOJfll5mZjYJlLmqZ2kaVvmlqeieiPhT3rDMzCyXMlf1dFM8FH0NxR27+0iaHxG+qsfMrA2V6eo5Gzg8Iu4BkPRS4EJgXs7AzMwsjzJX9UwZSvoAEfFLfJWPmVnbKnPEv0rSecC30/zxQH++kMzMLKcyif/DwN8Dp1D08V8HfKXMxiWtAR6nuON3U0R0SZoOXATMpjhv8K6IeGy8gZuZWWPKXNXzNPD59GpET0Q8UjO/CFgeEYslLUrzH29w22ZmNk4jJn5JF0fEuyQNUIzNs5mIeFWDdR4FdKfppUAfTvxmZk2jiC1yevGGNDMi1kp6cb33I+I3Y25cuh94jOI/jq9FxBJJGyJiWs0yj0XE7nXWXQgsBOjs7JzX29tbqkHDDQ4O0tHR0dC67cztrha3O6+BhzZmr6OeObOm1i0v2+6enp7+iOgaXj7iEX9ErE2TJ0XEZkfkacTOMkfph0TEw5L2Aq6R9IsS6wzVvwRYAtDV1RXd3d1lV91MX18fja7bztzuanG781qw6MrsddSz5vjuuuUTbXeZyznfUqfsiDIbj4iH09/1wGXA64B1kmZC8asCWF8uVDMz2xpGTPySPpz6918m6faa1/3A7WNtWNKu6TGNSNqV4mHtd1CM+zM/LTYfuHyijTAzs/JGu6rnO8BVwP+huPJmyOMR8fsS2+4ELpM0VM93IuKHkm4GLpZ0IvAAcGxDkZuZWUNG6+PfCGwEjgNI/fQ7AR2SOiLigdE2HBH3Aa+uU/4ocNhEgjYzs8aN2ccv6e2S7gXuB66luOnqqsxxmZlZJmVO7n4GOBj4ZUTsS3G0fkPWqMzMLJsyif9PqXtmO0nbRcQKYG7muMzMLJMyY/VskNRBMUbPBZLWA5vyhmVmZrmUOeI/iuK5u6cCPwR+DfxNzqDMzCyfMon/UxHxXERsioilEfElPLaOmVnbynrnrpmZbXtGG53zw8BJwH6Sau/U3Q1f1WNm1rZy3rlrZmbboBG7eiJiY0SsAT4J/Hcahnlf4H2Spo20npmZbdvK9PF/D3hW0v7AeRTJ/ztZozIzs2zKJP7nImIT8LfAFyPiVGBm3rDMzCyXUnfuSjoO+ABwRSqbki8kMzPLqUziPwF4PfDZiLhf0r7Af+UNy8zMchlzyIaIuAs4pWb+fmBxzqDMzCyfMkf8ZmY2iTjxm5lVzGjP3P12+vuR5oVjZma5jXbEP0/Si4EPStpd0vTaV9kKJG0v6VZJV6T5fSXdJOleSRdJ2nGijTAzs/JGS/znUgzD/HKgf9hr1Tjq+Ahwd838WcAXIuIA4DHgxPEEbGZmEzPakA1fiohXAN+IiJdExL41r5eU2bikvYG3AV9P8wLeBFySFlkKHD2hFpiZ2bgoIsZeSHo18D/S7HURcftoy9esdwnFIG+7AR8DFgArI2L/9P4+wFURcVCddRcCCwE6Ozvn9fb2lqlyC4ODg3R0dDS0bjtzu6vF7c5r4KGN2euoZ86sqXXLy7a7p6enPyK6hpePeR2/pFMoEvClqegCSUsi4j/GWO9vgPUR0S+pe6i4zqJ1/+eJiCXAEoCurq7o7u6ut9iY+vr6aHTdduZ2V4vbndeCRVdmr6OeNcd31y2faLvLPHP374C/iognACSdBfwMGDXxA4cA75B0JLAT8ELgi8A0STuk8X/2Bh5uNHgzMxu/MolfwLM1889S/8h9MxHxCeATAOmI/2MRcbyk7wLHAL3AfODyccZsZhU1e9iR9+lzNrXsaLydlUn83wRuknRZmj+aYnjmRn0c6JX0GeDWCW7LzMzGqcxYPZ+X1AccSnGkf0JE3DqeSiKiD+hL0/cBrxtvoGZmtnWUOeInIm4Bbskci5mZNYHH6jEzqxgnfjOzihk18adxdn7crGDMzCy/URN/RDwLPCmp/u1jZmbWdsqc3P0jMCDpGuCJocKIOGXkVczMbFtVJvFfmV5mZjYJlLmOf6mknYEXRcQ9TYjJzMwyGvOqHklvB1ZTjM2PpLmSluUOzMzM8ihzOeeZFHfabgCIiNXAvhljMjOzjMok/k0RMXww6rEH8Tczs21SmZO7d0h6L7C9pAOAU4Ab84ZlZma5lDniPxl4JfA0cCHwB+CjOYMyM7N8ylzV8yTwj+kBLBERj+cPy8y2VcPHxLf2U+aqnr+UNADcTnEj122S5uUPzczMcijTx38ecFJE/BRA0qEUD2d5Vc7AzMwsjzJ9/I8PJX2AiLgecHePmVmbGvGIX9Jr0+TPJX2N4sRuAO8mPU3LzMzaz2hdPWcPmz+jZnrM6/gl7QRcB7wg1XNJRJwhaV+KB61Pp3iq1/sj4plxRW1mZg0bMfFHRM8Et/008KaIGJQ0Bbhe0lXAacAXIqJX0rnAicBXJ1iXmZmVNObJXUnTgA8As2uXH2tY5ogIYDDNTkmvAN4EvDeVL6UYEsKJ38ysSVTk51EWkG4EVgIDwHND5RGxdMyNS9sD/cD+wJeBzwErI2L/9P4+wFURcVCddRcCCwE6Ozvn9fb2lmzS5gYHB+no6Gho3XbmdldLM9s98NDwEVxap3NnWPdUq6PIZ86s+s/AKvt59/T09EdE1/DyMpdz7hQRp5VYbgvpCV5z06+Gy4BX1FtshHWXAEsAurq6oru7u5EQ6Ovro9F125nbXS3NbPeCbegGrtPnbOLsgTJprD2tOb67bvlEP+8yl3N+W9KHJM2UNH3oNZ5KImIDxZVABwPTJA19UnsDD48rYjMzm5Ayif8Zii6an1F02/QDq8ZaSdKe6Uif9CCXNwN3AyuAY9Ji84HLxx+2mZk1qsxvpNOA/SPikXFueyawNPXzbwdcHBFXSLoL6JX0GeBWijuDzcysScok/juBJ8e74Yi4HXhNnfL7KB7sYmZmLVAm8T8LrJa0guLafGDsyznNzGzbVCbxfz+9zMxsEigzHv+Y1+ubmVn7KHPn7v3UudY+Il6SJSIzM8uqTFdP7V1fOwHHUgywZmZmbahMV8+jw4q+KOl64FN5QjJrH618DOGaxW9rWd3W3sp09by2ZnY7il8Au2WLyMzMsirT1VM7Lv8mYA3wrizRmJlZdmW6eiY6Lr+ZmW1DynT1vAB4J1uOx/8v+cIyM7NcynT1XA5spBic7ekxljUzs21cmcS/d0S8NXskZmbWFGWGZb5R0pzskZiZWVOUOeI/FFiQ7uB9GhDFI3VflTUyMzPLokziPyJ7FGZm1jRlLuf8TTMCMTOz5pi8Tyk2m+Rqh4s4fc6mbeoh6LZtK3Ny18zMJpFsiV/SPpJWSLpb0p2SPpLKp0u6RtK96e/uuWIwM7Mt5Tzi3wScHhGvAA4G/l7SgcAiYHlEHAAsT/NmZtYk2RJ/RKyNiFvS9OPA3cAs4Chg6KleS4Gjc8VgZmZbUsQWD9fa+pVIs4HrgIOAByJiWs17j0XEFt09khYCCwE6Ozvn9fb2NlT34OAgHR0dDa3bztzu5hh4aGPT6hpN586w7qlWR9F8k73dc2ZNrVte9nve09PTHxFdw8uzJ35JHcC1wGcj4lJJG8ok/lpdXV2xatWqhurv6+uju7u7oXXbmdvdHK18EEut0+ds4uyB6l2kN9nbPdLDdsp+zyXVTfxZr+qRNAX4HnBBRFyaitdJmpnenwmszxmDmZltLudVPQLOA+6OiM/XvLUMmJ+m51OM/mlmZk2S8zfSIcD7gQFJq1PZ/wYWAxdLOhF4gOLh7WZm1iTZEn9EXE8xoFs9h+Wqd1vhh3A319D+9h2sZmPznbtmZhXjxG9mVjFO/GZmFePEb2ZWMU78ZmYV48RvZlYxTvxmZhUzeQe5sKbbVsatMbPR+YjfzKxinPjNzCrGid/MrGKc+M3MKsaJ38ysYpz4zcwqxonfzKxinPjNzCrGid/MrGJyPnP3G5LWS7qjpmy6pGsk3Zv+7p6rfjMzqy/nkA3nA+cA36opWwQsj4jFkhal+Y9njKGSZi+60o8gNLMRZTvij4jrgN8PKz4KWJqmlwJH56rfzMzqa3Yff2dErAVIf/dqcv1mZpWniMi3cWk2cEVEHJTmN0TEtJr3H4uIuv38khYCCwE6Ozvn9fb2NhTD4OAgHR0dDa07EQMPbWx6nbU6d4Z1T7U0hJZwu6tlsrd7zqypdcvL5rWenp7+iOgaXt7sYZnXSZoZEWslzQTWj7RgRCwBlgB0dXVFd3d3QxX29fXR6LoT0er+9dPnbOLsgeqNuu12V8tkb/ea47vrlk80rzW7q2cZMD9Nzwcub3L9ZmaVl+2/SkkXAt3ADEkPAmcAi4GLJZ0IPAAcm6v+IQMPbWz50beZ2bYkW+KPiONGeOuwXHWamdnYfOeumVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxTjxm5lVjBO/mVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxTjxm5lVjBO/mVnFOPGbmVWME7+ZWcU48ZuZVYwTv5lZxTjxm5lVTEsSv6S3SrpH0q8kLWpFDGZmVdX0xC9pe+DLwBHAgcBxkg5sdhxmZlXViiP+1wG/ioj7IuIZoBc4qgVxmJlVUisS/yzgtzXzD6YyMzNrgh1aUKfqlMUWC0kLgYVpdlDSPQ3WNwN4pMF129YpbneluN2Tk84a8a2y7X5xvcJWJP4HgX1q5vcGHh6+UEQsAZZMtDJJqyKia6LbaTdud7W43dUy0Xa3oqvnZuAASftK2hF4D7CsBXGYmVVS04/4I2KTpP8J/AjYHvhGRNzZ7DjMzKqqFV09RMQPgB80qboJdxe1Kbe7WtzuaplQuxWxxXlVMzObxDxkg5lZxUyaxD/WMBCSXiDpovT+TZJmNz/Kra9Eu0+TdJek2yUtl1T38q52U3bYD0nHSApJk+LKjzLtlvSu9JnfKek7zY4xhxLf8xdJWiHp1vRdP7IVcW5Nkr4hab2kO0Z4X5K+lPbJ7ZJeW3rjEdH2L4qTxL8GXgLsCNwGHDhsmZOAc9P0e4CLWh13k9rdA+ySpj9clXan5XYDrgNWAl2tjrtJn/cBwK3A7ml+r1bH3aR2LwE+nKYPBNa0Ou6t0O43AK8F7hjh/SOBqyjujToYuKnstifLEX+ZYSCOApam6UuAwyTVu5msnYzZ7ohYERFPptmVFPdNtLuyw358Gvg34I/NDC6jMu3+EPDliHgMICLWNznGHMq0O4AXpump1Lk3qN1ExHXA70dZ5CjgW1FYCUyTNLPMtidL4i8zDMTzy0TEJmAjsEdTostnvMNfnEhxhNDuxmy3pNcA+0TEFc0MLLMyn/dLgZdKukHSSklvbVp0+ZRp95nA+yQ9SHHF4MnNCa2lGh7+piWXc2ZQZhiIUkNFtJnSbZL0PqALeGPWiJpj1HZL2g74ArCgWQE1SZnPeweK7p5uil93P5V0UERsyBxbTmXafRxwfkScLen1wLdTu5/LH17LNJzTJssRf5lhIJ5fRtIOFD8HR/sZ1Q5KDX8h6c3APwLviIinmxRbTmO1ezfgIKBP0hqK/s9lk+AEb9nv+eUR8aeIuB+4h+I/gnZWpt0nAhcDRMTPgJ0oxrOZzEr9+69nsiT+MsNALAPmp+ljgJ9EOkPSxsZsd+ry+BpF0p8M/b0wRrsjYmNEzIiI2RExm+LcxjsiYlVrwt1qynzPv09xQh9JMyi6fu5rapRbX5l2PwAcBiDpFRSJ/3dNjbL5lgEfSFf3HAxsjIi1ZVacFF09McIwEJL+BVgVEcuA8yh+/v2K4kj/Pa2LeOso2e7PAR3Ad9O57Aci4h0tC3orKNnuSadku38EHC7pLuBZ4H9FxKOti3riSrb7dOA/JZ1K0d2xoN0P7CRdSNFlNyOduzgDmAIQEedSnMs4EvgV8CRwQultt/m+MTOzcZosXT1mZlaSE7+ZWcU48ZuZVYwTv5lZxTjxm5lVjBP/JCfpFEl3S7pgK2xrgaS/qJn/uqQDJ7rdUep7gaQfS1ot6d256hkPSdMknVQz3y1pXMNCDN+P2zJJa9L9AKMtc2z6jq2Q1CXpS1up7gWSztka27LNTYrr+G1UJwFHpLs4nydphzRm0XgsAO4g3R0YEX+3VSIc2WuAKRExN3M9QOl9Mo1in35lAlUtoGY/jleDn11OJwInRcSKNN/uN8pNej7in8QknUsxlO0ySadKOlPSEklXA9+SNFvSTyXdkl5/XbPuP0gakHSbpMWSjqEY6+eCdAS+s6S+oWEQJB2Xlr9D0lk12xmU9Nm0nZWSOuvEOV3S99OY4islvUrSXsB/AXNTffvVLL+fpFtq5g+Q1J+m50m6VlK/pB8NjVYo6UOSbk5xfE/SLqn8fEmfl7QCOEvSG1N9q1WM7b7bsHAXA/ul9z+XyjokXSLpF5IuULpTTtKnUp13pP2uevtx2L4oG+euKsZrvznFucXopMN/jUg6R9KCNL1G0j+nz31A0stT+R6Srk7b/Bo148FIep+kn6e4vyZpe0mfAg4FzpX0udo60/ftG+l7cp+kU0bbVio/QdIvJV0LHDK8TbaVtHrMab/yvoA1wIw0fSbQD+yc5ncBdkrTB1DcBQlwBHAjfx7Hf3r620fNuPZD88BfUNwyvyfFr8ifAEenZQJ4e5r+N+CTdWL8D+CMNP0mYHWa7gauGKFdK4C5afpfKUZjnJLi3jOVv5viLk+APWrW/Qxwcpo+H7gC2D7N/z/gkDTdAewwrN7Z1IyPnmLcSDFOynbAz4BDa/dbmv52zX7YbD8O237ZOP8VeF+angb8Eth12LY223/AORR3tA59L4a2fRLw9TT9JeBTafpt6fObAbwi7Zsp6b2vAB8Y3p7aOim+bzcCL0jbeDR9RnW3Bczkz9+jHYEbgHNa/W9oMr7c1VM9yyLiqTQ9BThH0lyK2/tfmsrfDHwz0jj+ETHWYHZ/CfRFxO8AVJxPeAPFuDHPUCQsKP7TeUud9Q8F3pnq+kk66pw6Rp1fB06QdBpFgn8d8DKKwdmuSQfd2wNDY5ccJOkzFEmyg+L2/yHfjYhn0/QNwOdTGy6NiAfHiAPg50PLSVpN8Z/D9UCPpH+g+A92OnAnRcIbTdk4DwfeIeljaX4n4EXA3SXiHXJp+tsP/G2afsPQdERcKemxVH4YMA+4Oe3bnYEyYz9dGcXAgE9LWg90jrKtv2Lz79FF/Pk7aVuRE3/1PFEzfSqwDng1xdHq0ANLxPiGrB7tgTZ/inRYR/GfS73vXCPDy36PYuySnwD9EfGoihOmd0bE6+ssfz7Fr5DbUndHd817z++TiFgs6UqKMVBWSnpzRPxijFhqRzx9FthB0k4UR7JdEfFbSWdSJOexlIqTYp+9MyLuGWVbm9i8O3d4/UNxD/9c6u17AUsj4hOj1FfPFvtmpG1JOnqEum0rcx9/tU0F1kYxZvn7KY6QAa4GPljTvzw9lT9OMeTxcDcBb5Q0I/XVHgdcO444rgOOT3V1A49ExB9GWyEi/khxNPxV4Jup+B5gTxXjsSNpiqRXpvd2A9ZKmjJUVz2S9ouIgYg4i+Ik5cuHLTLSPhhuKMk+IqmDYkTYMtsoFSdF20+uOZ/wmjrL/AY4UMXVUVNJo1eOofazOALYPZUvB45Rce5l6LxMo89vHmlbNwHd6RffFODYBrdvY3Dir7avAPMlraT4Sf0EQET8kGLI11Wp62KoO+F8ipN4m52UjGIo2E9Q9LvfBtwSEZePI44zgS5Jt1OcPJ0/+uLPu4DiCPHqFMczFAn2LEm3AauBoRPW/0SRWK4BRjuC/2g6GXsb8BTDnlgWxUiXN6RlPld3C8VyG4D/BAYourxurnn7fOrsx3HG+WmKrrrbVTyM+9N1YvgtxRj1t1Psq1tH2d6QfwbeoOLk+eEUfe5ExF3AJ4Gr0+d0DUWf/LiNtK30PTqT4jzJj4FbRtyITYhH57S2lfq3p0bEP7U6FrN24j5+a0uSLgP2o7gKyMzGwUf8ZmYV4z5+M7OKceI3M6sYJ34zs4px4jczqxgnfjOzinHiNzOrmP8Pey3O4cYIZxwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "null_stations = pd.isnull(year_station_df).mean()\n",
    "null_stations.hist()\n",
    "plt.xlabel(\"fraction of years that are undefined\")\n",
    "plt.ylabel(\"number of stations\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.537762Z",
     "start_time": "2020-03-25T14:08:22.386101Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "null_years = pd.isnull(year_station_df).mean(axis=1)\n",
    "null_years.plot()\n",
    "plt.xlabel(\"years\")\n",
    "plt.ylabel(\"fraction of stations that are undefined\")\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estimate effect of year vs effect of location"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:22.641659Z",
     "start_time": "2020-03-25T14:08:22.537762Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total RMS =  753.145449514645\n",
      "\n",
      "RMS removing mean-by-station =  701.5445524111517 \n",
      "\treduction =  51.60089710349325\n",
      "\n",
      "RMS removing mean-by-year =  531.4823870967629 \n",
      "\treduction= 221.6630624178821\n"
     ]
    }
   ],
   "source": [
    "def RMS(mat):\n",
    "    return np.sqrt(np.nanmean(mat ** 2))\n",
    "\n",
    "mean_by_year = np.nanmean(year_station_df, axis=1)\n",
    "mean_by_station = np.nanmean(year_station_df, axis=0)\n",
    "df_minus_year = (year_station_df.transpose() - mean_by_year).transpose()\n",
    "df_minus_station = year_station_df - mean_by_station\n",
    "\n",
    "print('total RMS = ', RMS(year_station_df))\n",
    "print('\\nRMS removing mean-by-station = ',RMS(df_minus_station),\n",
    "      '\\n\\treduction = ', RMS(year_station_df) - RMS(df_minus_station))\n",
    "print('\\nRMS removing mean-by-year = ', RMS(df_minus_year),\n",
    "      '\\n\\treduction=', RMS(year_station_df) - RMS(df_minus_year))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "The effect of time is about four times as large as the effect of the location."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Iterative Reduction\n",
    "\n",
    "Using alternating minimization, to remove the combined effect of location and time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:08:23.029108Z",
     "start_time": "2020-03-25T14:08:22.641659Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial RMS :  753.145449514645\n",
      "0 : after removing mean_by_year = 531.4823870967629\n",
      "0 : after removing mean_by_stations  = 490.58111029852466\n",
      "1 : after removing mean_by_year = 490.11149394795615\n",
      "1 : after removing mean_by_stations  = 490.08333935027764\n",
      "2 : after removing mean_by_year = 490.08010695597295\n",
      "2 : after removing mean_by_stations  = 490.0794679008307\n",
      "3 : after removing mean_by_year = 490.0793022334803\n",
      "3 : after removing mean_by_stations  = 490.0792548676868\n",
      "4 : after removing mean_by_year = 490.0792408655066\n",
      "4 : after removing mean_by_stations  = 490.0792366758466\n"
     ]
    }
   ],
   "source": [
    "print(\"Initial RMS : \", RMS(year_station_df))\n",
    "\n",
    "for i in range(5):\n",
    "    mean_by_year = np.nanmean(year_station_df, axis=1)\n",
    "    year_station_df = (year_station_df.transpose() - mean_by_year).transpose()\n",
    "    print(\"{} : after removing mean_by_year = {}\".format(i, RMS(year_station_df)))\n",
    "    \n",
    "    mean_by_station = np.nanmean(year_station_df, axis=0)\n",
    "    year_station_df = year_station_df - mean_by_station\n",
    "    print(\"{} : after removing mean_by_stations  = {}\".format(i, RMS(year_station_df)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:10:01.749351Z",
     "start_time": "2020-03-25T14:10:01.725390Z"
    }
   },
   "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>station</th>\n",
       "      <th>USC00300023</th>\n",
       "      <th>USC00300055</th>\n",
       "      <th>USC00300063</th>\n",
       "      <th>USC00300077</th>\n",
       "      <th>USC00300085</th>\n",
       "      <th>USC00300093</th>\n",
       "      <th>USC00300183</th>\n",
       "      <th>USC00300220</th>\n",
       "      <th>USC00300321</th>\n",
       "      <th>USC00300331</th>\n",
       "      <th>...</th>\n",
       "      <th>USW00014771</th>\n",
       "      <th>USW00014786</th>\n",
       "      <th>USW00014798</th>\n",
       "      <th>USW00094725</th>\n",
       "      <th>USW00094728</th>\n",
       "      <th>USW00094745</th>\n",
       "      <th>USW00094789</th>\n",
       "      <th>USW00094790</th>\n",
       "      <th>USW00094794</th>\n",
       "      <th>mean_by_year</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1961</th>\n",
       "      <td>155.322668</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-526.216490</td>\n",
       "      <td>810.626258</td>\n",
       "      <td>-279.421801</td>\n",
       "      <td>142.485364</td>\n",
       "      <td>-202.849172</td>\n",
       "      <td>125.324071</td>\n",
       "      <td>129.805635</td>\n",
       "      <td>...</td>\n",
       "      <td>173.516964</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.791976</td>\n",
       "      <td>-1149.083768</td>\n",
       "      <td>-59.314214</td>\n",
       "      <td>1018.590878</td>\n",
       "      <td>NaN</td>\n",
       "      <td>99.364223</td>\n",
       "      <td>84.199898</td>\n",
       "      <td>-0.115733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1962</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>194.606475</td>\n",
       "      <td>-162.816631</td>\n",
       "      <td>42.324686</td>\n",
       "      <td>274.137805</td>\n",
       "      <td>-166.422647</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>170.828388</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>958.883579</td>\n",
       "      <td>-127.318680</td>\n",
       "      <td>-0.142607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1963</th>\n",
       "      <td>132.330101</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>490.657507</td>\n",
       "      <td>542.845673</td>\n",
       "      <td>301.262188</td>\n",
       "      <td>188.568372</td>\n",
       "      <td>294.201819</td>\n",
       "      <td>197.149024</td>\n",
       "      <td>148.849368</td>\n",
       "      <td>...</td>\n",
       "      <td>230.112110</td>\n",
       "      <td>NaN</td>\n",
       "      <td>59.291399</td>\n",
       "      <td>-208.082968</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>624.245549</td>\n",
       "      <td>-5.480156</td>\n",
       "      <td>-0.143292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1964</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>243.330515</td>\n",
       "      <td>435.685952</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-121.596578</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>-294.370434</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-121.120404</td>\n",
       "      <td>-359.614187</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-242.508398</td>\n",
       "      <td>-0.149812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1965</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36.324879</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-732.826859</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>-69.694317</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>135.953933</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-128.899506</td>\n",
       "      <td>-0.130511</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 269 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "station  USC00300023  USC00300055  USC00300063  USC00300077  USC00300085  \\\n",
       "year                                                                       \n",
       "1961      155.322668          NaN          NaN  -526.216490   810.626258   \n",
       "1962             NaN          NaN          NaN   194.606475  -162.816631   \n",
       "1963      132.330101          NaN          NaN   490.657507   542.845673   \n",
       "1964             NaN          NaN          NaN          NaN   243.330515   \n",
       "1965             NaN          NaN          NaN          NaN    36.324879   \n",
       "\n",
       "station  USC00300093  USC00300183  USC00300220  USC00300321  USC00300331  ...  \\\n",
       "year                                                                      ...   \n",
       "1961     -279.421801   142.485364  -202.849172   125.324071   129.805635  ...   \n",
       "1962       42.324686   274.137805  -166.422647          NaN          NaN  ...   \n",
       "1963      301.262188   188.568372   294.201819   197.149024   148.849368  ...   \n",
       "1964      435.685952          NaN  -121.596578          NaN          NaN  ...   \n",
       "1965             NaN          NaN  -732.826859          NaN          NaN  ...   \n",
       "\n",
       "station  USW00014771  USW00014786  USW00014798  USW00094725  USW00094728  \\\n",
       "year                                                                       \n",
       "1961      173.516964          0.0    79.791976 -1149.083768   -59.314214   \n",
       "1962             NaN          NaN          NaN   170.828388          NaN   \n",
       "1963      230.112110          NaN    59.291399  -208.082968          NaN   \n",
       "1964     -294.370434          NaN  -121.120404  -359.614187          NaN   \n",
       "1965      -69.694317          NaN          NaN   135.953933          NaN   \n",
       "\n",
       "station  USW00094745  USW00094789  USW00094790  USW00094794  mean_by_year  \n",
       "year                                                                       \n",
       "1961     1018.590878          NaN    99.364223    84.199898     -0.115733  \n",
       "1962             NaN          NaN   958.883579  -127.318680     -0.142607  \n",
       "1963             NaN          NaN   624.245549    -5.480156     -0.143292  \n",
       "1964             NaN          NaN          NaN  -242.508398     -0.149812  \n",
       "1965             NaN          NaN          NaN  -128.899506     -0.130511  \n",
       "\n",
       "[5 rows x 269 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "year_station_df[\"mean_by_year\"] = mean_by_year\n",
    "year_station_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-25T14:11:34.520581Z",
     "start_time": "2020-03-25T14:11:34.328605Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[10, 6])\n",
    "plt.plot(year_station_df[\"mean_by_year\"])\n",
    "plt.title(\"snow fall has been getting early\")\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}