NFDRSV4-TechDoc-LiveFuelMoistureEvaluation-Calibration.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Documentation to support the publication: \"Modernizing the US National Fire Danger Rating System (Version 4): Simplified Fuel Models and Improved Live and Dead Fuel Moisture Calculations\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Purpose: Evaluate the Growing Season Index-based Live Fuel Moisture Model used in NFDRS Version 4.0 and show a simple method for calibrating the model used field measurements of LFMC."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Original Notebook edited to add three more LFMC sites for comparison as per reviewer recommendations. (WMJ, 25 July 2024)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font size=5>NFDRSV4-TechDoc-LiveFuelMoistureEvaluation-Calibration.ipynb</font><br>\n",
    "W. Matt Jolly, PhD, william.jolly@usda.gov<br>\n",
    "Patrick H. Freeborn, PhD,  patrick.h.freeborn@usda.gov<br>\n",
    "USDA Forest Service<br>\n",
    "RMRS Fire Sciences Laboratory<br>\n",
    "Fire, Fuel and Smoke Program<br>\n",
    "5775 Hwy 10 W<br>\n",
    "Missoula, MT 59808<br>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math, sys, os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "from matplotlib import pyplot as plt\n",
    "from datetime import *\n",
    "from scipy.stats import spearmanr,stats,linregress\n",
    "from sklearn.metrics import mean_absolute_error,r2_score\n",
    "from pandas.plotting import register_matplotlib_converters\n",
    "import matplotlib.cm as cm\n",
    "register_matplotlib_converters()\n",
    "from matplotlib.dates import DateFormatter\n",
    "from src.NFDRSV4PubGraphicsOptions import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "pd.set_option('display.max_rows', 200)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Basic met functions, photoperiod and conversions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "## fToC Converts Fahrenheit to Celcius\n",
    "## Inputs: f: Temperature in deg F\n",
    "## Returns: c: Temperature in deg C\n",
    "def fToc(f):\n",
    "    return (f -32) * 5/9\n",
    "\n",
    "## cTof Converts Celcius to Fahrenheit\n",
    "## Inputs: c: Temperature in deg C\n",
    "## Returns: f: Temperature in deg F\n",
    "def cTof(c):\n",
    "    return (c * 9/5) + 32\n",
    "INC = 0\t#GSI Function Types, Increasing (Tmin, Dayl)\n",
    "DEC = 1 #GSI Function Types, Decreasing (VPD)\n",
    "MM_2_IN = 0.0393701  # Conversion factor from mm to inches\n",
    "KPH_2_MPH = 0.621371 # Conversion factor from KPH to MPH\n",
    "\n",
    "## Calculate the Photoperiod/Daylength from latitude and day of year (yearday)\n",
    "## Inputs: lat: Latitude in Degrees (90 to -90)\n",
    "##         yday: Yearday (1-366)\n",
    "def CalcDayl(lat,yday):\n",
    "    RADPERDAY = 0.017214\n",
    "    RADPERDEG = 0.01745329\n",
    "    MINDECL = -0.4092797\n",
    "    SECPERRAD = 13750.9871\n",
    "    DAYSOFF = 10.25\n",
    "    # Daylength function from MT-CLIM */\n",
    "    lat = lat * RADPERDEG\n",
    "    if lat > 1.5707:\n",
    "        lat = 1.5707\n",
    "    if lat < -1.5707:\n",
    "        lat = -1.5707\n",
    "    coslat = math.cos(lat)\n",
    "    sinlat = math.sin(lat)\n",
    "\n",
    "    #* calculate cos and sin of declination */\n",
    "    decl = MINDECL * math.cos((yday + DAYSOFF) * RADPERDAY)\n",
    "    cosdecl = math.cos(decl)\n",
    "    sindecl = math.sin(decl)\n",
    "    cosegeom = coslat * cosdecl\n",
    "    sinegeom = sinlat * sindecl\n",
    "    coshss = -(sinegeom) / cosegeom\n",
    "    if coshss < -1.0:\n",
    "        coshss = -1.0  # 24-hr daylight */\n",
    "    if coshss > 1.0:\n",
    "        coshss = 1.0    # 0-hr daylight */\n",
    "    hss = math.acos(coshss)                # hour angle at sunset (radians) */\n",
    "    #* daylength (seconds) */\n",
    "    return 2.0 * hss * SECPERRAD\n",
    "\n",
    "# Function to calculate the Saturation Vapor Pressure for a given temperature.\n",
    "# Note: Function converts internally from F to C.\n",
    "# Input: tempF: Temperature in Deg F\n",
    "# Return: Saturation Vapor Pressure (Pa)\n",
    "def CalcVP(tempF):\n",
    "    tmpC =  (tempF - 32.0) / 1.8\n",
    "    vp = 610.7 * math.exp((17.38 * tmpC)/(239 + tmpC))\n",
    "    return vp\n",
    "\n",
    "# Calculate the VPD from RH and temperature\n",
    "# Inputs: RH: Relative Humidity in Percent (1-100)\n",
    "#         TempF: Temperature in Deg F\n",
    "# Return: Vapor Pressure Deficit (Pa)\n",
    "def CalcVPD(RH, TempF):\n",
    "    vp = CalcVP(TempF)\n",
    "    vpd = vp - (RH / 100) * vp\n",
    "    if(vpd < 0.0):\n",
    "        vpd = 0.0;\n",
    "    return vpd\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GSI Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# A class to contains all of the parameters used to derived GSI.  This makes it easier to iterate through parameter spaces for\n",
    "# for optimization\n",
    "class GSILimits:\n",
    "    TminLow = -2       # Lower limit for minimum temperature (C)\n",
    "    TminUp = 5         # Upper limit for minimum temperature (C)\n",
    "    DaylLow = 36000    # Lower limit for daylength (seconds)\n",
    "    DaylUp = 39600     # Upper limit for daylength (seconds)\n",
    "    VPDLow = 900       # Lower limit for VPD (pascals)\n",
    "    VPDUp = 4100       # Upper limit for VPD (pascals)\n",
    "    PrcpRTLow = 0    # Lower limit for running total precip (inches)\n",
    "    PrcpRTUp = 0.393701  # Upper limit for running total precip (inches)... defaults to 10mm\n",
    "    PrcpRTPeriod = 28  # Running total period for precipitaiton (days)\n",
    "    GSIPeriod = 28     # Running average period for final GSI (days)\n",
    "    GUThresh = 0.2    # Green-up threshold (dim)\n",
    "    LFMMax = 200       # Maximum fuel moisture (% dry wt)\n",
    "    LFMMin = 60        # Minimum fuel moisture (% dry wt)\n",
    "    Lat = 45           # Station latitude (degrees)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper function to set the GSI Limits class values based on the list of calibrated parameters\n",
    "# gsilim: An instance of the GSILimits class\n",
    "# gsiparams: an 11 element list of GSI parameters\n",
    "# LFMMin: minimum live fuel moisture limit for model\n",
    "# LFMMax: maximum live fuel moisture limit for model\n",
    "# Lat: Latitude of the estimatation location (used in photoperiod calculations)\n",
    "def SetGSILimits(gsilim,gsiparams,LFMMin,LFMMax, Lat):\n",
    "    gsilim.TminLow = gsiparams[0] \n",
    "    gsilim.TminUp = gsiparams[1]\n",
    "    gsilim.VPDLow = gsiparams[2]\n",
    "    gsilim.VPDUp = gsiparams[3]\n",
    "    gsilim.DaylLow = gsiparams[4]\n",
    "    gsilim.DaylUp = gsiparams[5]\n",
    "    gsilim.PrcpRTLow = gsiparams[6]\n",
    "    gsilim.PrcpRTUp = gsiparams[7]\n",
    "    gsilim.GSIPeriod = gsiparams[8]\n",
    "    gsilim.GUThresh = gsiparams[9]\n",
    "    gsilim.PrcpRTPeriod = gsiparams[10]\n",
    "    gsilim.LFMMin = LFMMin\n",
    "    gsilim.LFMMax = LFMMax\n",
    "    gsilim.Lat = Lat\n",
    "    return gsilim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ramp function for GSI calculations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# GSI indicator/ramp function\n",
    "def Ind(Var,Low,Up):\n",
    "    # Make sure all the input variables are numbers\n",
    "    Var = float(Var)\n",
    "    Low = float(Low)\n",
    "    Up = float(Up)\n",
    "    if(Up == Low):  # Upper (Up) and Lower (Low) can't be the same\n",
    "        return 0\n",
    "    if( Var < Low):  # If the variables is less than the lower limit, return 0\n",
    "        return 0\n",
    "    elif(Var > Up):  # If the variables is greater than the upper limit, return 1\n",
    "        return 1\n",
    "    else:            # If the variables is between the lower and upper limits, return the proportion\n",
    "        return (Var - Low) / (Up - Low)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transform GSI to LFMC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GSI is scaled to LFMC using a simple linear transform above a threshold value.  The min and max LFMC values are fuel type specific and are carried over from previous versions of the USNFDRS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Function to transform GSI to LFMC\n",
    "def CalcLFMFromGSI(gsi,GUThresh,LFMMin,LFMMax):\n",
    "    m = (LFMMax - LFMMin) / (1 - GUThresh)\n",
    "    b = LFMMax - m\n",
    "    # Scale GSI between min and max LFMC if gsi is above the green-up threshold\n",
    "    if(gsi >= GUThresh):\n",
    "        LFM = m*gsi + b\n",
    "    # Below the Green-up Threshold, set the value to minimum LFM\n",
    "    else:\n",
    "        LFM = LFMMin\n",
    "    return(LFM)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to calculate the Growing Season Index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to calculate GSI, Scaled GSI and Live Fuel Moisture on a DataFrame\n",
    "# df: Pandas DataFrame of daily weather data\n",
    "# assumes the following labeled columns: \"Tmin: minimum temperature (deg F), VPDMax: Max daily VPD (Pa)\n",
    "#                                          and Prcp: 24 hour total precipitation (Inches)\"\n",
    "# gsilim: An instance of the GSILimits class \n",
    "# PLowLim and PUpperLim: Date range for model predictions\n",
    "\n",
    "def CalcGSI(df,gsilim,PLowLim='2014-01-01',PUpperLim = '2020-12-31'):\n",
    "    # Filter the dataset for the period of interest (5 year period by default but can be adjusted with PLowLim and PUpperLim)\n",
    "    df = df[(df['DateTime'] > PLowLim) & (df['DateTime'] <= PUpperLim)].copy()\n",
    "    # Compute and store the day of the year for use in calculating Daylength\n",
    "    df['JDay'] = df.DateTime.dt.dayofyear\n",
    "    \n",
    "    # Minimum Temperature\n",
    "    df['TminC'] =  (df.Tmin - 32.0) * 5.0 / 9.0; # Convert Tmin from Fahrenheit to celcuius\n",
    "    df['TminInd'] = df.apply(lambda row: Ind(row['TminC'],gsilim.TminLow,gsilim.TminUp),axis=1)\n",
    "    # Vapor Pressure Deficit\n",
    "    df['VPDInd']= df.apply(lambda row: 1 - Ind(row['VPDMax'],gsilim.VPDLow,gsilim.VPDUp),axis=1)\n",
    "    # Photoperiod / Daylength\n",
    "    # First, calculate the Daylength for each observation day (this uses site latitude and julian day)\n",
    "    df['Dayl'] = df.apply(lambda row: CalcDayl(gsilim.Lat,row['JDay']),axis=1)  \n",
    "    df['DaylInd']= df.apply(lambda row: Ind(row['Dayl'],gsilim.DaylLow,gsilim.DaylUp),axis=1)\n",
    "    # Running Total Precipitation\n",
    "    df['Prcp_RT'] = df['Prcp'].rolling(gsilim.PrcpRTPeriod).sum()\n",
    "    df['PrcpInd'] = df.apply(lambda row: Ind(row['Prcp_RT'],gsilim.PrcpRTLow,gsilim.PrcpRTUp),axis=1)\n",
    "\n",
    "    #################################################################                  \n",
    "    # Daily GSI for four ramp model\n",
    "    df['iGSI_PE'] = df['TminInd'] * df['VPDInd'] * df['DaylInd'] * df['PrcpInd']\n",
    "    # Smoothed GSI (running average over the GSIPeriod)\n",
    "    df['GSI_PE'] = df['iGSI_PE'].rolling(gsilim.GSIPeriod).mean()\n",
    "    # Rescale the final GSI time series\n",
    "    df['GSI_PE_RS'] = df['GSI_PE']/ df['GSI_PE'].quantile(1)\n",
    "\n",
    "    # Calculate LFMC from Rescaled GSI\n",
    "    df['LFMC'] = df.apply(lambda row: CalcLFMFromGSI(row['GSI_PE_RS'],gsilim.GUThresh,gsilim.LFMMin, gsilim.LFMMax),axis=1)\n",
    "    return df\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to calculate and compare GSI using weather obs, measured LFMC and a set of GSI-LFM parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to provide a bounded list of lower and upper values that ensures \n",
    "# that min is always less than max\n",
    "# To be used in a Grid Search Optimization for GSI parameters\n",
    "def MakeParamRange(minValue,maxValue,interv):\n",
    "    lower = []\n",
    "    upper = []\n",
    "    for x in range(minValue,maxValue+1,interv): \n",
    "        for y in range(minValue,maxValue+1,interv):\n",
    "            if(x < y):\n",
    "                #print(paste(x,y))\n",
    "                lower.append(x)\n",
    "                upper.append(y)\n",
    "    d = {'Lower': lower, 'Upper': upper}\n",
    "    df = pd.DataFrame(data=d)\n",
    "    return(df)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Produce a table entry comparing 78 and GSI/GSIp LFM to Observations\n",
    "def MakeGSILFMCompareTable4Param(Temp,LFMObs,BestParams,label=\"\",UseLFMMinMax = True,Herb=True,Lat=45,PLowLim='2015-01-01',PUpperLim='2019-12-31'):\n",
    "    #PLowLim='2015-01-01'\n",
    "    #PUpperLim = '2019-12-31'\n",
    "    \n",
    "    # Label the DataFrame columns so they match the GSI function\n",
    "    Temp = Temp.rename(columns={\"MinT\": \"Tmin\", \"VPDM\": \"VPDMax\", \"Rain\":\"Prcp\"})\n",
    "    # Lookup the appropriate live fuel moisture ranges\n",
    "    LFMMin,LFMMax = GetLFMRange(LFMObs,UseLFMMinMax,Herb)\n",
    "\n",
    "    # Create the GSILimits structure as a default and then overwrite the variables we need to change\n",
    "    gsilim = GSILimits\n",
    "    newgsilim = SetGSILimits(gsilim,BestParams,LFMMin,LFMMax,Lat)\n",
    "    \n",
    "    # Calculate the GSI\n",
    "    t = CalcGSI(Temp,newgsilim,PLowLim,PUpperLim)\n",
    "    \n",
    "    # Assign the components back to the dataframe\n",
    "    Temp['GSI_PE_RS'] = t['GSI_PE_RS']\n",
    "    Temp['LFMC'] = t['LFMC']  \n",
    "    Temp['TminInd'] = t['TminInd']\n",
    "    Temp['VPDInd'] = t['VPDInd']\n",
    "    Temp['DaylInd'] = t['DaylInd']\n",
    "    Temp['PrcpInd'] = t['PrcpInd']\n",
    "    \n",
    "    # Merge the GSI and LFM obs dataframes\n",
    "    MetLFMMerge = LFMObs.merge(Temp, left_on='DateTime', right_on='DateTime', suffixes=(False, False))\n",
    "   \n",
    "    # Trim the dataset to the analysis date range\n",
    "    MetLFMMerge = MetLFMMerge[(MetLFMMerge['DateTime'] > PLowLim) & (MetLFMMerge['DateTime'] <= PUpperLim)]\n",
    "    \n",
    "    # Compare the GSI-LFMC (NFDRS-V4) to measured LFMC\n",
    "    s2,p2 = spearmanr(MetLFMMerge['Percent'],MetLFMMerge['LFMC'],nan_policy=\"omit\")\n",
    "    mae2 = mean_absolute_error(MetLFMMerge['Percent'], MetLFMMerge['LFMC'])\n",
    "    slope, intercept, r_value, p_value_def, std_err = linregress(MetLFMMerge['Percent'],MetLFMMerge['LFMC'])\n",
    "    r2 = r_value**2                                        \n",
    "    print(r2)\n",
    "    # Compare the old LFMC (NFDRS V2) to measure LFMC\n",
    "    if Herb:\n",
    "        s3,p3 = spearmanr(MetLFMMerge['Percent'],MetLFMMerge['FMH'])\n",
    "        slope, intercept, r_value, p_value_def, std_err = linregress(MetLFMMerge['Percent'],MetLFMMerge['FMH'])\n",
    "        r3 = r_value**2\n",
    "        mae3 = mean_absolute_error(MetLFMMerge['Percent'], MetLFMMerge['FMH'])\n",
    "    else:\n",
    "        s3,p3 = spearmanr(MetLFMMerge['Percent'],MetLFMMerge['FMW'])\n",
    "        slope, intercept, r_value, p_value_def, std_err = linregress(MetLFMMerge['Percent'],MetLFMMerge['FMW'])\n",
    "        r3 = r_value**2\n",
    "        mae3 = mean_absolute_error(MetLFMMerge['Percent'], MetLFMMerge['FMW'])\n",
    "    \n",
    "    return([label,round(s2,3),round(p2,3),round(mae2,3),round(r2,3),round(s3,3),round(p3,3),round(mae3,3),round(r3,3),len(MetLFMMerge['Percent'])])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to perform a Grid Search Optimization of the GSI by best fit to field measurements"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def GridSearchOptimizeGSILFM(Temp,LFMDat,label,maxSim = 2,mySeed = 123456,UseLFMMinMax = False,Herb=True,PInt = 20,PLowLim='2015-01-01',PUpperLim='2019-12-31',Lat=45):\n",
    "    DEBUG = False\n",
    "      \n",
    "    # Create a range of smoothing / running precip period length ranges intervals\n",
    "    smint = range(21,60,7)\n",
    "    dfsmint = pd.DataFrame(data={'SMInt': smint})\n",
    "    \n",
    "    #### Get a Random Set of Parameters\n",
    "    # Make the VPD ranges\n",
    "    iVPDP = MakeParamRange(500, 9000, 500)  # VPD ranges from 500 to 9000 Pascals\n",
    "    # Make the Temperature ranges\n",
    "    iTMinP = MakeParamRange(-5, 10,1)  # MinT ranges from -5 to 10 deg C\n",
    "    # Make the Daylength Ranges\n",
    "    iDaylP = MakeParamRange(32400,46800,3600)  # Dayl ranges from 9 to 13 hours\n",
    "    # Make the Prcp ranges\n",
    "    iPrcpP = MakeParamRange(0,50,1)\n",
    "    iPrcpP = (iPrcpP/10)  # Precip ranges from 0 to 5 inches\n",
    "\n",
    "    # Make the green-up threshold \n",
    "    iThreshP = []\n",
    "    for i in range(0,81,10):\n",
    "        iThreshP.append(float(i/100))\n",
    "    iThreshP = pd.DataFrame(data={'iThresh': iThreshP})    \n",
    "    \n",
    "    BestS = 0\n",
    "    BestParams = []\n",
    "    \n",
    "    for i in range(0,maxSim):\n",
    "        \n",
    "        rs = dfsmint.sample(n=1,random_state=mySeed+i)\n",
    "        smint = rs.iloc[0].SMInt.astype(int)\n",
    "        \n",
    "        # VPD Params\n",
    "        rs = iVPDP.sample(n=1,random_state=mySeed+i)\n",
    "        iVPDMin = rs.iloc[0].Lower\n",
    "        iVPDMax = rs.iloc[0].Upper\n",
    "        \n",
    "        # TMin Params\n",
    "        rs = iTMinP.sample(n=1,random_state=mySeed+i)\n",
    "        iTminMin = rs.iloc[0].Lower\n",
    "        iTminMax = rs.iloc[0].Upper\n",
    "\n",
    "        # Daylength Params\n",
    "        rs = iDaylP.sample(n=1,random_state=mySeed+i)\n",
    "        iDaylMin = rs.iloc[0].Lower\n",
    "        iDaylMax = rs.iloc[0].Upper\n",
    "\n",
    "        # RT Precip Params\n",
    "        rs = iPrcpP.sample(n=1,random_state=mySeed+i)\n",
    "        iPrcpMin = rs.iloc[0].Lower\n",
    "        iPrcpMax = rs.iloc[0].Upper\n",
    "\n",
    "        # Greenup Threshold \n",
    "        rs = iThreshP.sample(n=1,random_state=mySeed+i)\n",
    "        iThreshVal = rs.iloc[0].iThresh.astype(float)\n",
    "        \n",
    "        Params = [iTminMin,iTminMax,iVPDMin,iVPDMax,iDaylMin,iDaylMax,iPrcpMin,iPrcpMax,smint,iThreshVal,smint]\n",
    "               \n",
    "        S =  MakeGSILFMCompareTable4Param(Temp,LFMDat,Params,label=label,UseLFMMinMax = UseLFMMinMax,Herb=Herb,PLowLim=PLowLim,PUpperLim=PUpperLim,Lat=Lat)\n",
    "        \n",
    "        if(S[1] > BestS):\n",
    "            BestS = S[1]\n",
    "            BestParams = Params\n",
    "           \n",
    "        if i % PInt == 0:\n",
    "            print(S,BestS,BestParams)\n",
    "    return(BestParams)\n",
    "       "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Input file helper functions for observed LFMC and met data files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper function for reading the NFMD measured live fuel moisture data \n",
    "def ReadNFMDData(filename):\n",
    "        # Read in data\n",
    "        data = pd.read_csv(filename, sep='\\t')\n",
    "        data['DateTime']=pd.to_datetime(data['Date'].astype(str) + \" 13\", format='%Y-%m-%d %H')\n",
    "        return data \n",
    "\n",
    "# Helper function for reading the export files from the Fuel Moisture Repository\n",
    "def ReadFMRData(filename):\n",
    "        # Read in data\n",
    "        #colspecs = [(0, 8), (9, 13), (13, 25), (25, 53), (53, 65), (65, 73)]\n",
    "        data = pd.read_csv(filename)\n",
    "        data['DateTime']=pd.to_datetime(data['DATE'].astype(str) + \" 13\", format='%Y-%m-%d %H')\n",
    "        return data \n",
    "\n",
    "\n",
    "# Helper function for reading the NFMD measured live fuel moisture data \n",
    "def ReadNFMDData(filename):\n",
    "        # Read in data\n",
    "        colspecs = [(0, 8), (9, 13), (13, 25), (25, 53), (53, 65), (65, 73)]\n",
    "        data = pd.read_csv(filename, sep='\\t')\n",
    "        data['DateTime']=pd.to_datetime(data['Date'].astype(str) + \" 13\", format='%Y-%m-%d %H')\n",
    "        return data \n",
    "# Helper function to derive historical min and max LFMC value from a NFMD set of measurements\n",
    "def GetLFMRange(LFMObs,UseLFMMinMax,Herb):\n",
    "    # Set the default green-up threshold and min / max live fuel moisture\n",
    "    if not UseLFMMinMax:\n",
    "        if Herb:\n",
    "            LFMMin = 30\n",
    "            LFMMax = 250\n",
    "        else:\n",
    "            LFMMin = 60\n",
    "            LFMMax = 200\n",
    "    else:\n",
    "        LFMMin = LFMObs['Percent'].quantile(0.01).astype(int)\n",
    "        LFMMax = LFMObs['Percent'].quantile(0.99).astype(int)\n",
    "    return(LFMMin,LFMMax)\n",
    "\n",
    "# Helper function for reading the FireFamily+ weather data and formatting columns corrected for use as Pandas dataframes\n",
    "def GetMet(inFileName,Lat,delLast=False):\n",
    "    Met = pd.read_csv(inFileName)\n",
    "\n",
    "    Met.rename(columns=lambda x: x.strip(), inplace=True)\n",
    "    ind = len(Met.columns)\n",
    "    print(ind)\n",
    "    if delLast:\n",
    "        Met=Met.drop(Met.columns[[ind - 1]],axis=1)\n",
    "    Met['DateTime']=pd.to_datetime(Met['DATE'].astype(str) + \" 13\", format='%m/%d/%Y %H')\n",
    "    MaxRain = float(Met['Rain'].quantile([0.999]).values)\n",
    "    #print(MaxRain)\n",
    "    Met['Rain'].values[Met['Rain'] > MaxRain] = MaxRain\n",
    "    Met['RunSumPrcp'] = Met['Rain'].rolling(28).sum()\n",
    "    # Data Prep\n",
    "    Met['MinTC'] = fToc(Met['MinT'])\n",
    "    JDay = []\n",
    "    Dayl = []\n",
    "    for index, row in Met.iterrows():\n",
    "        jday = int(row.DateTime.strftime(format=\"%j\"))\n",
    "        JDay.append(jday)\n",
    "        dayl = CalcDayl(Lat,jday) \n",
    "        Dayl.append(dayl)\n",
    "    Met['Dayl'] = Dayl\n",
    "    Met['JDay'] = JDay\n",
    "    return Met"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Simple GSI calc function that works on the "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def CalcGSIPandas(Temp,Params,LFMMin,LFMMax,Lat):\n",
    "    # Label the DataFrame columns so they match the GSI function\n",
    "    Temp = Temp.rename(columns={\"MinT\": \"Tmin\", \"VPDM\": \"VPDMax\", \"Rain\":\"Prcp\"})\n",
    "    gsilim = GSILimits\n",
    "    newgsilim = SetGSILimits(gsilim,Params,LFMMin,LFMMax,Lat)\n",
    "    \n",
    "    # Calculate the GSI\n",
    "    t = CalcGSI(Temp,newgsilim,PLowLim='2014-01-01',PUpperLim = '2024-12-31')\n",
    "\n",
    "    Temp['GSI_PE_RS'] = t['GSI_PE_RS']\n",
    "    Temp['LFMC'] = t['LFMC']  \n",
    "    Temp['TminInd'] = t['TminInd']\n",
    "    Temp['VPDInd'] = t['VPDInd']\n",
    "    Temp['DaylInd'] = t['DaylInd']\n",
    "    Temp['PrcpInd'] = t['PrcpInd']\n",
    "    \n",
    "    # Merge the GSI and LFM obs dataframes\n",
    "    #MetLFMMerge = LFMObs.merge(Temp, left_on='DateTime', right_on='DateTime', suffixes=(False, False))\n",
    "    return(Temp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def MakeGSILFMCompareNew(Temp,LFMObs,BestParams,UseLFMMinMax = True,PLowLim='2015-01-01',PUpperLim = '2019-12-31',UsePrcp=False,Lat=45,Herb=False):\n",
    "   \n",
    "    # Label the DataFrame columns so they match the GSI function\n",
    "    Temp = Temp.rename(columns={\"MinT\": \"Tmin\", \"VPDM\": \"VPDMax\", \"Rain\":\"Prcp\"}).copy()\n",
    "   \n",
    "    # Set the default green-up threshold and min / max live fuel moisture \n",
    "    LFMMin,LFMMax = GetLFMRange(LFMObs,UseLFMMinMax,Herb)\n",
    "    #print({\"UseLFMMinMax\":UseLFMMinMax,\"Herb\":Herb,\"LFMMin\":LFMMin,\"LFMMax\":LFMMax})\n",
    "    gsilim = GSILimits\n",
    "    newgsilim = SetGSILimits(gsilim,BestParams,LFMMin,LFMMax,Lat)\n",
    "    \n",
    "    # Calculate the GSI\n",
    "    t = CalcGSI(Temp,newgsilim,PLowLim=PLowLim,PUpperLim = PUpperLim)\n",
    "    Temp['GSI_PE_RS'] = t['GSI_PE_RS']\n",
    "    Temp['LFMC'] = t['LFMC']  \n",
    "    Temp['TminInd'] = t['TminInd']\n",
    "    Temp['VPDInd'] = t['VPDInd']\n",
    "    Temp['DaylInd'] = t['DaylInd']\n",
    "    Temp['PrcpInd'] = t['PrcpInd']\n",
    "    \n",
    "    # Merge and filter to limit date range\n",
    "    MetLFMMerge = LFMObs.merge(Temp, left_on='DateTime', right_on='DateTime', suffixes=(False, False))\n",
    "    MetLFMMerge = MetLFMMerge[(MetLFMMerge['DateTime'] > PLowLim) & (MetLFMMerge['DateTime'] <= PUpperLim)]\n",
    "        \n",
    "    if MetLFMMerge.LFMC.std() == 0:\n",
    "        s2 = [-999.9]\n",
    "        mae = 999.9\n",
    "    else:\n",
    "        s2,p2 = spearmanr(MetLFMMerge['Percent'],MetLFMMerge['LFMC']*100,nan_policy=\"omit\")\n",
    "        mae = mean_absolute_error(MetLFMMerge['Percent'], MetLFMMerge['LFMC'])\n",
    "\n",
    "    return([s2,mae,MetLFMMerge])\n",
    "\n",
    "def GetMetMerge(Met,LFMObs,BestParams,UseLFMMinMax,Lat=45,Herb=False,PLowLim='2015-01-01',PUpperLim = '2019-12-31'):\n",
    "    UsePrcp = True\n",
    "    LFMMin,LFMMax = GetLFMRange(LFMObs,UseLFMMinMax,Herb)\n",
    "    #print(LFMMin,LFMMax)\n",
    "    t = MakeGSILFMCompareNew(Met,LFMObs,BestParams,UseLFMMinMax,UsePrcp=UsePrcp,Lat=Lat,Herb=Herb,PLowLim=PLowLim,PUpperLim=PUpperLim)\n",
    "    return( t[2])\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Station-based analyses and local Grid Search Optimization "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the surface weather observations from the RAWS stations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11\n",
      "11\n",
      "8\n",
      "8\n",
      "8\n"
     ]
    }
   ],
   "source": [
    "# Read in the weather data and format them into Pandas dataframes\n",
    "# Note: These data were created using FireFamily+ and they include the weather, Max VPD and the calculated NFDRS Version 2 live fuel moisture content\n",
    "\n",
    "Lat421501 = 39.58\n",
    "FName = r'./data/421501.csv'\n",
    "Met421501 = GetMet(FName,Lat421501,True)\n",
    "\n",
    "Lat045433 = 34.08\n",
    "FName = r'./data/045433.csv'\n",
    "Met045433 = GetMet(FName,Lat045433,True) \n",
    "Met045433 = Met045433.dropna()\n",
    "\n",
    "LatSSFS = 44\n",
    "FName = r'./data/SSFS.csv'\n",
    "MetSSFS = GetMet(FName,LatSSFS,True)\n",
    "MetSSFS = MetSSFS.drop('DATE',axis=1)\n",
    "\n",
    "#https://wrcc.dri.edu/cgi-bin/rawMAIN.pl?sdTLUF \n",
    "Lat413509 = 31.313\n",
    "FName = r'./data/413509.csv'\n",
    "Met413509 = GetMet(FName,Lat413509,True)\n",
    "Met413509 = Met413509.drop('DATE',axis=1)\n",
    "\n",
    "Lat101303 = 46.5\n",
    "FName = r'./data/101303.csv'\n",
    "Met101303 = GetMet(FName,Lat101303,True)\n",
    "Met101303 = Met101303.drop('DATE',axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the Live Fuel Moisture Content field measurements from either the National Fuel Moisture Database or the Fuel Moisture Repository Webportal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read field-measured live fuel moisture values obtained from the US National Fuel Moisture Database\n",
    "# These files were exported from the National Fuel Moisture Database\n",
    "UT_ArtTri = ReadNFMDData(\"./data/EGBC_UT_FillmoreFieldOffice_SevierReservoir_SagebrushBasinBig_data.txt\")\n",
    "UT_BroTec = ReadNFMDData(\"./data/EGBC_UT_FillmoreFieldOffice_SevierReservoir_Cheatgrass_data.txt\")\n",
    "CA_AdeFas = ReadNFMDData(\"./data/SOCC_CA_LACounty_LaurelCanyon,Mt.Olympus_data.txt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Additional sites added per peer review recommendation (WMJ, 25 July 2024)\n",
    "# Fuel Moisture Repository Webportal\n",
    "# https://nfmdb.org/?&target=_blank\n",
    "ID_PhyCap = ReadFMRData(\"./data/FMR_202407191615.csv\").rename(columns={'PERCENT':'Percent'})\n",
    "TX_IleVom = ReadFMRData(\"./data/FMR_202407221747.csv\").rename(columns={'PERCENT':'Percent'})\n",
    "SD_BroIne = ReadFMRData(\"./data/FMR_202407221751.csv\").rename(columns={'PERCENT':'Percent'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# GSI-LFMC default parameters\n",
    "GU = 0.2\n",
    "Days = 28\n",
    "defParams = [-2, 5, 900, 4100, 36000, 39600, 0.394, 0.787, Days,GU,Days]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Max number of simulations for Grid Search Optimizations\n",
    "maxSims = 10000\n",
    "## Random seed number to ensure results are random but repeatable\n",
    "mySeed = 123456"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These simulations have already been run and the results saved below. Using the mySeed parameter, they are randomly sampled but repeatable, so enabling SIM would re-run the Grid Search Optimization for each of the three species but should yield the same final parameter set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "SIM = False\n",
    "if SIM:\n",
    "    results_brotec = GridSearchOptimizeGSILFM(Met421501,UT_BroTec,\"UT_BroTec\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=True,Lat=421501)\n",
    "    results_arttri = GridSearchOptimizeGSILFM(Met421501,UT_ArtTri,\"UT_ArtTri\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=False,Lat=421501)\n",
    "    results_adefas = GridSearchOptimizeGSILFM(Met045433,CA_AdeFas,\"CA_AdeFas\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=False,Lat=Lat045433)\n",
    "    results_ilevom = GridSearchOptimizeGSILFM(Met413509,TX_IleVom,\"TX_IleVom\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=False,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=Lat413509)\n",
    "    results_broine = GridSearchOptimizeGSILFM(MetSSFS,SD_BroIne,\"SD_BroIne\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=True,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=LatSSFS)\n",
    "    results_phycap = GridSearchOptimizeGSILFM(Met101303,ID_PhyCap,\"ID_PhyCap\",maxSim = maxSims,mySeed=mySeed,UseLFMMinMax = True,Herb=False,PLowLim='2019-01-01',PUpperLim='2023-12-31',Lat=Lat101303)\n",
    "else:\n",
    "    # Load the stored Grid Search Optimized parameter sets for each of the three species. These can be re-run by setting SIM to True in this cell\n",
    "    results_adefas = [0, 6, 500, 7000, 32400, 36000, 1.2, 4.8, 56, 0.0, 56]#[-4, 0, 2500, 4000, 32400, 36000, 1.5, 3.0, 55, 0.0, 55]\n",
    "    results_brotec = [-5, -2, 1000, 4000, 32400, 36000, 0.7, 1.0, 49, 0.0, 49]#[-4, -1, 2000, 5500, 32400, 43200, 0.2, 0.8, 50, 0.2, 50]\n",
    "    results_arttri = [-4, 0, 1000, 2500, 32400, 36000, 0.6, 4.3, 49, 0.0, 49]#[-5, -1, 2000, 4000, 32400, 43200, 0.1, 0.4, 59, 0.2, 59]\n",
    "    results_ilevom = [-3, 7, 1000, 6000, 39600, 43200, 1.2, 4.4, 49, 0.5, 49]#[-3, 1, 1000, 8500, 36000, 43200, 1.1, 4.5, 50, 0.6, 50]\n",
    "    results_broine = [-1, 8, 1000, 2000, 32400, 43200, 0.5, 1.9, 21, 0.1, 21]#[-1, 3, 1500, 7500, 39600, 46800, 0.3, 0.5, 25, 0.4, 25]\n",
    "    results_phycap = [-5, -1, 2000, 4000, 32400, 43200, 0.1, 0.4, 42, 0.2, 42]#[-5, -2, 500, 2000, 36000, 46800, 1.2, 2.6, 48, 0.2, 48]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Store the 'best' parameters for each site/species "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "CA_AdeFasBestParams = results_adefas # Adenostam fasciculatum\n",
    "UT_ArtTriBestParams = results_arttri # Artemesia tridentata\n",
    "UT_BroTecBestParams = results_brotec # Bromus tectorum\n",
    "TX_IleVomBestParams = results_ilevom # Ilex vomitoria\n",
    "SD_BroIneBestParams = results_broine # Bromus inermis\n",
    "ID_PhyCapBestParams = results_phycap # Physiocarpus capitatus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#MakeGSILFMCompareTable4Param(Met413509,TX_IleVom,[-4, -3, 2000, 6000, 39600, 43200, 0.5, 3.3, 45, 0.7, 45],label=\"TX_IleVomOpt\",UseLFMMinMax = True,Herb=False,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=Lat413509)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 6, 500, 7000, 32400, 36000, 1.2, 4.8, 56, 0.0, 56]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Chamise Parameters\n",
    "results_adefas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-5, -2, 1000, 4000, 32400, 36000, 0.7, 1.0, 49, 0.0, 49]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cheatgrass Parameters\n",
    "results_brotec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-4, 0, 1000, 2500, 32400, 36000, 0.6, 4.3, 49, 0.0, 49]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sagebrush Parameters\n",
    "results_arttri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-3, 7, 1000, 6000, 39600, 43200, 1.2, 4.4, 49, 0.5, 49]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Yaupon Parameters\n",
    "results_ilevom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-1, 8, 1000, 2000, 32400, 43200, 0.5, 1.9, 21, 0.1, 21]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Smooth Brome parameters\n",
    "results_broine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-5, -1, 2000, 4000, 32400, 43200, 0.1, 0.4, 42, 0.2, 42]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ninebark Parameters\n",
    "results_phycap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create tables and figures from the NFDRS V4 tech doc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6381955722489746\n",
      "0.8756704883622166\n",
      "0.5467308942649306\n",
      "0.7386899743386844\n",
      "0.5132834208839246\n",
      "0.7162495275313058\n",
      "0.20450188363337793\n",
      "0.46075964802954106\n",
      "0.6195839127766164\n",
      "0.5827159521408624\n",
      "0.44108235448895805\n",
      "0.6001318783618874\n"
     ]
    }
   ],
   "source": [
    "# Cheatgrass\n",
    "temp1a =  MakeGSILFMCompareTable4Param(Met421501,UT_BroTec,defParams,label=\"UT_BroTecDefaults\",UseLFMMinMax = False,Herb=True,Lat=Lat421501)\n",
    "temp1b = MakeGSILFMCompareTable4Param(Met421501,UT_BroTec,UT_BroTecBestParams,label=\"UT_BroTecGSO\",UseLFMMinMax = True,Herb=True,Lat=Lat421501)\n",
    "\n",
    "# Sagebrush\n",
    "temp2a =  MakeGSILFMCompareTable4Param(Met421501,UT_ArtTri,defParams,label=\"UT_ArtTriDefaults\",UseLFMMinMax = False,Herb=False,Lat=Lat421501)\n",
    "temp2b =  MakeGSILFMCompareTable4Param(Met421501,UT_ArtTri,UT_ArtTriBestParams,label=\"UT_ArtTriGSO\",UseLFMMinMax = True,Herb=False,Lat=Lat421501)\n",
    "\n",
    "# Chamise\n",
    "temp3a =  MakeGSILFMCompareTable4Param(Met045433,CA_AdeFas,defParams,label=\"CA_AdeFasDefaults\",UseLFMMinMax = False,Herb=False,Lat=Lat045433)\n",
    "temp3b =  MakeGSILFMCompareTable4Param(Met045433,CA_AdeFas,CA_AdeFasBestParams,label=\"CA_AdeFasGSO\",UseLFMMinMax = True,Herb=False,Lat=Lat045433)\n",
    "\n",
    "# Yaupon\n",
    "temp4a =  MakeGSILFMCompareTable4Param(Met413509,TX_IleVom,defParams,label=\"TX_IleVomDefaults\",UseLFMMinMax = False,Herb=False,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=Lat413509)\n",
    "temp4b =  MakeGSILFMCompareTable4Param(Met413509,TX_IleVom,TX_IleVomBestParams,label=\"TX_IleVomGSO\",UseLFMMinMax = True,Herb=False,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=Lat413509)\n",
    "\n",
    "# Smooth Brome \n",
    "temp5a =  MakeGSILFMCompareTable4Param(MetSSFS,SD_BroIne,defParams,label=\"SD_BroIneDefaults\",UseLFMMinMax = False,Herb=True,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=LatSSFS)\n",
    "temp5b =  MakeGSILFMCompareTable4Param(MetSSFS,SD_BroIne,SD_BroIneBestParams,label=\"SD_BroIneGSO\",UseLFMMinMax = True,Herb=True,PLowLim='2020-01-01',PUpperLim='2022-12-31',Lat=LatSSFS)\n",
    "\n",
    "# Ninebark\n",
    "temp6a =  MakeGSILFMCompareTable4Param(Met101303,ID_PhyCap,defParams,label=\"ID_PhyCapDefaults\",UseLFMMinMax = False,Herb=False,PLowLim='2019-01-01',PUpperLim='2023-12-31',Lat=Lat101303)\n",
    "temp6b =  MakeGSILFMCompareTable4Param(Met101303,ID_PhyCap,ID_PhyCapBestParams,label=\"ID_PhyCapGSO\",UseLFMMinMax = True,Herb=False,PLowLim='2019-01-01',PUpperLim='2023-12-31',Lat=Lat101303)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reproduce stats for Table 10 from the Technical Document"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Site_Species</th>\n",
       "      <th>rho_GSI-LFMC</th>\n",
       "      <th>GSI-LFMC-pval</th>\n",
       "      <th>GSI-LFMC-MAE</th>\n",
       "      <th>GSI-LFMC-R2</th>\n",
       "      <th>rho_LFMC-V2</th>\n",
       "      <th>LFMC-V2-pval</th>\n",
       "      <th>LFMC-V2-MAE</th>\n",
       "      <th>LFMC-V2-R2</th>\n",
       "      <th>N</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>UT_BroTecDefaults</td>\n",
       "      <td>0.789</td>\n",
       "      <td>0.000</td>\n",
       "      <td>27.558</td>\n",
       "      <td>0.638</td>\n",
       "      <td>0.289</td>\n",
       "      <td>0.032</td>\n",
       "      <td>68.580</td>\n",
       "      <td>0.078</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>UT_BroTecGSO</td>\n",
       "      <td>0.924</td>\n",
       "      <td>0.000</td>\n",
       "      <td>23.857</td>\n",
       "      <td>0.876</td>\n",
       "      <td>0.289</td>\n",
       "      <td>0.032</td>\n",
       "      <td>68.580</td>\n",
       "      <td>0.078</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>UT_ArtTriDefaults</td>\n",
       "      <td>0.749</td>\n",
       "      <td>0.000</td>\n",
       "      <td>37.858</td>\n",
       "      <td>0.547</td>\n",
       "      <td>-0.165</td>\n",
       "      <td>0.228</td>\n",
       "      <td>59.627</td>\n",
       "      <td>0.000</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>UT_ArtTriGSO</td>\n",
       "      <td>0.880</td>\n",
       "      <td>0.000</td>\n",
       "      <td>38.023</td>\n",
       "      <td>0.739</td>\n",
       "      <td>-0.165</td>\n",
       "      <td>0.228</td>\n",
       "      <td>59.627</td>\n",
       "      <td>0.000</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA_AdeFasDefaults</td>\n",
       "      <td>0.586</td>\n",
       "      <td>0.000</td>\n",
       "      <td>17.847</td>\n",
       "      <td>0.513</td>\n",
       "      <td>0.180</td>\n",
       "      <td>0.062</td>\n",
       "      <td>25.868</td>\n",
       "      <td>0.007</td>\n",
       "      <td>108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>CA_AdeFasGSO</td>\n",
       "      <td>0.821</td>\n",
       "      <td>0.000</td>\n",
       "      <td>13.258</td>\n",
       "      <td>0.716</td>\n",
       "      <td>0.180</td>\n",
       "      <td>0.062</td>\n",
       "      <td>25.868</td>\n",
       "      <td>0.007</td>\n",
       "      <td>108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>TX_IleVomDefaults</td>\n",
       "      <td>0.381</td>\n",
       "      <td>0.001</td>\n",
       "      <td>34.164</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.467</td>\n",
       "      <td>0.000</td>\n",
       "      <td>30.180</td>\n",
       "      <td>0.195</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>TX_IleVomGSO</td>\n",
       "      <td>0.727</td>\n",
       "      <td>0.000</td>\n",
       "      <td>17.815</td>\n",
       "      <td>0.461</td>\n",
       "      <td>0.467</td>\n",
       "      <td>0.000</td>\n",
       "      <td>30.180</td>\n",
       "      <td>0.195</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>SD_BroIneDefaults</td>\n",
       "      <td>0.781</td>\n",
       "      <td>0.000</td>\n",
       "      <td>34.312</td>\n",
       "      <td>0.620</td>\n",
       "      <td>0.865</td>\n",
       "      <td>0.000</td>\n",
       "      <td>69.995</td>\n",
       "      <td>0.653</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>SD_BroIneGSO</td>\n",
       "      <td>0.881</td>\n",
       "      <td>0.000</td>\n",
       "      <td>46.315</td>\n",
       "      <td>0.583</td>\n",
       "      <td>0.865</td>\n",
       "      <td>0.000</td>\n",
       "      <td>69.995</td>\n",
       "      <td>0.653</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>ID_PhyCapDefaults</td>\n",
       "      <td>0.643</td>\n",
       "      <td>0.000</td>\n",
       "      <td>64.317</td>\n",
       "      <td>0.441</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.699</td>\n",
       "      <td>78.708</td>\n",
       "      <td>0.000</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>ID_PhyCapGSO</td>\n",
       "      <td>0.792</td>\n",
       "      <td>0.000</td>\n",
       "      <td>36.018</td>\n",
       "      <td>0.600</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.699</td>\n",
       "      <td>78.708</td>\n",
       "      <td>0.000</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Site_Species  rho_GSI-LFMC  GSI-LFMC-pval  GSI-LFMC-MAE  GSI-LFMC-R2  \\\n",
       "0   UT_BroTecDefaults         0.789          0.000        27.558        0.638   \n",
       "1        UT_BroTecGSO         0.924          0.000        23.857        0.876   \n",
       "2   UT_ArtTriDefaults         0.749          0.000        37.858        0.547   \n",
       "3        UT_ArtTriGSO         0.880          0.000        38.023        0.739   \n",
       "4   CA_AdeFasDefaults         0.586          0.000        17.847        0.513   \n",
       "5        CA_AdeFasGSO         0.821          0.000        13.258        0.716   \n",
       "6   TX_IleVomDefaults         0.381          0.001        34.164        0.205   \n",
       "7        TX_IleVomGSO         0.727          0.000        17.815        0.461   \n",
       "8   SD_BroIneDefaults         0.781          0.000        34.312        0.620   \n",
       "9        SD_BroIneGSO         0.881          0.000        46.315        0.583   \n",
       "10  ID_PhyCapDefaults         0.643          0.000        64.317        0.441   \n",
       "11       ID_PhyCapGSO         0.792          0.000        36.018        0.600   \n",
       "\n",
       "    rho_LFMC-V2  LFMC-V2-pval  LFMC-V2-MAE  LFMC-V2-R2    N  \n",
       "0         0.289         0.032       68.580       0.078   55  \n",
       "1         0.289         0.032       68.580       0.078   55  \n",
       "2        -0.165         0.228       59.627       0.000   55  \n",
       "3        -0.165         0.228       59.627       0.000   55  \n",
       "4         0.180         0.062       25.868       0.007  108  \n",
       "5         0.180         0.062       25.868       0.007  108  \n",
       "6         0.467         0.000       30.180       0.195   69  \n",
       "7         0.467         0.000       30.180       0.195   69  \n",
       "8         0.865         0.000       69.995       0.653   21  \n",
       "9         0.865         0.000       69.995       0.653   21  \n",
       "10        0.065         0.699       78.708       0.000   38  \n",
       "11        0.065         0.699       78.708       0.000   38  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = [temp1a,temp1b,temp2a,temp2b,temp3a,temp3b,temp4a,temp4b,temp5a,temp5b,temp6a,temp6b]\n",
    "pd.DataFrame(data, columns=[\"Site_Species\",\"rho_GSI-LFMC\",\"GSI-LFMC-pval\",\"GSI-LFMC-MAE\",\"GSI-LFMC-R2\",\"rho_LFMC-V2\",\"LFMC-V2-pval\",\"LFMC-V2-MAE\",\"LFMC-V2-R2\",\"N\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "MM_UT_BroTecOpt = GetMetMerge(Met421501, UT_BroTec,UT_BroTecBestParams,UseLFMMinMax=True,Lat=Lat421501,Herb=True)\n",
    "MM_UT_ArtTriOpt = GetMetMerge(Met421501, UT_ArtTri,UT_ArtTriBestParams,UseLFMMinMax=True,Lat=Lat421501)\n",
    "MM_CA_AdeFasOpt = GetMetMerge(Met045433,CA_AdeFas,results_adefas,UseLFMMinMax=True,Lat=Lat045433)\n",
    "MM_TX_IleVomOpt = GetMetMerge(Met413509,TX_IleVom,results_ilevom,UseLFMMinMax=True,Lat=Lat413509,PLowLim='2020-01-01',PUpperLim = '2023-12-31')\n",
    "MM_SD_BroIneOpt = GetMetMerge(MetSSFS,SD_BroIne,results_broine,UseLFMMinMax=True,Lat=LatSSFS,PLowLim='2020-01-01',PUpperLim = '2022-12-31')\n",
    "MM_ID_PhyCapOpt = GetMetMerge(Met101303,ID_PhyCap,ID_PhyCapBestParams,UseLFMMinMax=True,Lat=Lat101303,PLowLim='2019-01-01',PUpperLim = '2023-12-31')\n",
    "\n",
    "# Assign either the Herb or Woody FM from Version 2 to the column \"FM\" for comparison across fuel types\n",
    "MM_UT_BroTecOpt['FM'] = MM_UT_BroTecOpt['FMH']\n",
    "MM_UT_ArtTriOpt['FM'] = MM_UT_ArtTriOpt['FMW']\n",
    "MM_CA_AdeFasOpt['FM'] = MM_CA_AdeFasOpt['FMW']\n",
    "MM_TX_IleVomOpt['FM'] = MM_TX_IleVomOpt['FMW']\n",
    "MM_SD_BroIneOpt['FM'] = MM_SD_BroIneOpt['FMH']\n",
    "MM_ID_PhyCapOpt['FM'] = MM_ID_PhyCapOpt['FMW']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LFMC Obs vs GSI-LFMC Sample Size Total:  346\n"
     ]
    }
   ],
   "source": [
    "print(\"LFMC Obs vs GSI-LFMC Sample Size Total: \",\n",
    "      MM_UT_BroTecOpt.index.size + MM_UT_ArtTriOpt.index.size + \n",
    "      MM_CA_AdeFasOpt.index.size + MM_TX_IleVomOpt.index.size + \n",
    "      MM_SD_BroIneOpt.index.size + MM_ID_PhyCapOpt.index.size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "MM_UT_BroTecOpt['DateTime'] = pd.to_datetime(MM_UT_BroTecOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')\n",
    "MM_UT_ArtTriOpt['DateTime'] = pd.to_datetime(MM_UT_ArtTriOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')\n",
    "MM_CA_AdeFasOpt['DateTime'] = pd.to_datetime(MM_CA_AdeFasOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')\n",
    "MM_TX_IleVomOpt['DateTime'] = pd.to_datetime(MM_TX_IleVomOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')\n",
    "MM_SD_BroIneOpt['DateTime'] = pd.to_datetime(MM_SD_BroIneOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')\n",
    "MM_ID_PhyCapOpt['DateTime'] = pd.to_datetime(MM_ID_PhyCapOpt['DateTime'],format='%Y-%m-%d %H:%M:%S')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Figure 5 from the NFDRS V4 technical document paper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x1200 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f,ax = plt.subplots(6,1,figsize=(5,12))\n",
    "## Plot BroTec\n",
    "MM_UT_BroTecOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[0],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf)\n",
    "MM_UT_BroTecOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[0],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf)\n",
    "ax[0].set_xlim([datetime(2015, 1, 1), datetime(2019, 12, 31)])\n",
    "ax[0].set_xlabel(\"\")\n",
    "ax[0].set_title(\"[A] Bromus tectorum (Cheatgrass)\", style='italic')\n",
    "\n",
    "# Define the date format\n",
    "date_form = DateFormatter(\"%Y\")\n",
    "ax[0].xaxis.set_major_formatter(date_form)\n",
    "# Setting the number of ticks\n",
    "ax[0].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[0].set_xlabel('')\n",
    "ax[0].set_ylabel('LFMC (% dry wt)')\n",
    "\n",
    "## Plot ArtTri\n",
    "ax[1].set_title(\"[B] Artemisia tridentata (Basin Big Sagebrush)\", style='italic')\n",
    "MM_UT_ArtTriOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[1],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf)\n",
    "MM_UT_ArtTriOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[1],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf)\n",
    "ax[1].set_xlim([datetime(2015, 1, 1), datetime(2019, 12, 31)])\n",
    "ax[1].xaxis.set_major_formatter(date_form)\n",
    "ax[1].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[1].set_xlabel('')\n",
    "ax[1].set_ylabel('LFMC (% dry wt)')\n",
    "\n",
    "## Plot AdeFas\n",
    "ax[2].set_title(\"[C] Adenostoma fasciculatum (Chamise)\", style='italic')\n",
    "MM_CA_AdeFasOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[2],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf,label='Modeled LFMC',legend=False)\n",
    "MM_CA_AdeFasOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[2],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf,label='Measured LFMC',legend=False)\n",
    "ax[2].set_xlim([datetime(2015, 1, 1), datetime(2019, 12, 31)])\n",
    "ax[2].set_xlabel('')\n",
    "ax[2].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[2].xaxis.set_major_formatter(date_form)\n",
    "ax[2].set_xlabel('')\n",
    "ax[2].set_ylabel('LFMC (% dry wt)')\n",
    "\n",
    "## Plot IleVom\n",
    "plotind = 3\n",
    "ax[plotind].set_title(\"[D] Ilex vomitoria (Yaupon)\", style='italic')\n",
    "MM_TX_IleVomOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[plotind],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf,label='Modeled LFMC',legend=False)\n",
    "MM_TX_IleVomOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[plotind],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf,label='Measured LFMC',legend=False)\n",
    "ax[plotind].set_xlim([datetime(2020, 1, 1), datetime(2022, 12, 31)])\n",
    "ax[plotind].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[plotind].xaxis.set_major_formatter(date_form)\n",
    "ax[plotind].set_xlabel(\"\")\n",
    "ax[plotind].set_ylabel('LFMC (% dry wt)')\n",
    "\n",
    "## Plot BroIne\n",
    "plotind = 4\n",
    "ax[plotind].set_title(\"[E] Bromus inermis (Smooth Brome)\", style='italic')\n",
    "MM_SD_BroIneOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[plotind],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf,label='Modeled LFMC',legend=False)\n",
    "MM_SD_BroIneOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[plotind],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf,label='Measured LFMC',legend=False)\n",
    "ax[plotind].set_xlim([datetime(2020, 1, 1), datetime(2022, 12, 31)])\n",
    "ax[plotind].set_xlabel(\"Date\")\n",
    "ax[plotind].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[plotind].xaxis.set_major_formatter(date_form)\n",
    "ax[plotind].set_xlabel('Date')\n",
    "ax[plotind].set_ylabel('LFMC (% dry wt)')\n",
    "\n",
    "\n",
    "## Plot PhyCap\n",
    "plotind = 5\n",
    "ax[plotind].set_title(\"[F] Physiocarpus capitatus (Pacific Ninebark)\", style='italic')\n",
    "MM_ID_PhyCapOpt.plot.scatter(x='DateTime',y='LFMC',ax=ax[plotind],marker=mkr_nf1,color=clr_nf1, edgecolor=clr_edg_nf,label='Modeled LFMC')\n",
    "MM_ID_PhyCapOpt.plot.scatter(x='DateTime',y='Percent',ax=ax[plotind],marker=mkr_nf2,color=clr_nf2, edgecolor=clr_edg_nf,label='Measured LFMC')\n",
    "ax[plotind].set_xlim([datetime(2020, 1, 1), datetime(2022, 12, 31)])\n",
    "ax[plotind].set_xlabel(\"Date\")\n",
    "ax[plotind].xaxis.set_major_locator(plt.MaxNLocator(7))\n",
    "ax[plotind].xaxis.set_major_formatter(date_form)\n",
    "ax[plotind].set_xlabel('Date')\n",
    "ax[plotind].set_ylabel('LFMC (% dry wt)')\n",
    "ax[5].legend(loc='upper center', bbox_to_anchor=(0.5, -0.37),\n",
    "          fancybox=True, shadow=True, ncol=5)\n",
    "\n",
    "## Write out the figure\n",
    "f.tight_layout()\n",
    "fout = \"./outputs/CombinedGSILFMPlot6panel.png\" \n",
    "if fout != \"\":\n",
    "    plt.savefig(fout,dpi=300, bbox_inches='tight',facecolor='white', transparent=False )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Note these use the default parameters for each species\n",
    "MM_UT_BroTec = GetMetMerge(Met421501, UT_BroTec,defParams,UseLFMMinMax=True,Lat=Lat421501,Herb=True)\n",
    "MM_UT_ArtTri = GetMetMerge(Met421501, UT_ArtTri,defParams,UseLFMMinMax=True,Lat=Lat421501)\n",
    "MM_CA_AdeFas = GetMetMerge(Met045433,CA_AdeFas,defParams,UseLFMMinMax=True,Lat=Lat045433)\n",
    "MM_TX_IleVom = GetMetMerge(Met413509,TX_IleVom,defParams,UseLFMMinMax=True,Lat=Lat413509,PLowLim='2020-01-01',PUpperLim = '2023-12-31')\n",
    "MM_SD_BroIne = GetMetMerge(MetSSFS,SD_BroIne,defParams,UseLFMMinMax=True,Lat=LatSSFS,PLowLim='2020-01-01',PUpperLim = '2022-12-31')\n",
    "MM_ID_PhyCap = GetMetMerge(Met101303,ID_PhyCap,defParams,UseLFMMinMax=True,Lat=Lat101303,PLowLim='2019-01-01',PUpperLim = '2023-12-31')\n",
    "\n",
    "# Fix the first comparions against V2 LFMC's\n",
    "MM_UT_BroTec['FM'] = MM_UT_BroTec['FMH']\n",
    "MM_UT_ArtTri['FM'] = MM_UT_ArtTri['FMW']\n",
    "MM_CA_AdeFas['FM'] = MM_CA_AdeFas['FMW']\n",
    "MM_TX_IleVom['FM'] = MM_TX_IleVom['FMW']\n",
    "MM_SD_BroIne['FM'] = MM_SD_BroIne['FMH']\n",
    "MM_ID_PhyCap['FM'] = MM_ID_PhyCap['FMW']\n",
    "\n",
    "# Create a simple dataframe for the correlation for both the default and Grid Search Op\n",
    "MM_GSIDef = pd.concat([MM_UT_BroTec,MM_UT_ArtTri,MM_CA_AdeFas,MM_TX_IleVom,MM_SD_BroIne,MM_ID_PhyCap])\n",
    "MM_GSIDef = MM_GSIDef[['Percent','LFMC','FM']]\n",
    "\n",
    "MM_GSIOpt= pd.concat([MM_UT_BroTecOpt,MM_UT_ArtTriOpt,MM_CA_AdeFasOpt,MM_TX_IleVomOpt,MM_SD_BroIneOpt,MM_ID_PhyCapOpt])\n",
    "MM_GSIOpt = MM_GSIOpt[['Percent','LFMC','FM']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "MM_Herb_GSIOpt =  pd.concat([MM_UT_BroTecOpt,MM_SD_BroIneOpt])\n",
    "MM_Wood_GSIOpt = pd.concat([MM_UT_ArtTriOpt,MM_CA_AdeFasOpt,MM_TX_IleVomOpt,MM_ID_PhyCapOpt])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ranges of values for measured woody LFMC for Yaupon"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00     91.0\n",
       "0.05    104.4\n",
       "0.50    121.0\n",
       "0.95    151.0\n",
       "1.00    186.0\n",
       "Name: Percent, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MM_TX_IleVom.Percent.quantile([0,0.05,0.5,0.95,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ranges of values for measured herbaceous LFMC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00      4.00\n",
       "0.05     10.25\n",
       "0.50     74.50\n",
       "0.95    184.00\n",
       "1.00    279.00\n",
       "Name: Percent, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MM_Herb_GSIOpt.Percent.quantile([0,0.05,0.5,0.95,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ranges of values for measured woody LFMC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00     55.0\n",
       "0.05     59.0\n",
       "0.50    108.0\n",
       "0.95    196.0\n",
       "1.00    304.0\n",
       "Name: Percent, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MM_Wood_GSIOpt.Percent.quantile([0,0.05,0.5,0.95,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.7739404438857747"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2_score(MM_GSIDef['Percent'],MM_GSIDef['FM'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Correlations across all species using the Default GSI parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Percent</th>\n",
       "      <th>LFMC</th>\n",
       "      <th>FM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Percent</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.779313</td>\n",
       "      <td>0.329860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LFMC</th>\n",
       "      <td>0.779313</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.473494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FM</th>\n",
       "      <td>0.329860</td>\n",
       "      <td>0.473494</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Percent      LFMC        FM\n",
       "Percent  1.000000  0.779313  0.329860\n",
       "LFMC     0.779313  1.000000  0.473494\n",
       "FM       0.329860  0.473494  1.000000"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Pooled correlation for all three species using default parameters\n",
    "MM_GSIDef.corr(method='spearman')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'p_value' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[42], line 2\u001b[0m\n\u001b[0;32m      1\u001b[0m slope, intercept, r_value_fm, p_value_def, std_err \u001b[38;5;241m=\u001b[39m linregress(MM_GSIDef[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mPercent\u001b[39m\u001b[38;5;124m'\u001b[39m],MM_GSIDef[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mFM\u001b[39m\u001b[38;5;124m'\u001b[39m])\n\u001b[1;32m----> 2\u001b[0m \u001b[38;5;28mprint\u001b[39m(r_value_fm\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m2\u001b[39m,\u001b[43mp_value\u001b[49m)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'p_value' is not defined"
     ]
    }
   ],
   "source": [
    "slope, intercept, r_value_fm, p_value_def, std_err = linregress(MM_GSIDef['Percent'],MM_GSIDef['FM'])\n",
    "print(r_value_fm**2,p_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "slope, intercept, r_value_def, p_value_def, std_err = linregress(MM_GSIDef['Percent'],MM_GSIDef['LFMC'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "r_value_def**2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Correlations across all species using the Grid Search Optimized GSI parameters at each site"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "MM_GSIOpt.corr(method='spearman')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "slope, intercept, r_value_opt, p_value, std_err = linregress(MM_GSIOpt['Percent'],MM_GSIOpt['LFMC'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "r_value_opt**2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Figure 4 from the NFDRFS V4 technical document paper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f,ax = plt.subplots(3,1,figsize=(5,12))\n",
    "MM_UT_BroTec.plot.scatter(x='Percent',y='FMH',ax=ax[0],label=\"BroTec\",marker=mkr_nf1,color=clr_nf1,edgecolor=clr_edg_nf)\n",
    "MM_UT_ArtTri.plot.scatter(x='Percent',y='FMW',ax=ax[0],label=\"ArtTri\",marker=mkr_nf2,color=clr_nf2,edgecolor=clr_edg_nf)\n",
    "MM_CA_AdeFas.plot.scatter(x='Percent',y='FMW',ax=ax[0],label=\"AdeFas\",marker=mkr_nf3,color=clr_nf3,edgecolor=clr_edg_nf)\n",
    "MM_TX_IleVom.plot.scatter(x='Percent',y='FMW',ax=ax[0],label=\"IleVom\",marker=mkr_nf4,color=clr_nf4,edgecolor=clr_edg_nf)\n",
    "MM_SD_BroIne.plot.scatter(x='Percent',y='FMW',ax=ax[0],label=\"BroIne\",marker=mkr_nf5,color=clr_nf5,edgecolor=clr_edg_nf)\n",
    "MM_ID_PhyCap.plot.scatter(x='Percent',y='FMW',ax=ax[0],label=\"PhyCap\",marker=mkr_nf6,color=clr_nf6,edgecolor=clr_edg_nf)\n",
    "ax[0].plot([0,250],[0,250],color='black')\n",
    "ax[0].legend(loc='upper left')\n",
    "ax[0].set_xlabel(\"\")\n",
    "ax[0].set_ylabel(\"Modeled LFMC (% dry wt)\")\n",
    "ax[0].set_title('NFDRS Version 2 LFMC Model')\n",
    "\n",
    "MM_UT_BroTec.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf1,color=clr_nf1,edgecolor=clr_edg_nf)\n",
    "MM_UT_ArtTri.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf2,color=clr_nf2,edgecolor=clr_edg_nf)\n",
    "MM_CA_AdeFas.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf3,color=clr_nf3,edgecolor=clr_edg_nf)\n",
    "MM_TX_IleVom.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf4,color=clr_nf4,edgecolor=clr_edg_nf)\n",
    "MM_SD_BroIne.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf5,color=clr_nf5,edgecolor=clr_edg_nf)\n",
    "MM_ID_PhyCap.plot.scatter(x='Percent',y='LFMC',ax=ax[1],marker=mkr_nf6,color=clr_nf6,edgecolor=clr_edg_nf)\n",
    "ax[1].plot([0,250],[0,250],color='black')\n",
    "#ax[1].legend(loc='lower left',bbox_to_anchor=(1.0, 1.0))\n",
    "ax[1].set_xlabel(\"\")\n",
    "ax[1].set_ylabel(\"Modeled LFMC (% dry wt)\")\n",
    "ax[1].set_title('NFDRS Version 4 Default GSI Parameters')\n",
    "\n",
    "MM_UT_BroTecOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf1,color=clr_nf1,edgecolor=clr_edg_nf)\n",
    "MM_UT_ArtTriOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf2,color=clr_nf2,edgecolor=clr_edg_nf)\n",
    "MM_CA_AdeFasOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf3,color=clr_nf3,edgecolor=clr_edg_nf)\n",
    "MM_TX_IleVomOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf4,color=clr_nf4,edgecolor=clr_edg_nf)\n",
    "MM_SD_BroIneOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf5,color=clr_nf5,edgecolor=clr_edg_nf)\n",
    "MM_ID_PhyCapOpt.plot.scatter(x='Percent',y='LFMC',ax=ax[2],marker=mkr_nf6,color=clr_nf6,edgecolor=clr_edg_nf)\n",
    "ax[2].plot([0,250],[0,250],color='black')\n",
    "#ax[1].legend(loc='lower left',bbox_to_anchor=(1.0, 1.0))\n",
    "ax[2].set_xlabel(\"Measured LFMC (% dry wt)\")\n",
    "ax[2].set_ylabel(\"Modeled LFMC (% dry wt)\")\n",
    "ax[2].set_title('NFDRS Version 4 Grid-Search Optimized GSI Parameters')\n",
    "\n",
    "plt.tight_layout()\n",
    "fout = './outputs/FuelMoistureComparisons_3Grid.png'\n",
    "if fout != \"\":\n",
    "    plt.savefig(fout,dpi=300, bbox_inches='tight',facecolor='white', transparent=False )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The following code is just to demonstrate some aspects of the GSI calculations such as differences in the cumulative precip between locations and choice of threshold and smoothing period for optimizations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Met421501RunSumPrcp = Met421501.RunSumPrcp.dropna()\n",
    "Met045433RunSumPrcp = Met045433.RunSumPrcp.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the cumulative proportion of the data that falls below each value\n",
    "cumulative = np.linspace(0, 1, len(Met421501RunSumPrcp))\n",
    "cumulative2 = np.linspace(0, 1, len(Met045433RunSumPrcp))\n",
    "# Sort the data in ascending order\n",
    "sorted_data = np.sort(Met421501RunSumPrcp)\n",
    "sorted_data2 = np.sort(Met045433RunSumPrcp)\n",
    "\n",
    "# Calculate the cumulative proportion of the sorted data\n",
    "cumulative_data = np.cumsum(sorted_data) / np.sum(sorted_data)\n",
    "cumulative_data2 = np.cumsum(sorted_data2) / np.sum(sorted_data2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot the CDF\n",
    "plt.plot(sorted_data, cumulative_data,label='Servier Reservoir, UT')\n",
    "plt.plot(sorted_data2, cumulative_data2,label='Beverly Hills, CA')\n",
    "# Add labels and title\n",
    "plt.xlabel(\"Running Total Precip (in)\")\n",
    "plt.ylabel(\"Cumulative Proportion\")\n",
    "plt.legend()\n",
    "plt.title(\"Cumulative Distribution Function (CDF) of Running Total Precip (in)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Look up the percentiles of Cum Rainfall\n",
    "print(stats.percentileofscore(Met421501.RunSumPrcp.dropna(), 0.393701))\n",
    "print(stats.percentileofscore(Met421501.RunSumPrcp.dropna(), 1.1))\n",
    "print(stats.percentileofscore(Met045433.RunSumPrcp.dropna(), 0.393701))\n",
    "print(stats.percentileofscore(Met045433.RunSumPrcp.dropna(), 3.6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(stats.percentileofscore(Met421501.RunSumPrcp.dropna(), 1.1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "SIM = False\n",
    "if SIM:\n",
    "    # Test the default Green-up threshold \n",
    "    corrs = []\n",
    "    DayRange = np.arange(7,100,7,dtype='int')\n",
    "    for i in DayRange:\n",
    "        print('Days: ', i)\n",
    "        for j in range(0,10):\n",
    "            GU = j / 10\n",
    "            Days = int(i)\n",
    "            #defParams = [-2, 5, 900, 4100, 36000, 39600, 0, 0.393701, Days,GU,Days]\n",
    "            \n",
    "            MM_UT_BroTec = GetMetMerge(Met421501, UT_BroTec,defParams,UseLFMMinMax=False,Lat=Lat421501, Herb=True)\n",
    "            MM_UT_ArtTri = GetMetMerge(Met421501, UT_ArtTri,defParams,UseLFMMinMax=False,Lat=Lat421501)\n",
    "            MM_CA_AdeFas = GetMetMerge(Met045433,CA_AdeFas,defParams,UseLFMMinMax=False,Lat=Lat045433)\n",
    "            \n",
    "            MM_GSIDef = pd.concat([MM_UT_BroTec,MM_UT_ArtTri,MM_CA_AdeFas])\n",
    "            MM_GSIDef = MM_GSIDef.drop(columns=['GACC','State','Group','Site','Date','Fuel','Unnamed: 7','DateTime',\n",
    "                                                'DATE','Tmin','MaxT','MnRH','MxRH','VPDMax','VPDA','Prcp',\n",
    "                                                'RunSumPrcp','MinTC','Dayl','JDay','GSI_PE_RS',\n",
    "                                               'TminInd','VPDInd','DaylInd','PrcpInd','FMW','FMH'])\n",
    "            # Append the correlation to the list\n",
    "            corrs.append(MM_GSIDef.corr(method='spearman')['Percent'][1])\n",
    "\n",
    "if SIM:\n",
    "    arr = np.array(corrs)\n",
    "    # Use around() to round the array\n",
    "    rounded_corrs = np.around(arr,decimals=3)    \n",
    "else:\n",
    "    rounded_corrs = [0.552, 0.543, 0.5  , 0.48 , 0.466, 0.424, 0.394, 0.333, 0.293, 0.276, 0.691, 0.669, 0.646, 0.624, 0.571, 0.525, 0.435, 0.368,\n",
    "       0.314, 0.292, 0.766, 0.765, 0.753, 0.724, 0.694, 0.604, 0.511, 0.412, 0.332, 0.299, 0.786, 0.791, 0.808, 0.779, 0.756, 0.719,\n",
    "       0.606, 0.538, 0.335, 0.291, 0.771, 0.778, 0.802, 0.793, 0.785, 0.731, 0.606, 0.549, 0.436, 0.297, 0.755, 0.773, 0.795, 0.784,\n",
    "       0.796, 0.796, 0.696, 0.55 , 0.526, 0.317, 0.722, 0.744, 0.763, 0.768, 0.768, 0.765, 0.737, 0.538, 0.489, 0.338, 0.697, 0.726,\n",
    "       0.747, 0.749, 0.756, 0.767, 0.732, 0.576, 0.471, 0.339, 0.66 , 0.696, 0.708, 0.727, 0.728, 0.73 , 0.704, 0.571, 0.42 , 0.298,\n",
    "       0.613, 0.639, 0.656, 0.674, 0.682, 0.682, 0.66 , 0.502, 0.389, 0.3  , 0.568, 0.592, 0.611, 0.631, 0.639, 0.642, 0.626, 0.491,\n",
    "       0.393, 0.329, 0.517, 0.535, 0.556, 0.576, 0.602, 0.603, 0.565, 0.488, 0.41 , 0.332, 0.461, 0.478, 0.498, 0.516, 0.543, 0.549,\n",
    "       0.516, 0.472, 0.377, 0.332, 0.4  , 0.419, 0.438, 0.46 , 0.489, 0.508, 0.459, 0.471, 0.408, 0.343]\n",
    "X = np.arange(0, 1, 0.1)\n",
    "Y = np.arange(7,100,7,dtype='int')\n",
    "corrs_new = np.array(rounded_corrs).reshape(len(Y), len(X))\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "CS = ax.contour(X, Y, corrs_new)\n",
    "ax.clabel(CS, inline=True, fontsize=10)\n",
    "ax.set_title('Default parameters correlation')\n",
    "ax.set_xlabel(\"Green-up Threshold (dim)\")\n",
    "ax.set_ylabel(\"Smoothing Period (days)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  },
  "toc-autonumbering": true
 },
 "nbformat": 4,
 "nbformat_minor": 4
}