Catchment-selection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.lines import Line2D\n",
    "from trendmaster import trend\n",
    "import itertools\n",
    "from scipy.stats import linregress\n",
    "import datetime\n",
    "import pickle\n",
    "import descartes\n",
    "import geopandas as gpd\n",
    "from shapely.geometry import Point,Polygon"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Other"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def saveDict(dictionary,filename):\n",
    "    \"\"\"\n",
    "    Saves dictionary to pickle file in working directory.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    dictionary: dict\n",
    "    filename: str\n",
    "        filename without .pkl ending\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    Nothing\n",
    "    \"\"\"\n",
    "    f = open(f\"{filename}.pkl\",\"wb\")\n",
    "    pickle.dump(dictionary,f)\n",
    "    f.close()\n",
    "\n",
    "def openDict(filename):\n",
    "    \"\"\"\n",
    "    Opens dictionary from pickle file in working directory.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    filename: str\n",
    "        filename without .pkl ending\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    dictionary\n",
    "    \"\"\"\n",
    "    pickle_in = open(f\"{filename}.pkl\",\"rb\")\n",
    "    loadedDict = pickle.load(pickle_in)\n",
    "    return loadedDict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Metadata table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load table of metadata, i.e. final station selection of Vormoor et al.\n",
    "catchm = pd.read_csv(\"final_station_selection.txt\",header=0,sep=\"\\t\")\n",
    "df = pd.DataFrame(catchm)\n",
    "\n",
    "# adding information from NEVINA\n",
    "df.loc[df.snumber==200607,\"areal\"] = 126.41 #must be removed\n",
    "df.loc[df.snumber==200607,\"altitude\"] = 933 #must be removed\n",
    "df.loc[df.snumber==8400020,\"areal\"] = 5.14\n",
    "df.loc[df.snumber==8400020,\"altitude\"] = 673\n",
    "df.loc[df.snumber==2400008,\"areal\"] = 121\n",
    "df.loc[df.snumber==2400008,\"altitude\"] = 325 #Hmin change to H50!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "feltparam = pd.read_csv(\"feltparam.txt\",header=0,sep=\"\\t\", usecols=[0,2,7,12,17,20,21,24,25,26,27,28])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "df[\"BREPRO\"] = np.nan\n",
    "for c in df.snumber:\n",
    "    cpad = int(str(c)+\"000\")\n",
    "    val = float(feltparam.loc[feltparam.SNUMBER==cpad,\"BREPRO\"].mean())\n",
    "    df.loc[df.snumber==c,\"BREPRO\"] = val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regions ['ostlandet' 'sorlandet' 'vestlandet' 'trondelag' 'nordland' 'finnmark']\n",
      "Highest station: 1546.0\n",
      "Lowest station: 25.0\n",
      "Missing years: [0 1 3 4 2 6 5]\n",
      "Shortest timeseries from 1982\n",
      "Longest timeseries from 1960\n"
     ]
    }
   ],
   "source": [
    "# some stats\n",
    "runoffRegion = df[\"runoff.region\"].unique()\n",
    "print(\"Regions\",runoffRegion)\n",
    "regionNames = [\"Østlandet\",\"Sørlandet\",\"Vestlandet\",\"Trøndelag\",\"Nordland\",\"Finnmark\"]\n",
    "print(\"Highest station:\",df[\"altitude\"].max())\n",
    "print(\"Lowest station:\",df[\"altitude\"].min())\n",
    "print(\"Missing years:\",df[\"missing.yr\"].unique())\n",
    "print(\"Shortest timeseries from\",df[\"start.yr\"].max())\n",
    "print(\"Longest timeseries from\",df[\"start.yr\"].min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'Altitude')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[\"altitude\"].plot.hist()\n",
    "plt.close()\n",
    "\n",
    "plt.figure()\n",
    "df[\"missing.yr\"].plot.hist()\n",
    "plt.close()\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(df[\"lon\"],df[\"lat\"],\"x\")\n",
    "plt.close()\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(df[\"areal\"],df[\"altitude\"], \"x\")\n",
    "plt.xlabel(\"Area\")\n",
    "plt.ylabel(\"Altitude\")\n",
    "#plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Splitting by region\n",
    "Splitting by region and sorting\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def byRegion(data,region):\n",
    "    # select region from df\n",
    "    regionData = data[data[\"runoff.region\"]==region]\n",
    "    # sort by altitude, high to low\n",
    "    regionData = regionData.sort_values(\"altitude\",ascending=False)\n",
    "    # add station number according to altitude (for plotting later)\n",
    "    regionData[\"station.order\"]=np.arange(regionData.shape[0],0,-1)\n",
    "    return regionData"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "ostlandet = byRegion(df,\"ostlandet\")\n",
    "vestlandet = byRegion(df,\"vestlandet\")\n",
    "sorlandet = byRegion(df,\"sorlandet\")\n",
    "trondelag = byRegion(df,\"trondelag\")\n",
    "nordland = byRegion(df,\"nordland\")\n",
    "finnmark = byRegion(df,\"finnmark\")\n",
    "\n",
    "regionDF = [ostlandet,sorlandet,vestlandet,trondelag,nordland,finnmark]\n",
    "regions = dict(zip(runoffRegion,regionDF))\n",
    "regionLabels = dict(zip(runoffRegion,regionNames))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importing data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Organising data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def readRunoff(catchmentNo):\n",
    "    # reading data\n",
    "    runoff = pd.read_csv(f\"runoff/{catchmentNo}.txt\",\n",
    "                         header=None, sep=\" \", names=[\"date\",\"runoff\"],usecols=[0,1],index_col=0)\n",
    "    # converting indices to datetime objects\n",
    "    runoff.index = pd.to_datetime(runoff.index)\n",
    "    # replacing invalid values with nan\n",
    "    runoff = runoff.replace(-9999.,np.nan)\n",
    "    return runoff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def readSnow(regine,main):\n",
    "    # create filepath\n",
    "    file = f\"seNorge/{regine}.{main}/{regine}.{main}_SeNorge_qsw_1959_2014.dta\"\n",
    "    # read file\n",
    "    snow = pd.read_table(file,delim_whitespace=True,header=None,names=[\"year\",\"month\",\"day\",\"qsw\"])\n",
    "    # list of dates\n",
    "    index = list(np.arange(len(snow)))\n",
    "    for i in range(len(snow)):\n",
    "        index[i] = f\"{snow.year.iloc[i]}-{snow.month.iloc[i]}-{snow.day.iloc[i]}\"\n",
    "    # index as timestamp objects\n",
    "    snow.index = pd.to_datetime(index)\n",
    "    # removing unnecessary columns\n",
    "    snow = snow.drop([\"year\",\"month\",\"day\"],axis = 1)\n",
    "    return snow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def readSeNorge(regine,main):\n",
    "    # create filepath\n",
    "    file = f\"seNorge/{regine}.{main}/{regine}.{main}_SeNorge_rr_tm_1959_2014.dta\"\n",
    "    # read file\n",
    "    data = pd.read_table(file,delim_whitespace=True,header=None,names=[\"year\",\"month\",\"day\",\"precip\",\"temp\"])\n",
    "    # list of dates\n",
    "    index = list(np.arange(len(data)))\n",
    "    for i in range(len(data)):\n",
    "        index[i] = f\"{data.year[i]}-{data.month[i]}-{data.day[i]}\"\n",
    "    # index as timestamp objects\n",
    "    data.index = pd.to_datetime(index)\n",
    "    # removing unnecessary columns\n",
    "    data = data.drop([\"year\",\"month\",\"day\"],axis = 1)\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def organiseData(region):\n",
    "    \"\"\"\n",
    "    Organises all data for a region into a dictionary.\n",
    "    \n",
    "    Parameters\n",
    "    -----------\n",
    "    region: pandas.DataFrame\n",
    "    \n",
    "    Returns\n",
    "    -----------\n",
    "    dictionary with all variables\n",
    "    \"\"\"\n",
    "    # create dictionary and adds station order by elevation\n",
    "    d = {\"order\":list(region.snumber),\n",
    "         \"data\":{},\n",
    "         \"metadata\":region}\n",
    "    # add all catchments as keys\n",
    "    for i in range(region.shape[0]):\n",
    "        d[\"data\"][region.snumber.iloc[i]] = {}\n",
    "    \n",
    "    for c in d[\"order\"]:\n",
    "        # runoff data for each cathcment\n",
    "        d[\"data\"][c][\"runoff\"] = readRunoff(c)\n",
    "        # snowmelt    \n",
    "        regine = int(region[region.snumber == c].regine)\n",
    "        main = int(region[region.snumber == c].main)\n",
    "        d[\"data\"][c][\"snow\"] = readSnow(regine,main)\n",
    "        # precipitation and temperature\n",
    "        pt = readSeNorge(regine,main)\n",
    "        d[\"data\"][c][\"temp\"] = pt.temp\n",
    "        d[\"data\"][c][\"precip\"] = pt.precip\n",
    "        \n",
    "    # return the finished dictionary\n",
    "    return d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set this to True if importing and organising data again\n",
    "importing = False\n",
    "\n",
    "\n",
    "if importing:\n",
    "    # imports and organises data into dictionaries\n",
    "    ost = organiseData(ostlandet)\n",
    "    sor = organiseData(sorlandet)\n",
    "    nord = organiseData(nordland)\n",
    "    vest = organiseData(vestlandet)\n",
    "    trond = organiseData(trondelag)\n",
    "    finn = organiseData(finnmark)\n",
    "else:\n",
    "    # opening dictionaries with organised data from file\n",
    "    sor = openDict(\"sorlandet_final\")\n",
    "    ost = openDict(\"ostlandet_final\")\n",
    "    vest = openDict(\"vestlandet_final\")\n",
    "    trond = openDict(\"trondelag_final\")\n",
    "    nord = openDict(\"nordland_final\")\n",
    "    finn = openDict(\"finnmark_final\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualising data quality"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Runoff quality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Earliest start year: 1871\n",
      "Latest start year: 1982\n",
      "Latest end year 2014\n",
      "Earliest end year 2012\n"
     ]
    }
   ],
   "source": [
    "start = []\n",
    "end = []\n",
    "for station in ost[\"order\"]:\n",
    "    #print(\"Station:\",station)\n",
    "    #print(\"First year:\\t\",ost['data'][station]['runoff'].index.year[0])\n",
    "    start.append(ost['data'][station]['runoff'].index.year[0])\n",
    "    #print(\"Last year:\\t\",ost['data'][station]['runoff'].index.year[-1],\"\\n\")\n",
    "    end.append(ost['data'][station]['runoff'].index.year[-1])\n",
    "\n",
    "for station in sor[\"order\"]:\n",
    "    start.append(sor['data'][station]['runoff'].index.year[0])\n",
    "    end.append(sor['data'][station]['runoff'].index.year[-1])\n",
    "    \n",
    "for station in vest[\"order\"]:\n",
    "    start.append(vest['data'][station]['runoff'].index.year[0])\n",
    "    end.append(vest['data'][station]['runoff'].index.year[-1])\n",
    "    \n",
    "for station in nord[\"order\"]:\n",
    "    start.append(nord['data'][station]['runoff'].index.year[0])\n",
    "    end.append(nord['data'][station]['runoff'].index.year[-1])\n",
    "\n",
    "for station in trond[\"order\"]:\n",
    "    start.append(trond['data'][station]['runoff'].index.year[0])\n",
    "    end.append(trond['data'][station]['runoff'].index.year[-1])\n",
    "\n",
    "for station in finn[\"order\"]:\n",
    "    start.append(finn['data'][station]['runoff'].index.year[0])\n",
    "    end.append(finn['data'][station]['runoff'].index.year[-1])\n",
    "\n",
    "print(\"Earliest start year:\",np.array(start).min())\n",
    "print(\"Latest start year:\",np.array(start).max())\n",
    "print(\"Latest end year\",np.array(end).max())\n",
    "print(\"Earliest end year\",np.array(end).min())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Missing days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def missingDays(df):\n",
    "    a = np.full(144,np.nan)\n",
    "    years = pd.unique(df.index.year)\n",
    "    for year in years:\n",
    "        part = df[str(year)]\n",
    "        a[year-1871] = part.isnull().sum()[0]\n",
    "    return a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def missingDaysRegion(data):\n",
    "    \"\"\"\n",
    "    Makes an array of missing days per year per region\n",
    "    Paramters\n",
    "    ---------\n",
    "    data: dictionary\n",
    "        a dictionary which contains data for a runoff region, especially: \n",
    "            data[\"order\"]: the station order \n",
    "            data[\"data\"][stationnumber][\"runoff\"]: runoff for specific station\n",
    "            \n",
    "    Returns\n",
    "    -------\n",
    "    array of days with missing runoff per year\n",
    "    \"\"\"\n",
    "    a = np.full((len(data[\"order\"]),144),np.nan)\n",
    "    for i in range(len(data[\"order\"])):\n",
    "        a[i] = missingDays(data[\"data\"][data[\"order\"][i]][\"runoff\"])\n",
    "    \n",
    "    return a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Missing years"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def missingYears(df):\n",
    "    \"\"\"\n",
    "    Finds gaps of 5 or more values in each year of timeseries. \n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    array of years with 3 possible values:\n",
    "        NaN: no data availiable for year\n",
    "        0: no data gaps in year\n",
    "        1: one or more gaps of 5 or more missing values found\n",
    "    \"\"\"\n",
    "    a = np.full(144,np.nan)\n",
    "    years = pd.unique(df.index.year)\n",
    "    for year in years:\n",
    "        runoff = df[str(year)].runoff\n",
    "        n = runoff.isnull().astype(int).groupby(runoff.notnull().astype(int).cumsum()).sum()\n",
    "        m = n>0 \n",
    "        if m.any():\n",
    "            a[year-1871] = 1\n",
    "        else:\n",
    "            a[year-1871] = 0\n",
    "    return a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def missingYearsRegion(data):\n",
    "    \"\"\"\n",
    "    Makes an array of missing days per year per region\n",
    "    Paramters\n",
    "    ---------\n",
    "    data: dictionary\n",
    "        a dictionary which contains data for a runoff region, especially: \n",
    "            data[\"order\"]: the station order \n",
    "            data[\"data\"][stationnumber][\"runoff\"]: runoff for specific station\n",
    "            \n",
    "    Returns\n",
    "    -------\n",
    "    array of days with missing runoff per year\n",
    "    \"\"\"\n",
    "    a = np.full((len(data[\"order\"]),144),np.nan)\n",
    "    for i in range(len(data[\"order\"])):\n",
    "        a[i] = missingYears(data[\"data\"][data[\"order\"][i]][\"runoff\"])\n",
    "    \n",
    "    return a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Summary\n",
    "\n",
    "Data is availiable for all catchments from 1982-2012\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## seNorge"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Appears that files contain *rainfall* and not *precipitation* data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def findMissingSeNorge(data):\n",
    "    \"\"\"\n",
    "    Finds the catchment with missing/problematic seNorge data\n",
    "    \"\"\"\n",
    "    missing = []\n",
    "    for c in data[\"order\"]:\n",
    "        if all(data[\"data\"][c][\"precip\"]==0):\n",
    "            missing.append(c)\n",
    "    return missing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def findMissingDays(data,startYear=1983,endYear=2012):\n",
    "    \"\"\"\n",
    "    Identifies catchments which have more than 10% missing data for the chosen period.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    data: dictionary with organised data for a runoff region\n",
    "    startYear: beginning of time period\n",
    "    endYear: end of time period\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    List of catchment numbers\n",
    "    \"\"\"\n",
    "    start = startYear-1871\n",
    "    end = endYear-1871\n",
    "    period = endYear - startYear\n",
    "    \n",
    "    arr = missingDaysRegion(data)[:,start:end]\n",
    "    \n",
    "    \n",
    "    # an nan signifies that no data is availiable for that year\n",
    "    if np.isnan(arr).any():\n",
    "        print(\"Warning: Chosen period contains years with no records\")\n",
    "    \n",
    "    # sum of missing days per catchment\n",
    "    b = arr.sum(axis=1)\n",
    "    # identifying catchments with more than 10% missing\n",
    "    c = (b/(365*period))>0.1\n",
    "    \n",
    "    return list(np.array(data[\"order\"])[c])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Updating metadata table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the functions to update metadata table with new values (only for the last 30 years?)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def updateMeta(data,startYear=1983,endYear=2012):\n",
    "    # resetting indeces\n",
    "    data[\"metadata\"].reset_index(drop=True,inplace=True)\n",
    "    \n",
    "    # seNorge\n",
    "    data[\"metadata\"][\"seNorge\"] = 1\n",
    "    for i in data[\"metadata\"].index:\n",
    "        if data[\"metadata\"].snumber[i] in findMissingSeNorge(data):\n",
    "            data[\"metadata\"].loc[i,\"seNorge\"] = 0\n",
    "    \n",
    "    # start and end years\n",
    "    start = startYear-1871\n",
    "    end = endYear-1871\n",
    "    data[\"metadata\"][\"missingData30yr\"] = np.nan\n",
    "    for i in data[\"metadata\"].index:\n",
    "        s = data[\"metadata\"].snumber[i]\n",
    "        \n",
    "        # missing data in 30 years\n",
    "        missing = missingDays(data[\"data\"][s][\"runoff\"])[start:end].sum()\n",
    "        data[\"metadata\"].loc[i,\"missingData30yr\"] = missing/(30*365)\n",
    "    \n",
    "        # updating start and end year\n",
    "        data[\"metadata\"].loc[i,\"start.yr\"] = data[\"data\"][s][\"runoff\"].index.year.min()\n",
    "        data[\"metadata\"].loc[i,\"end.yr\"] = data[\"data\"][s][\"runoff\"].index.year.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "updateMeta(sor)\n",
    "updateMeta(ost)\n",
    "updateMeta(vest)\n",
    "updateMeta(trond)\n",
    "updateMeta(nord)\n",
    "updateMeta(finn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Final catchment selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Additional catchments may have to be rejected because of permission rights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def recordsAvailiable(data,startYear,endYear):\n",
    "    \"\"\"\n",
    "    Returns list of catchments where runoff records are availiable for the chosen period.\n",
    "    \"\"\"\n",
    "    mask = (data[\"metadata\"][\"start.yr\"]<=startYear) & (data[\"metadata\"][\"end.yr\"]>=endYear)\n",
    "    qualified = list(data[\"metadata\"].snumber[mask])\n",
    "    return qualified"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dataSelection(data,years=30,startYear=1983,endYear=2012):\n",
    "    \"\"\"\n",
    "    Adds the final data selection to the organised data dictionary.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    data: dictionary of all catchment data\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    Nothing\n",
    "    \"\"\"\n",
    "    \n",
    "    # requirement 1: seNorge data is availiable\n",
    "    r1 = findMissingSeNorge(data)\n",
    "    # requirement 2: no more than 10 % missing data in period\n",
    "    r2 = findMissingDays(data,startYear=startYear,endYear=endYear)\n",
    "    \n",
    "    # disqualified catchments\n",
    "    disqual = r1 + r2\n",
    "    \n",
    "    # checking for availiable records in period\n",
    "    availiable = recordsAvailiable(data,startYear,endYear)\n",
    "    \n",
    "    # making copy of original list\n",
    "    catchments = data[\"order\"].copy()\n",
    "    \n",
    "    # removing disqualified catchments from list\n",
    "    for c in data[\"order\"]:\n",
    "        if (c not in availiable) or (c in disqual):\n",
    "            catchments.remove(c)\n",
    "        \n",
    "    # adding new list to dictionary\n",
    "    data[f\"final{years}\"] = catchments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: Chosen period contains years with no records\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/skam2905/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:29: RuntimeWarning: invalid value encountered in greater\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: Chosen period contains years with no records\n",
      "Warning: Chosen period contains years with no records\n",
      "Warning: Chosen period contains years with no records\n",
      "Warning: Chosen period contains years with no records\n",
      "Warning: Chosen period contains years with no records\n"
     ]
    }
   ],
   "source": [
    "# 30 years\n",
    "dataSelection(ost)\n",
    "dataSelection(vest)\n",
    "dataSelection(sor)\n",
    "dataSelection(trond)\n",
    "dataSelection(nord)\n",
    "dataSelection(finn)\n",
    "\n",
    "# 50 years\n",
    "dataSelection(ost,years=50,startYear=1963)\n",
    "dataSelection(vest,years=50,startYear=1963)\n",
    "dataSelection(sor,years=50,startYear=1963)\n",
    "dataSelection(trond,years=50,startYear=1963)\n",
    "dataSelection(nord,years=50,startYear=1963)\n",
    "dataSelection(finn,years=50,startYear=1963)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 30 years\n",
    "finalSel = ost[\"final30\"] + sor[\"final30\"] + vest[\"final30\"] + trond[\"final30\"] + nord[\"final30\"] + finn[\"final30\"]\n",
    "newFinalSelection30 = df[df.snumber.isin(finalSel)]\n",
    "newFinalSelection30.to_csv(\"newFinalSelection_30years.csv\")\n",
    "removedCatchm30 = df[~df.snumber.isin(finalSel)]\n",
    "\n",
    "# 50 years\n",
    "finalSel = ost[\"final50\"] + sor[\"final50\"] + vest[\"final50\"] + trond[\"final50\"] + nord[\"final50\"] + finn[\"final50\"]\n",
    "newFinalSelection50 = df[df.snumber.isin(finalSel)]\n",
    "newFinalSelection50.to_csv(\"newFinalSelection_50years.csv\")\n",
    "# catchments removed from selection\n",
    "removedCatchm50 = df[~df.snumber.isin(finalSel)]\n",
    "\n",
    "finalSelectionList = {\"ost\":{30:ost[\"final30\"],50:ost[\"final50\"]},\n",
    "                      \"vest\":{30:vest[\"final30\"],50:vest[\"final50\"]},\n",
    "                      \"sor\":{30:sor[\"final30\"],50:sor[\"final50\"]},\n",
    "                      \"trond\":{30:trond[\"final30\"],50:trond[\"final50\"]},\n",
    "                      \"nord\":{30:nord[\"final30\"],50:nord[\"final50\"]},\n",
    "                      \"finn\":{30:finn[\"final30\"],50:finn[\"final50\"]}}\n",
    "saveDict(finalSelectionList,\"finalSelectionList\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Catchments in recent trend analysis: 207\n",
      "Catchments in long-term trend analysis: 107\n"
     ]
    }
   ],
   "source": [
    "print(\"Catchments in recent trend analysis:\",newFinalSelection30.shape[0])\n",
    "print(\"Catchments in long-term trend analysis:\",newFinalSelection50.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# saving dictionaries with organised data to file\n",
    "saveDict(sor,\"sorlandet_final\")\n",
    "saveDict(ost,\"ostlandet_final\")\n",
    "saveDict(vest,\"vestlandet_final\")\n",
    "saveDict(trond,\"trondelag_final\")\n",
    "saveDict(nord,\"nordland_final\")\n",
    "saveDict(finn,\"finnmark_final\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "frames = [finn[\"metadata\"],\n",
    "          nord[\"metadata\"],\n",
    "          trond[\"metadata\"],\n",
    "          vest[\"metadata\"],\n",
    "          ost[\"metadata\"],\n",
    "          sor[\"metadata\"]]\n",
    "          \n",
    "pd.concat(frames).to_csv(\"updated_stationselection.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Plot final catchment selection including disqualified cathcments in grey: *Already plotted, see runoffLength2 plots*\n",
    "- Visualise some of the metadata of the final catchment selection, e.g. area and altitude\n",
    "- Maps of locations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = {\"finn\":finn,\"nord\":nord,\"trond\":trond,\"vest\":vest,\"ost\":ost,\"sor\":sor}\n",
    "regionLabels = {'ost': 'Østlandet',\n",
    "                'sor': 'Sørlandet',\n",
    "                'vest': 'Vestlandet',\n",
    "                'trond': 'Trøndelag',\n",
    "                'nord': 'Nordland',\n",
    "                'finn': 'Finnmark'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# with subplotting version 2\n",
    "fig,ax = plt.subplots(nrows = 3,ncols = 2,sharex = True,sharey=True,figsize=(10,7))\n",
    "plt.xlim(0,70)\n",
    "\n",
    "for i in range(len(data.keys())):\n",
    "    reg = list(data.keys())[i]\n",
    "    df = data[reg]\n",
    "    dfr30 = df[\"metadata\"][df[\"metadata\"].snumber.isin(df[f\"final30\"])]\n",
    "    dfr50 = df[\"metadata\"][df[\"metadata\"].snumber.isin(df[f\"final50\"])]\n",
    "    rm = df[\"metadata\"][~df[\"metadata\"].snumber.isin(df[f\"final50\"])]\n",
    "    if i < 3:\n",
    "        ax[i][0].bar(rm[\"station.order\"],rm[\"altitude\"],width=1,color=\"lightgrey\",label=\"Eliminated\")\n",
    "        ax[i][0].bar(dfr30[\"station.order\"],dfr30[\"altitude\"],width=1,color=\"k\",label=\"Selected 30year\")\n",
    "        ax[i][0].bar(dfr50[\"station.order\"],dfr50[\"altitude\"],width=0.4,color=\"c\",label=\"Selected 50year\")\n",
    "        ax[i][0].set_title(regionLabels[reg])\n",
    "    else:\n",
    "        ax[i-3][1].bar(rm[\"station.order\"],rm[\"altitude\"],width=1,color=\"lightgrey\",label=\"Eliminated\")\n",
    "        ax[i-3][1].bar(dfr30[\"station.order\"],dfr30[\"altitude\"],width=1,color=\"k\",label=\"Recent\")\n",
    "        ax[i-3][1].bar(dfr50[\"station.order\"],dfr50[\"altitude\"],width=0.4,color=\"c\",label=\"Long-term\")\n",
    "        ax[i-3][1].set_title(regionLabels[reg])\n",
    "        \n",
    "fig.text(0.5, 0.04, \"Station number\", ha=\"center\",fontsize=12)\n",
    "fig.text(0.04, 0.5, \"Median altitude of catchment in $m.a.s.l.$\", va=\"center\", rotation=\"vertical\",fontsize=12)\n",
    "plt.legend()\n",
    "fig.savefig(f\"Plots/elevation_selectedStations2_final\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "marker = itertools.cycle((\"v\", \"x\", \"+\", \"D\", \"^\",\".\"))\n",
    "for label, dfr in newFinalSelection30.groupby(\"runoff.region\"):\n",
    "    plt.figure()\n",
    "    plt.hist(dfr.altitude,color=\"grey\")\n",
    "    plt.title(label)\n",
    "    plt.xlabel(\"Elevation in $m$\")\n",
    "    plt.ylabel(\"Frequency\")\n",
    "    plt.savefig(f\"Plots/final_hist_{label}_altitude.png\",dpi=400,bbox_inches='tight')\n",
    "    plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'finnmark'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-34-20a7b1b83592>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdfr\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnewFinalSelection30\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroupby\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"runoff.region\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m     \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdfr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mareal\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdfr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maltitude\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\".\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mregionLabels\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mylabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Elevation in $m$\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxlabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Area\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlegend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'finnmark'"
     ]
    }
   ],
   "source": [
    "for label, dfr in newFinalSelection30.groupby(\"runoff.region\"):\n",
    "    plt.plot(dfr.areal,dfr.altitude,\".\",marker=next(marker),label=regionLabels[label])\n",
    "plt.ylabel(\"Elevation in $m$\")\n",
    "plt.xlabel(\"Area\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/final_scatter_altitude_area_30year.png\",dpi=400,bbox_inches='tight')\n",
    "\n",
    "plt.figure()\n",
    "for label, dfr in newFinalSelection50.groupby(\"runoff.region\"):\n",
    "    plt.plot(dfr.areal,dfr.altitude,\".\",marker=next(marker),label=regionLabels[label])\n",
    "plt.ylabel(\"Elevation in $m$\")\n",
    "plt.xlabel(\"Area\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/final_scatter_altitude_area_50year.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'finnmark'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-35-46b5a3527dee>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfigure\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfigsize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnewFinalSelection30\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroupby\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"runoff.region\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m     \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maltitude\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\".\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mregionLabels\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolors\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxlabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Elevation in $m$\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mylabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Latitude\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'finnmark'"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 360x720 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = itertools.cycle((\"c\", \"darkslategrey\"))\n",
    "plt.figure(figsize=(5,10))\n",
    "for label, df in newFinalSelection30.groupby(\"runoff.region\"):\n",
    "    plt.plot(df.altitude,df.lat,\".\",marker=next(marker),label=regionLabels[label],color=next(colors))\n",
    "plt.xlabel(\"Elevation in $m$\")\n",
    "plt.ylabel(\"Latitude\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/final_scatter_lat_altitude_30year.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'finnmark'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-36-ac9d14013b05>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnewFinalSelection30\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroupby\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"runoff.region\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m     \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mareal\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\".\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmarker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mregionLabels\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxlabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Area\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mylabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Latitude\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlegend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'finnmark'"
     ]
    }
   ],
   "source": [
    "for label, df in newFinalSelection30.groupby(\"runoff.region\"):\n",
    "    plt.plot(df.areal,df.lat,\".\",marker=next(marker),label=regionLabels[label])\n",
    "plt.xlabel(\"Area\")\n",
    "plt.ylabel(\"Latitude\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/final_scatter_lat_area_30year.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plotRunoffLengthFinal(data,title,filename):\n",
    "    cmap=plt.cm.get_cmap(\"RdYlGn_r\")\n",
    "    a = missingDaysRegion(data)\n",
    "    plt.figure(figsize=(15,a.shape[0]/10))\n",
    "    # if more than 30 days are missing, they are red\n",
    "    plt.imshow(a,cmap=cmap,vmin=0,vmax=366,aspect=1.5)\n",
    "    plt.xlabel(\"Years\")\n",
    "    plt.xticks(range(-1,145,10),range(1870,2014,10))\n",
    "    plt.ylabel(\"Station ordered by altitude\")\n",
    "    cbar = plt.colorbar(shrink=1/(a.shape[0])*20, label=\"Missing days per year\")\n",
    "    cbar.ax.invert_yaxis()\n",
    "    plt.title(title)\n",
    "    for i in range(len(data[\"order\"])):\n",
    "        if data[\"order\"][i] not in data[\"final50\"]:\n",
    "            plt.axhline(i,color=\"lightgrey\",xmin=(1963-1870)/144,xmax=(1982-1870)/144,linewidth=3,label=\"Removed from 30 year analysis\")\n",
    "    for i in range(len(data[\"order\"])):\n",
    "        if data[\"order\"][i] not in data[\"final30\"]:\n",
    "            plt.axhline(i,color=\"grey\",xmin=(1983-1870)/144,linewidth=3)\n",
    "    plt.savefig(f\"Plots/runoffLengthFinal_{filename}.png\",dpi=300,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "legend_elements = [Line2D([0], [0], color=\"grey\",label=\"Recent\"),\n",
    "                       Line2D([0], [0], color=\"lightgrey\",label=\"Long-term analysis\")]\n",
    "plt.legend(handles=legend_elements, loc=\"center\",title=\"Cathcment eliminated from:                       \",ncol=2)\n",
    "plt.savefig(f\"Plots/runoffLengthFinal_legend.png\",dpi=300,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "plotRunoffLengthFinal(sor,\"Sørlandet\",\"sorlandet\")\n",
    "plotRunoffLengthFinal(ost,\"Østlandet\",\"ostlandet\")\n",
    "plotRunoffLengthFinal(vest,\"Vestlandet\",\"vestlandet\")\n",
    "plotRunoffLengthFinal(nord,\"Nordland\",\"nordland\")\n",
    "plotRunoffLengthFinal(trond,\"Trøndelag\",\"trondelag\")\n",
    "plotRunoffLengthFinal(finn,\"Finnmark\",\"finnmark\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "norge = gpd.read_file(\"gis/Norge.shp\")\n",
    "regionLines = gpd.read_file(\"gis/Grense_Avrenningsregioner.shp\")\n",
    "stations = gpd.read_file(\"gis/Gauging_Stations.shp\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "regine = list(newFinalSelection30.regine)\n",
    "main = list(newFinalSelection30.main)\n",
    "station30 = []\n",
    "for i in range(len(regine)):\n",
    "    station30.append(f\"{str(regine[i])}.{str(main[i])}\")\n",
    "\n",
    "regine = list(removedCatchm30.regine)\n",
    "main = list(removedCatchm30.main)\n",
    "removed30 = []\n",
    "for i in range(len(regine)):\n",
    "    removed30.append(f\"{str(regine[i])}.{str(main[i])}\")\n",
    "    \n",
    "regine = list(newFinalSelection50.regine)\n",
    "main = list(newFinalSelection50.main)\n",
    "station50 = []\n",
    "for i in range(len(regine)):\n",
    "    station50.append(f\"{str(regine[i])}.{str(main[i])}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask = stations.STASJON_NR.isin(station30)\n",
    "stations30 = stations[mask]\n",
    "mask = stations.STASJON_NR.isin(removed30)\n",
    "rstations30 = stations[mask]\n",
    "mask = stations.STASJON_NR.isin(station50)\n",
    "stations50 = stations[mask]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig,ax = plt.subplots(figsize=(30,10))\n",
    "norge.plot(ax=ax,color=\"white\", edgecolor=\"lightgrey\")\n",
    "regionLines.plot(ax=ax,color=\"darksalmon\",label=\"Runoff region\")\n",
    "rstations30.plot(ax=ax,color=\"grey\",marker=\"x\",markersize=20,label=\"Removed\")\n",
    "stations30.plot(ax=ax,marker=\"v\",markersize=40,color=\"k\",label=\"Recent\",alpha=.8)\n",
    "stations50.plot(ax=ax,marker=\"^\",markersize=40,color=\"c\",label=\"Long-term\",alpha=.8)\n",
    "plt.setp(ax.get_xticklabels(), visible=False)\n",
    "plt.setp(ax.get_yticklabels(), visible=False)\n",
    "plt.setp(ax.get_xticklines(), visible=False)\n",
    "plt.setp(ax.get_yticklines(), visible=False)\n",
    "plt.setp(ax.spines.values(), visible=False)\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.savefig(\"Plots/final&removed_map.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Moving average"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Window comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "catchm = trond[\"final30\"][5]\n",
    "runoff = trond[\"data\"][catchm][\"runoff\"]\n",
    "\n",
    "# moving average with different window sizes\n",
    "runoff5day = runoff.rolling(5,center=True).mean()\n",
    "runoff10day = runoff.rolling(10,center=True).mean()\n",
    "runoff15day = runoff.rolling(15,center=True).mean()\n",
    "runoff20day = runoff.rolling(20,center=True).mean()\n",
    "runoff30day = runoff.rolling(30,center=True).mean()\n",
    "\n",
    "line = itertools.cycle((\"--\",\"-.\",\":\"))\n",
    "colors = itertools.cycle((\"c\",\"salmon\",\"olive\"))\n",
    "\n",
    "# plotting\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.plot(runoff.runoff[\"2013\"],\"darkslategray\",label=\"Original\")\n",
    "plt.plot(runoff5day.runoff[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"5day\")\n",
    "plt.plot(runoff10day.runoff[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"10day\")\n",
    "#plt.plot(runoff15day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"15day\")\n",
    "#plt.plot(runoff20day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"20day\")\n",
    "plt.plot(runoff30day.runoff[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"30day\")\n",
    "plt.ylabel(\"Mean daily streamflow ($m^3 \\ s^{-1}$)\")\n",
    "plt.title(f\"Catchment: {catchm} (Trøndelag)\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/movingAverage_example_streamflow.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "catchm = trond[\"final30\"][5]\n",
    "runoff = trond[\"data\"][catchm][\"snow\"]\n",
    "\n",
    "# moving average with different window sizes\n",
    "runoff5day = runoff.rolling(5,center=True).mean()\n",
    "runoff10day = runoff.rolling(10,center=True).mean()\n",
    "runoff15day = runoff.rolling(15,center=True).mean()\n",
    "runoff20day = runoff.rolling(20,center=True).mean()\n",
    "runoff30day = runoff.rolling(30,center=True).mean()\n",
    "\n",
    "line = itertools.cycle((\"--\",\"-.\",\":\"))\n",
    "colors = itertools.cycle((\"c\",\"salmon\",\"olive\"))\n",
    "\n",
    "# plotting\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.plot(runoff[\"2013\"],\"darkslategray\",label=\"Original\")\n",
    "plt.plot(runoff5day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"5day\")\n",
    "plt.plot(runoff10day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"10day\")\n",
    "#plt.plot(runoff15day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"15day\")\n",
    "#plt.plot(runoff20day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"20day\")\n",
    "plt.plot(runoff30day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"30day\")\n",
    "plt.ylabel(\"Daily snowmelt ($mm$)\")\n",
    "plt.title(f\"Catchment: {catchm} (Trøndelag)\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/movingAverage_example_snowmelt.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "catchm = trond[\"final30\"][5]\n",
    "runoff = trond[\"data\"][catchm][\"precip\"]\n",
    "\n",
    "# moving average with different window sizes\n",
    "runoff5day = runoff.rolling(5,center=True).mean()\n",
    "runoff10day = runoff.rolling(10,center=True).mean()\n",
    "runoff15day = runoff.rolling(15,center=True).mean()\n",
    "runoff20day = runoff.rolling(20,center=True).mean()\n",
    "runoff30day = runoff.rolling(30,center=True).mean()\n",
    "\n",
    "line = itertools.cycle((\"--\",\"-.\",\":\"))\n",
    "colors = itertools.cycle((\"c\",\"salmon\",\"olive\"))\n",
    "\n",
    "# plotting\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.plot(runoff[\"2013\"],\"darkslategray\",label=\"Original\")\n",
    "plt.plot(runoff5day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"5day\")\n",
    "plt.plot(runoff10day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"10day\")\n",
    "#plt.plot(runoff15day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"15day\")\n",
    "#plt.plot(runoff20day.runoff[-365:],next(line),linewidth=2.5,color=next(colors),label=\"20day\")\n",
    "plt.plot(runoff30day[\"2013\"],next(line),linewidth=2.5,color=next(colors),label=\"30day\")\n",
    "plt.ylabel(\"Daily rainfall ($mm$)\")\n",
    "plt.title(f\"Catchment: {catchm} (Trøndelag)\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Plots/movingAverage_example_rainfall.png\",dpi=400,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extract timeseries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def extractMA(timeseries, interval, startYear=1983, endYear = 2012):\n",
    "    \"\"\"\n",
    "    Exstracts moving average timeseries in given timeperiod \n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    timeseries: \n",
    "        a pandas timeseries, with datetime index\n",
    "    interval: int\n",
    "        number of days in window\n",
    "    startYear: int\n",
    "        start of timeseries\n",
    "    endYear: int\n",
    "        end of timeseries\n",
    "        \n",
    "    Returns\n",
    "    -------\n",
    "    timeseries for given years\n",
    "    \"\"\"\n",
    "    start = datetime.datetime(startYear,1,1)\n",
    "    end = datetime.datetime(endYear,12,31)\n",
    "    \n",
    "    series = timeseries.rolling(interval,center=True).mean()\n",
    "    \n",
    "    return series[start:end]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Runoff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "r = [\"ost\",\"vest\",\"sor\",\"nord\",\"trond\",\"finn\"]\n",
    "data = [ost,vest,sor,nord,trond,finn]\n",
    "data = dict(zip(r,data))\n",
    "\n",
    "\"\"\"runoff30 = {}\n",
    "for region in data.keys():\n",
    "    runoff30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final\"]:\n",
    "        ts = d[\"data\"][c][\"runoff\"]\n",
    "        runoff30yr[c] = {}\n",
    "        runoff30yr[c][\"5day\"] = extractMA(ts, interval = 5).runoff\n",
    "        runoff30yr[c][\"10day\"] = extractMA(ts, interval = 10).runoff\n",
    "        runoff30yr[c][\"30day\"] = extractMA(ts, interval = 30).runoff\n",
    "    \n",
    "    runoff30[region] = runoff30yr\"\"\"\n",
    "\n",
    "spesRunoff30 = {}\n",
    "for region in data.keys():\n",
    "    runoff30yr = {}\n",
    "    d = data[region]\n",
    "    meta = d[\"metadata\"]\n",
    "    for c in d[\"final30\"]:\n",
    "        area = float(d[\"metadata\"][d[\"metadata\"].snumber==c].areal)\n",
    "        ts = d[\"data\"][c][\"runoff\"]/area\n",
    "        runoff30yr[c] = {}\n",
    "        runoff30yr[c][\"5day\"] = extractMA(ts, interval = 5).runoff/area\n",
    "        runoff30yr[c][\"10day\"] = extractMA(ts, interval = 10).runoff/area\n",
    "        runoff30yr[c][\"30day\"] = extractMA(ts, interval = 30).runoff/area\n",
    "    \n",
    "    spesRunoff30[region] = runoff30yr\n",
    "    \n",
    "\"\"\"runoff50 = {}\n",
    "for region in data.keys():\n",
    "    runoff30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final\"]:\n",
    "        ts = d[\"data\"][c][\"runoff\"]\n",
    "        runoff30yr[c] = {}\n",
    "        runoff30yr[c][\"5day\"] = extractMA(ts, interval = 5, startYear=1963).runoff\n",
    "        runoff30yr[c][\"10day\"] = extractMA(ts, interval = 10, startYear=1963).runoff\n",
    "        runoff30yr[c][\"30day\"] = extractMA(ts, interval = 30, startYear=1963).runoff\n",
    "    \n",
    "    runoff50[region] = runoff30yr\"\"\"\n",
    "\n",
    "spesRunoff50 = {}\n",
    "for region in data.keys():\n",
    "    runoff30yr = {}\n",
    "    d = data[region]\n",
    "    meta = d[\"metadata\"]\n",
    "    for c in d[\"final50\"]:\n",
    "        area = float(d[\"metadata\"][d[\"metadata\"].snumber==c].areal)\n",
    "        ts = d[\"data\"][c][\"runoff\"]/area\n",
    "        runoff30yr[c] = {}\n",
    "        runoff30yr[c][\"5day\"] = extractMA(ts, interval = 5, startYear=1963).runoff/area\n",
    "        runoff30yr[c][\"10day\"] = extractMA(ts, interval = 10, startYear=1963).runoff/area\n",
    "        runoff30yr[c][\"30day\"] = extractMA(ts, interval = 30, startYear=1963).runoff/area\n",
    "    \n",
    "    spesRunoff50[region] = runoff30yr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Temperature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp30 = {}\n",
    "for region in data.keys():\n",
    "    temp30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final30\"]:\n",
    "        ts = d[\"data\"][c][\"temp\"]\n",
    "        temp30yr[c] = {}\n",
    "        temp30yr[c][\"5day\"] = extractMA(ts, interval = 5)\n",
    "        temp30yr[c][\"10day\"] = extractMA(ts, interval = 10)\n",
    "        temp30yr[c][\"30day\"] = extractMA(ts, interval = 30)\n",
    "    \n",
    "    temp30[region] = temp30yr\n",
    "    \n",
    "temp50 = {}\n",
    "for region in data.keys():\n",
    "    temp30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final50\"]:\n",
    "        ts = d[\"data\"][c][\"temp\"]\n",
    "        temp30yr[c] = {}\n",
    "        temp30yr[c][\"5day\"] = extractMA(ts, interval = 5, startYear=1963)\n",
    "        temp30yr[c][\"10day\"] = extractMA(ts, interval = 10, startYear=1963)\n",
    "        temp30yr[c][\"30day\"] = extractMA(ts, interval = 30, startYear=1963)\n",
    "    \n",
    "    temp50[region] = temp30yr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Rainfall\n",
    "\n",
    "**OBS!** \"Precipitation\" is possibly only rainfall data\n",
    "\n",
    "How should it be smoothed? Does MA makes sense?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rainfall30 = {}\n",
    "for region in data.keys():\n",
    "    rainfall30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final30\"]:\n",
    "        ts = d[\"data\"][c][\"precip\"]\n",
    "        rainfall30yr[c] = {}\n",
    "        rainfall30yr[c][\"5day\"] = extractMA(ts, interval = 5)\n",
    "        rainfall30yr[c][\"10day\"] = extractMA(ts, interval = 10)\n",
    "        rainfall30yr[c][\"30day\"] = extractMA(ts, interval = 30)\n",
    "    \n",
    "    rainfall30[region] = rainfall30yr\n",
    "    \n",
    "rainfall50 = {}\n",
    "for region in data.keys():\n",
    "    rainfall30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final50\"]:\n",
    "        ts = d[\"data\"][c][\"precip\"]\n",
    "        rainfall30yr[c] = {}\n",
    "        rainfall30yr[c][\"5day\"] = extractMA(ts, interval = 5, startYear=1963)\n",
    "        rainfall30yr[c][\"10day\"] = extractMA(ts, interval = 10, startYear=1963)\n",
    "        rainfall30yr[c][\"30day\"] = extractMA(ts, interval = 30, startYear=1963)\n",
    "    \n",
    "    rainfall50[region] = rainfall30yr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Snowmelt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "snow30 = {}\n",
    "for region in data.keys():\n",
    "    snow30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final30\"]:\n",
    "        ts = d[\"data\"][c][\"snow\"]\n",
    "        snow30yr[c] = {}\n",
    "        snow30yr[c][\"5day\"] = extractMA(ts, interval = 5).qsw\n",
    "        snow30yr[c][\"10day\"] = extractMA(ts, interval = 10).qsw\n",
    "        snow30yr[c][\"30day\"] = extractMA(ts, interval = 30).qsw\n",
    "    \n",
    "    snow30[region] = snow30yr\n",
    "    \n",
    "snow50 = {}\n",
    "for region in data.keys():\n",
    "    snow30yr = {}\n",
    "    d = data[region]\n",
    "    for c in d[\"final50\"]:\n",
    "        ts = d[\"data\"][c][\"snow\"]\n",
    "        snow30yr[c] = {}\n",
    "        snow30yr[c][\"5day\"] = extractMA(ts, interval = 5, startYear=1963).qsw\n",
    "        snow30yr[c][\"10day\"] = extractMA(ts, interval = 10, startYear=1963).qsw\n",
    "        snow30yr[c][\"30day\"] = extractMA(ts, interval = 30, startYear=1963).qsw\n",
    "    \n",
    "    snow50[region] = snow30yr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Saving to file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "saveDict(snow30,\"MA_snowmelt_30year\")\n",
    "saveDict(snow50,\"MA_snowmelt_50year\")\n",
    "saveDict(rainfall30,\"MA_rainfall_30year\")\n",
    "saveDict(rainfall50,\"MA_rainfall_50year\")\n",
    "saveDict(temp30,\"MA_temperature_30year\")\n",
    "saveDict(temp50,\"MA_temperature_50year\")\n",
    "saveDict(spesRunoff30,\"MA_streamflow_30year\")\n",
    "saveDict(spesRunoff50,\"MA_streamflow_50year\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}