BRDF_test.ipynb

{
 "cells": [
  {
   "cell_type": "raw",
   "id": "14782a8d-866d-43f4-9d12-3389cf61393c",
   "metadata": {},
   "source": [
    "# BRDF AHI to MODIS"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7fb7dcc-cdb0-42f1-904f-4d220a556b04",
   "metadata": {},
   "source": [
    "# Import libary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5ff8b134-8ec6-4ceb-87ed-c5c8f0d6d804",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sasktran as sk\n",
    "import numpy as np\n",
    "import datetime as dt\n",
    "import time as T\n",
    "import cv2\n",
    "from joblib import Parallel, delayed\n",
    "import numba as nb\n",
    "import os\n",
    "import paramiko\n",
    "from scp import SCPClient\n",
    "import subprocess\n",
    "from scipy.stats import gaussian_kde\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.basemap import Basemap\n",
    "import matplotlib.transforms as mtransforms"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb3c9112-1976-4548-8595-428538764d94",
   "metadata": {},
   "source": [
    "# Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e9da6bb5-24bc-4f50-b8cc-68df3b89def3",
   "metadata": {},
   "outputs": [],
   "source": [
    "class AHI_angle:\n",
    "    def __init__(self,date,):\n",
    "        self.date = date      \n",
    "        \n",
    "    def read_angle_data(self):\n",
    "    \n",
    "        AHI_date = self.date[4:11]\n",
    "        \n",
    "        sza_file_name = 'AHI_SZA_2018{}5.dat'.format(AHI_date)\n",
    "        saa_file_name = 'AHI_SAA_2018{}5.dat'.format(AHI_date)\n",
    "        \n",
    "        with open(SZA_PATH + sza_file_name,'rb') as fp:\n",
    "            AHI_SZA = np.frombuffer(fp.read(),dtype='u2').reshape(3000,3000) / 100\n",
    "            AHI_SZA=cv2.resize(np.array(AHI_SZA,dtype='float64'),(12000,12000),interpolation=cv2.INTER_NEAREST)\n",
    "        \n",
    "        with open(SAA_PATH + saa_file_name,'rb') as fp:\n",
    "            AHI_SAA = np.frombuffer(fp.read(),dtype='u2').reshape(3000,3000) / 100\n",
    "            AHI_SAA=cv2.resize(np.array(AHI_SAA,dtype='float64'),(12000,12000),interpolation=cv2.INTER_NEAREST)        \n",
    "        return AHI_SZA,AHI_SAA\n",
    "class H8_data:\n",
    "\n",
    "    def __init__(self , account , pw , band , band_number , date):\n",
    "        self.account = account\n",
    "        self.pw = pw\n",
    "        self.band = band\n",
    "        self.band_number = band_number\n",
    "        self.date = date\n",
    "    \n",
    "    def get_path(self):\n",
    "        return '/data01/GEO/ORGDATA/H8AHI/hmwr829gr.cr.chiba-u.ac.jp/gridded/FD/V20151105/' + self.date[0:6] + '/' + self.band.upper() + '/'\n",
    "\n",
    "    def get_filename(self):\n",
    "        return self.date + \".\" + self.band + \".\" + self.band_number + \".fld.geoss.bz2\"\n",
    "    \n",
    "    def DN2TBB(self,data):\n",
    "        LUT=np.loadtxt(DN_PATH + 'count2tbb_v102/' + self.band + \".\" + self.band_number)\n",
    "        return LUT[data,1]\n",
    "    \n",
    "    def file_path(self):\n",
    "        return self.get_path() + self.get_filename() \n",
    "                 \n",
    "    def download_H8data(self):\n",
    "        client = paramiko.SSHClient()\n",
    "        client.load_system_host_keys()\n",
    "        client.set_missing_host_key_policy(paramiko.AutoAddPolicy())\n",
    "        client.connect(hostname='10.4.123.47', port=22, username=self.account, password=self.pw)\n",
    "        scp = SCPClient(client.get_transport())\n",
    "        sftp = client.open_sftp()\n",
    "\n",
    "        try :\n",
    "            sftp.stat(self.file_path())\n",
    "\n",
    "        except FileNotFoundError:\n",
    "            print(\"File Not Found\")\n",
    "            return 'No data'\n",
    "\n",
    "        else:\n",
    "            scp.get(self.file_path(), folder_original+'/')\n",
    "            p = subprocess.Popen('lbzip2 -d {}{}'.format(folder_original+'/',self.file_path()[-33:]),shell=True)\n",
    "            p.communicate()\n",
    "            print ('Himawari8/AHI data Processed Finish')\n",
    "            return folder_original + '/' + self.get_filename()[:-4]\n",
    "            \n",
    "    def read_H8data(self):\n",
    "        H8_file_path = self.download_H8data()\n",
    "        if self.band == \"vis\":\n",
    "            sr = 12000\n",
    "        elif self.band == \"ext\":\n",
    "            sr = 24000\n",
    "        else:\n",
    "            sr = 6000\n",
    "        if H8_file_path != 'No data':\n",
    "            with open(H8_file_path,'rb') as fp:\n",
    "                data = np.frombuffer(fp.read(),dtype='>u2').reshape(sr,sr)\n",
    "                data = self.DN2TBB(data)\n",
    "                data = data/100\n",
    "\n",
    "            print(\"data reading finish\")\n",
    "            return data\n",
    "        else:\n",
    "            return 'No data'\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "39e21544-a1f1-488c-9bac-496a1f46f138",
   "metadata": {},
   "outputs": [],
   "source": [
    "def Time_split(time):\n",
    "    YYYY = time.strftime('%Y')\n",
    "    MM = time.strftime('%m')\n",
    "    DD = time.strftime('%d')\n",
    "    HH = time.strftime('%H')\n",
    "    MIN = time.strftime('%M')\n",
    "    date = YYYY + MM + DD + HH + MIN\n",
    "    return YYYY,MM,DD,HH,MIN,date\n",
    "\n",
    "def time2mjd(dateT):\n",
    "    t0=dt.datetime(1858,11,17,0,0,0,0)\n",
    "    mjd=(dateT-t0).days\n",
    "    mjd_s=dateT.hour*3600.0+dateT.minute*60.0+dateT.second+dateT.microsecond/1000000.0\n",
    "    return mjd+mjd_s/86400.0\n",
    "\n",
    "def Get_Kernel_Input_Parameter(date,band):\n",
    "    AHI_SZA,AHI_SAA = AHI_angle(date).read_angle_data()\n",
    "    RAA = abs(AHI_SAA - VAA)\n",
    "    RAA[RAA>180]=360-RAA[RAA>180]\n",
    "    if band == 4:\n",
    "        if os.path.exists(SR_PATH + '{}_AC/{}_b04.dat'.format(date,date)):\n",
    "            with open(SR_PATH + '{}_AC/{}_b04.dat'.format(date,date),'rb') as fp:\n",
    "                SR = np.frombuffer(fp.read(),dtype='f4').reshape(12000,12000)\n",
    "        else:\n",
    "            return 'None','None','None'\n",
    "    if band == 3:\n",
    "        if os.path.exists(SR_PATH + '{}_AC/{}_b03.dat'.format(date,date)):\n",
    "            with open(SR_PATH + '{}_AC/{}_b03.dat'.format(date,date),'rb') as fp:\n",
    "                SR = np.frombuffer(fp.read(),dtype='f4').reshape(24000,24000)\n",
    "                SR = SR.reshape(12000,2,12000,2).mean(-1).mean(1)\n",
    "        else:\n",
    "            return 'None','None','None'\n",
    "    # return SR[AHI_row-6:AHI_row+6,AHI_col-6:AHI_col+6],RAA[AHI_row-6:AHI_row+6,AHI_col-6:AHI_col+6],AHI_SZA[AHI_row-6:AHI_row+6,AHI_col-6:AHI_col+6]\n",
    "    return SR[2550:2750,4700:5000],RAA[2550:2750,4700:5000],AHI_SZA[2550:2750,4700:5000]\n",
    "\n",
    "def Volume_Kernel(wavel_nm,latitude,longitude,mjd,Cos_SZA,Cos_VZA,Cos_RAA):\n",
    "    brdf = sk.RossThickKernel()\n",
    "    vol = brdf.reflectance(wavel_nm,latitude,longitude,mjd,Cos_SZA,Cos_VZA,Cos_RAA)\n",
    "    return vol\n",
    "\n",
    "def Geometric_Kernel(wavel_nm,latitude,longitude,mjd,Cos_SZA,Cos_VZA,Cos_RAA):\n",
    "    brdf = sk.LiSparseReciprocalKernel(crown_shape=1.0, relative_height=2.0)\n",
    "    geo = brdf.reflectance(wavel_nm,latitude,longitude,mjd,Cos_SZA,Cos_VZA,Cos_RAA)\n",
    "    return geo\n",
    "\n",
    "\n",
    "\n",
    "def MODIS_Kernel_BRDF(band,idx_row,idx_col): \n",
    "    if band == 3:\n",
    "        if c_o_mask[idx_row,idx_col] == True:\n",
    "            kvol = Volume_Kernel(640,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "            kgeo = Geometric_Kernel(640,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "            return kvol,kgeo\n",
    "        else:\n",
    "            return np.nan,np.nan\n",
    "    elif band == 4:\n",
    "        if c_o_mask[idx_row,idx_col] == True:\n",
    "            kvol = Volume_Kernel(860,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "            kgeo = Geometric_Kernel(860,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "            return kvol,kgeo\n",
    "        else:\n",
    "            return np.nan,np.nan\n",
    "\n",
    "def Mask_Merge(date):\n",
    "    # Read cloud mask\n",
    "    # cloud 0, clear 1\n",
    "    if os.path.exists('/data01/people/liwei/Data/cm_120-150E_20-50N/{}_cloudmask.dat'.format(date)):\n",
    "        with open('/data01/people/liwei/Data/cm_120-150E_20-50N/{}_cloudmask.dat'.format(date),'rb') as fp:\n",
    "            cloudmask = np.frombuffer(fp.read(),dtype='<f4').reshape(1500,1500)\n",
    "            cloudmask = cv2.resize(np.array(cloudmask,dtype='float32'),(3000,3000),interpolation=cv2.INTER_NEAREST)\n",
    "\n",
    "        MASK = (Ocean_mask + cloudmask)\n",
    "        MASK = np.where(MASK==2,1,0)\n",
    "        MASK = MASK.astype(bool)\n",
    "        return MASK[1550:1750,1200:1500]\n",
    "        # return MASK[919:931,2038:2050]\n",
    "    \n",
    "@nb.jit()\n",
    "def get_water_idx():\n",
    "    Om = Ocean_mask.reshape(-1)\n",
    "    Water_idx = []\n",
    "    for i in range(200*300):\n",
    "        if Om[i] == True:\n",
    "            Water_idx.append(i)\n",
    "    return Water_idx\n",
    "\n",
    "def H8_Process(ACCOUNT,PW,Band,Band_number,Date):\n",
    "    data = H8_data(ACCOUNT,PW,Band,Band_number,Date).read_H8data()\n",
    "    return data\n",
    "\n",
    "import numpy.ma as ma"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96cf6e7f-9233-4fed-b9e8-59451ce666e6",
   "metadata": {},
   "source": [
    "# MAIN FUNCTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "05e8241f-6930-4b6c-ace1-739ef7b35b3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "SR_PATH = '/data01/people/liwei/AC_Result/'\n",
    "VZA_PATH = '/data01/GEO/INPUT/ANGLE/Viewer_Zenith_Angle/AHI_VZA_10.dat'\n",
    "VAA_PATH = '/data01/GEO/INPUT/ANGLE/Viewer_Azimuth_Angle/AHI_VAA_10.dat'\n",
    "LAND_MASK_PATH = '/data01/GEO/INPUT/LAND_MASK/Landmask_10.dat'\n",
    "SZA_PATH = '/data01/GEO/INPUT/ANGLE/Solar_Zenith_Angle_u2/'\n",
    "SAA_PATH = '/data01/GEO/INPUT/ANGLE/Solar_Azimuth_Angle_u2/'\n",
    "\n",
    "# AHI VZA/VAA Load\n",
    "with open(VZA_PATH,'rb') as fp:\n",
    "    VZA = np.frombuffer(fp.read(),dtype='u2').reshape(12000,12000)[2550:2750,4700:5000] / 100\n",
    "with open(VAA_PATH,'rb') as fp:\n",
    "    VAA = np.frombuffer(fp.read(),dtype='u2').reshape(12000,12000) / 100\n",
    "with open(LAND_MASK_PATH,'rb') as fp:\n",
    "    Ocean_mask = np.frombuffer(fp.read(),dtype='u1').reshape(12000,12000)[1000:4000,3500:6500]\n",
    "res = 0.01\n",
    "lat = np.linspace(50 - res/2 ,20 + res/2,3000)\n",
    "lon = np.linspace(120 + res/2 ,150 - res/2,3000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "32c0c3aa-299b-4fcd-987e-712a3d33cd0e",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "SR_PATH = '/data01/people/liwei/AC_Result/'\n",
    "VZA_PATH = '/data01/GEO/INPUT/ANGLE/Viewer_Zenith_Angle/AHI_VZA_10.dat'\n",
    "VAA_PATH = '/data01/GEO/INPUT/ANGLE/Viewer_Azimuth_Angle/AHI_VAA_10.dat'\n",
    "LAND_MASK_PATH = '/data01/GEO/INPUT/LAND_MASK/Landmask_10.dat'\n",
    "SZA_PATH = '/data01/GEO/INPUT/ANGLE/Solar_Zenith_Angle_u2/'\n",
    "SAA_PATH = '/data01/GEO/INPUT/ANGLE/Solar_Azimuth_Angle_u2/'\n",
    "\n",
    "# AHI VZA/VAA Load\n",
    "with open(VZA_PATH,'rb') as fp:\n",
    "    VZA = np.frombuffer(fp.read(),dtype='u2').reshape(12000,12000)[AHI_row-6:AHI_row+6,AHI_col-6:AHI_col+6] / 100\n",
    "with open(VAA_PATH,'rb') as fp:\n",
    "    VAA = np.frombuffer(fp.read(),dtype='u2').reshape(12000,12000) / 100\n",
    "with open(LAND_MASK_PATH,'rb') as fp:\n",
    "    Ocean_mask = np.frombuffer(fp.read(),dtype='u1').reshape(12000,12000)[1000:4000,3500:6500]\n",
    "res = 0.01\n",
    "lat = np.linspace(50 - res/2 ,20 + res/2,3000)\n",
    "lon = np.linspace(120 + res/2 ,150 - res/2,3000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72ed4839-d6a4-4926-b901-a809f4999ea1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-21 00:00:00\n",
      "time: 13.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 00:10:00\n",
      "time: 11.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 00:20:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 00:30:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 00:40:00\n",
      "time: 12.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 00:50:00\n",
      "time: 12.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 01:00:00\n",
      "time: 18.1 secs, 0.3 mins,0.0 hours\n",
      "2018-07-21 01:10:00\n",
      "time: 10.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 01:20:00\n",
      "time: 11.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 01:30:00\n",
      "time: 11.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 01:40:00\n",
      "time: 10.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 01:50:00\n",
      "time: 17.2 secs, 0.3 mins,0.0 hours\n",
      "2018-07-21 02:00:00\n",
      "time: 10.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 02:10:00\n",
      "time: 11.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 02:20:00\n",
      "time: 13.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 02:30:00\n",
      "time: 10.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 02:40:00\n",
      "2018-07-21 02:50:00\n",
      "time: 14.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:00:00\n",
      "time: 12.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:10:00\n",
      "time: 16.8 secs, 0.3 mins,0.0 hours\n",
      "2018-07-21 03:20:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:30:00\n",
      "time: 13.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:40:00\n",
      "time: 19.8 secs, 0.3 mins,0.0 hours\n",
      "2018-07-21 03:50:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:00:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:10:00\n",
      "time: 14.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:20:00\n",
      "time: 12.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:30:00\n",
      "time: 12.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:40:00\n",
      "time: 16.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-21 04:50:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:00:00\n",
      "time: 10.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:10:00\n",
      "time: 11.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:20:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:30:00\n",
      "time: 13.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:40:00\n",
      "time: 12.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:50:00\n",
      "time: 11.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 06:00:00\n",
      "time: 16.3 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 00:00:00\n",
      "time: 16.6 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 00:10:00\n",
      "time: 14.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 00:20:00\n",
      "time: 11.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 00:30:00\n",
      "time: 13.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 00:40:00\n",
      "time: 15.4 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 00:50:00\n",
      "time: 14.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 01:00:00\n",
      "time: 16.3 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 01:10:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 01:20:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 01:30:00\n",
      "time: 15.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 01:40:00\n",
      "time: 13.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 01:50:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 02:00:00\n",
      "time: 13.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 02:10:00\n",
      "time: 16.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 02:20:00\n",
      "time: 13.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 02:30:00\n",
      "time: 11.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 02:40:00\n",
      "2018-07-22 02:50:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:00:00\n",
      "time: 10.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:10:00\n",
      "time: 12.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:20:00\n",
      "time: 13.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:30:00\n",
      "time: 10.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:40:00\n",
      "time: 13.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 03:50:00\n",
      "time: 16.2 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 04:00:00\n",
      "time: 10.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 04:10:00\n",
      "time: 11.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 04:20:00\n",
      "time: 11.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 04:30:00\n",
      "time: 13.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 04:40:00\n",
      "time: 17.6 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 04:50:00\n",
      "time: 13.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 05:00:00\n",
      "time: 13.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 05:10:00\n",
      "time: 17.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-22 05:20:00\n",
      "time: 13.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 05:30:00\n",
      "time: 13.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 05:40:00\n",
      "time: 21.6 secs, 0.4 mins,0.0 hours\n",
      "2018-07-22 05:50:00\n",
      "time: 12.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-22 06:00:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:00:00\n",
      "time: 12.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:10:00\n",
      "time: 12.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:20:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:30:00\n",
      "time: 11.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:40:00\n",
      "time: 11.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 00:50:00\n",
      "time: 16.5 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 01:00:00\n",
      "time: 15.6 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 01:10:00\n",
      "time: 11.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 01:20:00\n",
      "time: 11.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 01:30:00\n",
      "time: 14.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 01:40:00\n",
      "time: 11.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 01:50:00\n",
      "time: 10.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 02:00:00\n",
      "time: 12.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 02:10:00\n",
      "time: 11.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 02:20:00\n",
      "time: 13.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 02:30:00\n",
      "time: 13.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 02:40:00\n",
      "2018-07-23 02:50:00\n",
      "time: 10.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 03:00:00\n",
      "time: 11.4 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 03:10:00\n",
      "time: 14.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 03:20:00\n",
      "time: 12.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 03:30:00\n",
      "time: 17.4 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 03:40:00\n",
      "time: 14.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 03:50:00\n",
      "time: 10.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 04:00:00\n",
      "time: 14.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 04:10:00\n",
      "time: 20.2 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 04:20:00\n",
      "time: 16.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 04:30:00\n",
      "time: 13.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 04:40:00\n",
      "time: 14.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 04:50:00\n",
      "time: 18.2 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 05:00:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 05:10:00\n",
      "time: 16.1 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 05:20:00\n",
      "time: 12.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 05:30:00\n",
      "time: 20.3 secs, 0.3 mins,0.0 hours\n",
      "2018-07-23 05:40:00\n",
      "time: 12.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 05:50:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-23 06:00:00\n",
      "time: 12.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 00:00:00\n",
      "time: 12.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 00:10:00\n",
      "time: 11.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 00:20:00\n",
      "time: 16.0 secs, 0.3 mins,0.0 hours\n",
      "2018-07-24 00:30:00\n",
      "time: 11.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 00:40:00\n",
      "time: 10.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 00:50:00\n",
      "time: 17.9 secs, 0.3 mins,0.0 hours\n",
      "2018-07-24 01:00:00\n",
      "time: 10.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 01:10:00\n",
      "time: 20.6 secs, 0.3 mins,0.0 hours\n",
      "2018-07-24 01:20:00\n",
      "time: 10.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 01:30:00\n",
      "time: 13.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 01:40:00\n",
      "time: 12.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 01:50:00\n",
      "time: 13.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 02:00:00\n",
      "time: 11.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 02:10:00\n",
      "time: 9.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 02:20:00\n",
      "time: 14.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 02:30:00\n",
      "time: 10.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 02:40:00\n",
      "2018-07-24 02:50:00\n",
      "time: 10.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:00:00\n",
      "time: 12.8 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:10:00\n",
      "time: 13.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:20:00\n",
      "time: 12.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:30:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:40:00\n",
      "time: 11.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 03:50:00\n",
      "time: 13.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-24 04:00:00\n"
     ]
    }
   ],
   "source": [
    "kvol = []\n",
    "kgeo = []\n",
    "Ref = []\n",
    "time_list = []\n",
    "for d in range(21,28):\n",
    "    \n",
    "\n",
    "    date_start = '2018-07-{} 00:00'.format(str(d))\n",
    "    date_end = '2018-07-{} 06:00'.format(str(d))\n",
    "    date_t = dt.timedelta(minutes=10)\n",
    "    # date_t = dt.timedelta(minutes=30)\n",
    "    date_s = dt.datetime.strptime(date_start, \"%Y-%m-%d %H:%M\")\n",
    "    date_e = dt.datetime.strptime(date_end, \"%Y-%m-%d %H:%M\")\n",
    "    date_time_now = date_s\n",
    "    b = 3\n",
    "    \n",
    "    # time_list = []\n",
    "    while date_time_now <= date_e:\n",
    "        print(date_time_now)\n",
    "        start_time = T.time()\n",
    "        YYYY,MM,DD,HH,MIN,date= Time_split(date_time_now)\n",
    "        c_o_mask = Mask_Merge(date)\n",
    "        \n",
    "        SR,RAA,SZA = Get_Kernel_Input_Parameter(date,b)\n",
    "        if SR[0][0] == 'N':\n",
    "            date_time_now = date_time_now + date_t\n",
    "            continue\n",
    "        MJD = time2mjd(date_time_now)\n",
    "        Kresult = Parallel(n_jobs=20)(delayed(MODIS_Kernel_BRDF)(b,i,j) for i in range(200) for j in range(300))\n",
    "        \n",
    "        kvol_current = ma.array(np.array(Kresult)[:,0].reshape(200,300),mask = ~c_o_mask,fill_value = np.nan).filled()\n",
    "        kgeo_current = ma.array(np.array(Kresult)[:,1].reshape(200,300),mask = ~c_o_mask,fill_value = np.nan).filled()\n",
    "        Ref_current =  ma.array(SR,mask = ~c_o_mask,fill_value = np.nan).filled()\n",
    "        kvol.append(kvol_current)\n",
    "        kgeo.append(kgeo_current)\n",
    "        Ref.append(Ref_current)\n",
    "        time_list.append(date)\n",
    "\n",
    "\n",
    "        end_time = T.time()\n",
    "        TIME = end_time - start_time\n",
    "        print('time: {:.1f} secs, {:.1f} mins,{:.1f} hours'.format(TIME,TIME/60,TIME/3600))\n",
    "        date_time_now = date_time_now + date_t"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d21fbe6d-34cd-4e20-8a83-754111e280d9",
   "metadata": {},
   "source": [
    "# Fitting Least squares / Multiple linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6153b4a2-c60d-414a-9f98-03bdde9d5f48",
   "metadata": {},
   "outputs": [],
   "source": [
    "def fitting(i,j):\n",
    "    r = np.array(Ref)[:,i,j].reshape(-1,1)\n",
    "    cm = ~np.isnan(r)\n",
    "    r = r[cm].reshape(-1,1)\n",
    "    k = np.hstack((np.array(kvol)[:,i,j].reshape(-1,1)[cm].reshape(-1,1),np.array(kgeo)[:,i,j].reshape(-1,1)[cm].reshape(-1,1)))\n",
    "    if r.size > 54 and not np.isnan(r[0]):\n",
    "        linreg = LinearRegression()\n",
    "        model=linreg.fit(k, r)\n",
    "        fi = linreg.intercept_[0]\n",
    "        fv, fg = linreg.coef_[0]\n",
    "        return fi,fv,fg\n",
    "    else:\n",
    "        return np.nan,np.nan,np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9ebac671-7145-420c-a204-bc72ca53cef1",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "result = Parallel(n_jobs=25)(delayed(fitting)(i,j) for i in range(200) for j in range(300))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4df82b3a-d272-4634-a184-f812dd829592",
   "metadata": {},
   "source": [
    "# Fitting Result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "de776079-9365-4258-96af-e50ffc6bb535",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[       nan        nan        nan ... 0.06533023        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " ...\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]]\n",
      "[[        nan         nan         nan ... -0.34955976         nan\n",
      "          nan]\n",
      " [        nan         nan         nan ...         nan         nan\n",
      "          nan]\n",
      " [        nan         nan         nan ...         nan         nan\n",
      "          nan]\n",
      " ...\n",
      " [        nan         nan         nan ...         nan         nan\n",
      "          nan]\n",
      " [        nan         nan         nan ...         nan         nan\n",
      "          nan]\n",
      " [        nan         nan         nan ...         nan         nan\n",
      "          nan]]\n",
      "[[       nan        nan        nan ... 0.14082167        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " ...\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]\n",
      " [       nan        nan        nan ...        nan        nan        nan]]\n"
     ]
    }
   ],
   "source": [
    "fiso = np.array(result)[:,0].reshape(200,300)\n",
    "fvol = np.array(result)[:,1].reshape(200,300)\n",
    "fgeo = np.array(result)[:,2].reshape(200,300)\n",
    "print(fiso)\n",
    "print(fvol)\n",
    "print(fgeo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "531ae2a8-24c9-42ed-bbf3-d992deeb1614",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-21 00:00:00\n",
      "time: 22.8 secs, 0.4 mins,0.0 hours\n",
      "2018-07-21 00:10:00\n",
      "time: 6.3 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 00:20:00\n",
      "time: 6.6 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 00:30:00\n",
      "time: 7.3 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 00:40:00\n",
      "time: 6.9 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 00:50:00\n",
      "time: 6.9 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:00:00\n",
      "time: 7.3 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:10:00\n",
      "time: 7.3 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:20:00\n",
      "time: 6.6 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:30:00\n",
      "time: 7.2 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:40:00\n",
      "time: 7.4 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 01:50:00\n",
      "time: 6.0 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 02:00:00\n",
      "time: 8.2 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 02:10:00\n",
      "time: 7.6 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 02:20:00\n",
      "time: 7.9 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 02:30:00\n",
      "time: 7.1 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 02:40:00\n",
      "2018-07-21 02:50:00\n",
      "time: 6.8 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 03:00:00\n",
      "time: 10.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:10:00\n",
      "time: 9.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:20:00\n",
      "time: 8.9 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 03:30:00\n",
      "time: 8.7 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 03:40:00\n",
      "time: 9.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 03:50:00\n",
      "time: 9.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:00:00\n",
      "time: 9.0 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 04:10:00\n",
      "time: 9.7 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:20:00\n",
      "time: 9.1 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:30:00\n",
      "time: 8.3 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 04:40:00\n",
      "time: 9.5 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 04:50:00\n",
      "time: 9.0 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:00:00\n",
      "time: 9.3 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:10:00\n",
      "time: 9.9 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:20:00\n",
      "time: 9.6 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:30:00\n",
      "time: 8.4 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 05:40:00\n",
      "time: 9.2 secs, 0.2 mins,0.0 hours\n",
      "2018-07-21 05:50:00\n",
      "time: 8.9 secs, 0.1 mins,0.0 hours\n",
      "2018-07-21 06:00:00\n",
      "time: 8.7 secs, 0.1 mins,0.0 hours\n"
     ]
    }
   ],
   "source": [
    "def Kernel_value(band,idx_row,idx_col): \n",
    "    if band == 3:\n",
    "        kvol = Volume_Kernel(640,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "        kgeo = Geometric_Kernel(640,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "        return kvol,kgeo\n",
    "    elif band == 4:\n",
    "        kvol = Volume_Kernel(860,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "        kgeo = Geometric_Kernel(860,lat[idx_row],lon[idx_col],MJD,np.cos(np.radians(SZA[idx_row,idx_col])),np.cos(np.radians(VZA[idx_row,idx_col])),np.cos(np.radians(RAA[idx_row,idx_col])))\n",
    "        return kvol,kgeo\n",
    "    \n",
    "    \n",
    "# kvol_valid = []\n",
    "# kgeo_valid = []\n",
    "# Ref_valid = []\n",
    "SZA_valid = []\n",
    "for d in range(21,22):\n",
    "    \n",
    "\n",
    "    date_start = '2018-07-{} 00:00'.format(str(d))\n",
    "    date_end = '2018-07-{} 06:00'.format(str(d))\n",
    "    date_t = dt.timedelta(minutes=10)\n",
    "    # date_t = dt.timedelta(minutes=30)\n",
    "    date_s = dt.datetime.strptime(date_start, \"%Y-%m-%d %H:%M\")\n",
    "    date_e = dt.datetime.strptime(date_end, \"%Y-%m-%d %H:%M\")\n",
    "    date_time_now = date_s\n",
    "    b = 3\n",
    "    # time_list = []\n",
    "    while date_time_now <= date_e:\n",
    "        print(date_time_now)\n",
    "        start_time = T.time()\n",
    "        YYYY,MM,DD,HH,MIN,date= Time_split(date_time_now)\n",
    "        c_o_mask = Mask_Merge(date)\n",
    "    \n",
    "        SR,RAA,SZA = Get_Kernel_Input_Parameter(date,b)\n",
    "        SZA_valid.append(SZA)\n",
    "        if SR[0][0] == 'N':\n",
    "            date_time_now = date_time_now + date_t\n",
    "            continue\n",
    "        # MJD = time2mjd(date_time_now)\n",
    "        # Kresult = Parallel(n_jobs=30)(delayed(Kernel_value)(b,i,j) for i in range(200) for j in range(300))\n",
    "        \n",
    "        # kvol_current = np.array(Kresult)[:,0].reshape(200,300)\n",
    "        # kgeo_current = np.array(Kresult)[:,1].reshape(200,300)\n",
    "        # Ref_current =  SR\n",
    "        # kvol_valid.append(kvol_current)\n",
    "        # kgeo_valid.append(kgeo_current)\n",
    "        # Ref_valid.append(Ref_current)\n",
    "        end_time = T.time()\n",
    "        TIME = end_time - start_time\n",
    "        print('time: {:.1f} secs, {:.1f} mins,{:.1f} hours'.format(TIME,TIME/60,TIME/3600))\n",
    "        date_time_now = date_time_now + date_t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ce32b4c9-6dbd-4d25-ae4a-6bd841dac9b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check Landcover"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 423,
   "id": "c8ac5bdd-a4e9-4927-84da-2911d8131d06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 5,  5,  5,  5,  5,  1,  1,  5,  5,  1],\n",
       "       [ 5,  5,  5,  5,  5,  1,  1,  1,  1,  1],\n",
       "       [ 5,  5,  1,  1,  1,  1,  1,  1,  1,  2],\n",
       "       [ 5,  5,  5,  1,  1,  5,  1,  1,  2,  2],\n",
       "       [ 1,  1,  1,  1,  1,  1,  2,  2,  5,  5],\n",
       "       [ 1,  1,  5,  5,  2,  2,  5,  5,  2,  2],\n",
       "       [ 1,  1,  1,  1,  2,  5,  5,  5,  5,  5],\n",
       "       [ 1,  1,  1,  2,  5,  2,  5,  5,  1,  1],\n",
       "       [ 1,  2,  2,  5,  1,  5,  5,  5,  1,  1],\n",
       "       [ 2, 10,  5,  2,  1,  1,  1,  2,  5,  5]], dtype=uint8)"
      ]
     },
     "execution_count": 423,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# with open( '/data01/people/liwei/Data/Landcover/MCD12Q1.006.LC.CEReS_H8AHI.01km.MEAN.NA.ann.bsq.byt','rb') as fp:      \n",
    "#     Landcover = np.frombuffer(fp.read(),dtype='uint8').reshape(12000, 12000)\n",
    "Landcover[2550:2750,4700:5000][i_idx-r:i_idx+r,j_idx-r:j_idx+r]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 422,
   "id": "8ed44fac-e44f-47b5-afca-9691db52ca73",
   "metadata": {},
   "outputs": [],
   "source": [
    "i_idx,j_idx = 79,185\n",
    "r = 5\n",
    "X = np.array(Ref_valid)[0:36,i_idx-r:i_idx+r,j_idx-r:j_idx+r]\n",
    "Y = fiso[i_idx-r:i_idx+r,j_idx-r:j_idx+r] + np.array(kvol_valid)[0:36,i_idx-r:i_idx+r,j_idx-r:j_idx+r]*fvol[i_idx-r:i_idx+r,j_idx-r:j_idx+r] + np.array(kgeo_valid)[0:36,i_idx-r:i_idx+r,j_idx-r:j_idx+r]*fgeo[i_idx-r:i_idx+r,j_idx-r:j_idx+r]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "c83b5635-28e9-431c-8f08-26086a23b973",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.colors as mcolors    \n",
    "color_list = list(mcolors.CSS4_COLORS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 425,
   "id": "30f40059-318c-40cc-bf27-820a1cbd1d93",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.6)"
      ]
     },
     "execution_count": 425,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "axis_min = 0\n",
    "axis_max = 0.6\n",
    "\n",
    "fig = plt.figure(figsize=(16,4))\n",
    "ax1 = fig.add_subplot(111)\n",
    "\n",
    "\n",
    "ax1.tick_params(axis=\"y\",which='major',length=10,direction='in',labelsize=8)\n",
    "ax1.tick_params(axis=\"x\",which='major',length=10,direction='in',labelsize=8)\n",
    "\n",
    "\n",
    "ax1.spines['right'].set_color('none')\n",
    "ax1.spines['top'].set_color('none')\n",
    "\n",
    "# ax1.plot(label,X,color = 'k',marker=\"o\",markersize=1,linewidth=1,label='Ref')\n",
    "for i in range(2*r):\n",
    "    for j in range(2*r):\n",
    "        ax1.plot(label,Y[:,i,j],color = 'r',marker=\"o\",markersize=1,linewidth=1,label='{},{}'.format(i,j))\n",
    "\n",
    "\n",
    "# ax1.set_xticks(np.linspace(0,105,106))\n",
    "ax1.set_yticks(np.arange(axis_min+0.1,1.1,0.1))\n",
    "\n",
    "ax1.set_xlabel(\"Local Obs Time \",fontsize=10)\n",
    "ax1.set_ylabel(\"BRDF fitting Ref\",fontsize=10)\n",
    "\n",
    "\n",
    "\n",
    "text_x = axis_min + (axis_max - axis_min) * 0.07\n",
    "text_y = axis_max - (axis_max - axis_min) * 0.22\n",
    "\n",
    "ax1.set_xlim(axis_min,36)\n",
    "ax1.set_ylim(axis_min,axis_max)\n",
    "# plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 382,
   "id": "29b20009-7544-4d54-9c4b-48aec1ff3aca",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import r2_score,mean_squared_error\n",
    "def add_right_cax(ax, pad, width):\n",
    "\n",
    "    axpos = ax.get_position()\n",
    "    caxpos = mtransforms.Bbox.from_extents(\n",
    "        axpos.x1 + pad,\n",
    "        axpos.y0,\n",
    "        axpos.x1 + pad + width,\n",
    "        axpos.y1\n",
    "    )\n",
    "    cax = ax.figure.add_axes(caxpos)\n",
    "\n",
    "    return cax\n",
    "def identifer(li):\n",
    "    result = []\n",
    "    for a in li:\n",
    "        mean = np.nanmean(a)\n",
    "        std = np.nanstd(a) \n",
    "        down = mean - 3 * std\n",
    "        up =  mean + 3 * std\n",
    "        n_a = np.where(a < down,np.nan,a)\n",
    "        n_a = np.where(n_a > up,np.nan,n_a)\n",
    "        result.append(n_a)\n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b062b918-8222-424b-b135-40ecaf049572",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.array(Ref).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a20ca56-8f7a-4dec-bce4-6e185166f28f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "DAY = 2\n",
    "\n",
    "X = np.array(Ref)[0:35,:,:]\n",
    "Y = fiso[:,:] + np.array(kvol)[0:35,:,:]*fvol[:,:] + np.array(kgeo)[0:35,:,:]*fgeo[:,:]\n",
    "\n",
    "nan_mask = ~np.isnan(X.reshape(-1))\n",
    "X = X.reshape(-1)[nan_mask]\n",
    "Y = Y.reshape(-1)[nan_mask]\n",
    "\n",
    "nan_mask = ~np.isnan(Y.reshape(-1))\n",
    "X = X.reshape(-1)[nan_mask]\n",
    "Y = Y.reshape(-1)[nan_mask]\n",
    "\n",
    "\n",
    "# rmse = np.sqrt(mean_squared_error(X ,Y))\n",
    "p = np.polyfit(X ,Y,1)    \n",
    "\n",
    "k = round(p[0],2)\n",
    "b = round(p[1],2)\n",
    "rmse = round(rmse.astype('float64'),3)\n",
    "\n",
    "axis_min = 0\n",
    "axis_max = 1\n",
    "\n",
    "fig = plt.figure(figsize=(4,4))\n",
    "ax1 = fig.add_subplot(111,aspect='equal')\n",
    "\n",
    "x = np.arange(axis_min,axis_max+1)\n",
    "y = 1 * x\n",
    "\n",
    "xx = np.arange(0,1+0.1,0.05) \n",
    "yy = k * xx + b\n",
    "\n",
    "# Calculate the point density\n",
    "xy = np.vstack([X,Y])\n",
    "z = gaussian_kde(xy)(xy)\n",
    "\n",
    "# Sort the points by density, so that the densest points are plotted last\n",
    "idx = z.argsort()\n",
    "X, Y, z = X[idx], Y[idx], z[idx]\n",
    "\n",
    "ax1.minorticks_on()\n",
    "# x_major_locator = plt.MultipleLocator(5)\n",
    "x_minor_locator = plt.MultipleLocator(0.05)\n",
    "ax1.xaxis.set_minor_locator(x_minor_locator)\n",
    "# ax.xaxis.set_major_locator(x_major_locator)\n",
    "ax1.yaxis.set_minor_locator(x_minor_locator)\n",
    "# ax.yaxis.set_major_locator(x_major_locator)\n",
    "\n",
    "ax1.tick_params(axis=\"y\",which='minor',length=5,direction='in',labelsize=8)\n",
    "ax1.tick_params(axis=\"y\",which='major',length=10,direction='in',labelsize=8)\n",
    "\n",
    "ax1.tick_params(axis=\"x\",which='minor',length=5,direction='in',labelsize=8)\n",
    "ax1.tick_params(axis=\"x\",which='major',length=10,direction='in',labelsize=8)\n",
    "\n",
    "\n",
    "ax1.spines['right'].set_color('none')\n",
    "ax1.spines['top'].set_color('none')\n",
    "\n",
    "\n",
    "im = ax1.scatter(X,Y,marker='o', c=z,s=15,cmap='Spectral_r')\n",
    "# im = ax1.scatter(X,Y,marker='o', c='k',s=15)\n",
    "\n",
    "ax1.set_xticks(np.arange(axis_min,axis_max+0.1,0.1))\n",
    "ax1.set_yticks(np.arange(axis_min+0.1,axis_max+0.1,0.1))\n",
    "\n",
    "ax1.set_xlabel(\"AHI Ref \",fontsize=10)\n",
    "ax1.set_ylabel(\"BRDF fitting Ref\",fontsize=10)\n",
    "\n",
    "\n",
    "ax1.plot(x,y,color='k',linewidth=2,linestyle='-',alpha=0.6)\n",
    "ax1.plot(xx,yy,color='r',linewidth=2,linestyle='-') \n",
    "\n",
    "\n",
    "text_x = axis_min + (axis_max - axis_min) * 0.07\n",
    "text_y = axis_max - (axis_max - axis_min) * 0.22\n",
    "\n",
    "ax1.text(text_x,text_y,s = 'N = {}\\nRMSE = {}\\ny = {} + {}x'.format(len(X),rmse,b,k),fontsize=8)\n",
    "\n",
    "cax = add_right_cax(ax1, pad=0.06, width=0.03)\n",
    "cb = fig.colorbar(im,cax=cax)\n",
    "cb.ax.set_xlabel('Count',rotation=360)\n",
    "ax1.set_xlim(axis_min,axis_max)\n",
    "ax1.set_ylim(axis_min,axis_max)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "362edb0b-a957-4ea5-904c-95b0b8fb3918",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 612,
   "id": "012a318c-eeef-4fa5-ac2c-306d6ae91bda",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 624,
   "id": "0ec865bb-5ec7-4357-8ae0-3123f6f1ed5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "FISO = fiso.reshape(-1)[~np.isnan(fiso.reshape(-1))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 625,
   "id": "16ef876a-191a-4269-9885-74d03398f6f1",
   "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>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>21040.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.380917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>32.345422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1588.200322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.096191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.193482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.594255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1221.407705</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0\n",
       "count  21040.000000\n",
       "mean       0.380917\n",
       "std       32.345422\n",
       "min    -1588.200322\n",
       "25%       -0.096191\n",
       "50%        0.193482\n",
       "75%        0.594255\n",
       "max     1221.407705"
      ]
     },
     "execution_count": 625,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(FISO)\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "10fe50e2-cbda-4073-a503-cbb0f568b708",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAF6CAYAAAAJaaMjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAAZhUlEQVR4nO3de2xk53nf8e/DmeGdHF7WRVOJtpRGRiPYLeISUpwUsNy4tRQEFoo6hoTWqVvbiwBR0yZpEhl1ZcUtnDhu0Dap6rSo3fSCxHGNNli4chTAlzpxbEMUFF8kW8FGSSrZCWxehsvldYb79A8Ox9zxLknRJMd89/sBBPKc8+5zHs7Z+ensyznnRGYiSTr7+nrdgCTpeBjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNBVrIj48Yh4OiLWIyIj4p+2v36i171JJ6Ha6wakkxAR9wH/DngS+LfAJvCZXvYknTQDXaX6od2vmfnV3ZUR8d3AWm9akk6Wga5S/SWAvWHeXv5yb9qRTp5z6CpKRDwcEQm8ur2cu//tWf5E158Zi4h/ERFfjIhLEbESEX8UEb8ZEX/9Gvt4Q0R8MiKW2/PzX4iIt0XEwGn8jNL1eIau0nyi/fVNwEuAn9tvcEQE8NvA9wGfBv4z0AJuZud/Cr8LPLFn/LuAtwHzwK8Dl4F7gHcBr42Iv52ZW8f200gvQHhzLpWofRb+qsyMrvUJ/N/MvKu9/HLg88BvZebf6RrbB9Qzc6m9/Erg94HngDsy88/b66vA/2Zn3v6fZ+a7TvBHk67LKRdpx3r3isy8shvmbf+o/fVf7YZ5e1wL+CngCvCWE+1S2oeBrhvd08AfAPdHxKci4mci4vsiov8aY1/R/vqx7g2Z+YfA88CtEVE/sW6lfRjouqFl5jbwN9n5rPqLgXcDnwLmI+JXImJ0z/DdoP6z65TbXT9x/J1KBzPQdcPLzKXM/InMnAFuY2fa5MvAA8B79wxdbn/9i9cp9R1d46RTZaBLe2Tmxcx8H/Aqdj7Bcu+ezU+2v97V/eci4rvY+WTMH2dm44TblK7JQNcNLSJujYjvvMamSWCAq39Z+v7217dHxIv21KgA/5qd99P7TqpX6SB+Dl03ur8G/K+IeBz4EvBV4EXsnJnX2JlTByAzfz8ifhH4GeCLEfEhYJWdz6G/DPg94D2n2770DQa6bnRzwC+wM8VyNztn5l9n52KiX87Mj+wdnJk/GxFPsjO//iPshP4fAW8HfsmLitRLXlgkSYVwDl2SCmGgS1IhDHRJKoSBLkmFMNAlqRA9+9jiuXPn8pZbbunV7iXpTHriiSfmM/NF19rWs0C/5ZZbmJub69Xupeu6dOkS4+PjvW5DuqaI+NPrbXPKReqyteW1QTqbDHRJKoSBLnWp130+hc4mA13q0mq1et2CdCQGutRldXW11y1IR2KgS1IhDHSpy/DwcK9bkI7EQJe61Gq1XrcgHYmBLnVZXvYZzzqbDHRJKoSBLnVxykVnlYEudfHCIp1VPiRaxYuIU9mPz+dVrxnoKt4LDdqIMJx1JjnlIkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklSIAwM9It4fEV+LiC9eZ3tExC9HxMWI+HxEvOL425QkHeQwZ+i/Bty9z/Z7gNva/50H3vuttyVJeqEODPTM/CSwuM+Qe4H/ljs+A0xExHccV4OSpMOpHkONm4Dn9iw/3173Z90DI+I8O2fxzMzMMD8/D8DIyAjVapXl5WUA+vv7GRsbY2FhAYC+vj6mpqZYXl6m2WwCMDExwebmJuvr650alUqFS5cuATAwMMDIyAiLi4tX1Wg0GrRaLQAmJydZX19nY2MDgNHRUSKClZUVAAYHBxkeHu7UqFQqTE5OsrS0xPb2NgBTU1Osra11aoyNjZGZXL58uVNjaGiIpaUlAKrVKhMTEywuLnLlypVOjdXVVTY3NwEYHx9ne3ub1dVVAIaGhhgYGKDRaABQq9Wo1+tX1ZienmZlZYWtrS0A6vU6rVarU2N4eJhardZ5jXdrLCwskJlEBNPT01e9xvV6nWazydra2g13nBqNhsfpDBynG/H9tJ/IzH0HAETELcCHM/Nl19j2YeAXMvP32ssfBX42M+f2qzk7O5tzc/sOkXoiIjjM+0LqhYh4IjNnr7XtOD7l8hVgZs/yze11kqRTdByBfgH4kfanXb4XWM7Mb5pukSSdrAPn0CPiN4C7gHMR8TzwDqAGkJm/CjwK/CBwEVgD/uFJNStJur4DAz0z7z9gewI/dmwdSZKOxCtFJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIcKtAj4u6IeCYiLkbEg9fY/uKI+HhEPBkRn4+IHzz+ViVJ+zkw0COiAjwC3APcDtwfEbd3DXs78MHM/B7gPuA/HHejkqT9HeYM/Q7gYmY+m5lbwAeAe7vGJDDe/r4OfPX4WpQkHUb1EGNuAp7bs/w8cGfXmIeB34mIfwyMAK+5VqGIOA+cB5iZmWF+fh6AkZERqtUqy8vLAPT39zM2NsbCwgIAfX19TE1Nsby8TLPZBGBiYoLNzU3W19c7NSqVCpcuXQJgYGCAkZERFhcXr6rRaDRotVoATE5Osr6+zsbGBgCjo6NEBCsrKwAMDg4yPDzcqVGpVJicnGRpaYnt7W0ApqamWFtb69QYGxsjM7l8+XKnxtDQEEtLSzsveLXKxMQEi4uLXLlypVNjdXWVzc1NAMbHx9ne3mZ1dRWAoaEhBgYGaDQaANRqNer1+lU1pqenWVlZYWtrC4B6vU6r1erUGB4eplardV7j3RoLCwtkJhHB9PT0Va9xvV6n2WyytrZ2wx2nRqPhcToDx+lGfD/tJzJz/wERrwfuzsy3tJffCNyZmQ/sGfOT7Vq/FBGvBN4HvCwzr1yv7uzsbM7Nze27b6kXIoKD3hdSr0TEE5k5e61th5ly+Qows2f55va6vd4MfBAgMz8NDALnXnirkqSjOkygPw7cFhG3RkQ/O7/0vNA15v8BPwAQEd/NTqB//TgblSTt78BAz8wW8ADwGPAldj7N8lREvDMiXtce9lPAWyPic8BvAG9K/80qSafqML8UJTMfBR7tWvfQnu+fBr7/eFuTJL0QXikqSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQlR73YD0QkxNTbG0tHTi+4mIE9/H5OQki4uLJ74f3TgMdJ0pS0tLZOaJ7qPZbFKr1U50H3A6/9PQjcUpF6lLs9nsdQvSkRjoUpe1tbVetyAdiYEuSYUw0KUuIyMjvW5BOhIDXepSrfpZAZ1NBrrUZXl5udctSEdioEtSIQx0qUt/f3+vW5COxECXuoyNjfW6BelIDHSpy8LCQq9bkI7EQJekQhjoUpe+Pt8WOpv8myt1mZqa6nUL0pEcKtAj4u6IeCYiLkbEg9cZ84aIeDoinoqIXz/eNqXT4+fQdVYdeElcRFSAR4C/BTwPPB4RFzLz6T1jbgPeBnx/Zi5FxF84qYalk+bdFnVWHeYM/Q7gYmY+m5lbwAeAe7vGvBV4JDOXADLza8fbpiTpIIe5acVNwHN7lp8H7uwa81KAiPgUUAEezszf7i4UEeeB8wAzMzPMz88DOzdDqlarnX/q9vf3MzY21vn4WF9fH1NTUywvL3fOniYmJtjc3GR9fb1To1KpcOnSJQAGBgYYGRnpPBFmt0aj0aDVagE7T4xZX19nY2MDgNHRUSKClZUVAAYHBxkeHu7UqFQqTE5OsrS0xPb2NrAz37q2ttapMTY2RmZy+fLlTo2hoaHOU3aq1SoTExMsLi5y5cqVTo3V1VU2NzcBGB8fZ3t7m9XVVQCGhoYYGBig0WgAUKvVqNfrV9WYnp5mZWWFra0tAOr1Oq1Wq1NjeHiYWq3WeY13aywsLJCZRATT09NXvcb1ep1ms9m5ney3y3Ha2Ng40eO0vb1No9E4lePUbDaLPU6+n07mOO0nDnr6S0S8Hrg7M9/SXn4jcGdmPrBnzIeBJvAG4Gbgk8DLM7Nxvbqzs7M5Nze3776lbhFx4k8sWl1dPZU7Lp7Gz6LyRMQTmTl7rW2HmXL5CjCzZ/nm9rq9ngcuZGYzM/8Y+EPgtqM0K/Xa7lmqdNYcJtAfB26LiFsjoh+4D7jQNea3gLsAIuIcO1Mwzx5fm5KkgxwY6JnZAh4AHgO+BHwwM5+KiHdGxOvawx4DFiLiaeDjwE9nptdP60zyARc6qw6cQz8pzqHrKE5j3nlra+tU7rjoHLqO4ludQ5duKLuf7JDOGgNdkgphoEtdBgYGet2CdCQGutTFX4rqrDLQpS67VzJKZ42BLkmFMNClLj7gQmeVf3OlLj7gQmeVgS512b0Ln3TWGOhSl93bwUpnjYEuSYUw0KUuk5OTvW5BOhIDXeri/dB1Vh3mEXTSt418xzg8XD/RfYyeaPVvyHeMn9KedKMw0HWmxM9dOvFbzs7Pz3Pu3LkT3Qe0b5/78InvRjcQp1ykLqOjp3WOLh0vA13qEhG9bkE6EgNd6rKystLrFqQjMdAlqRAGutRlcHCw1y1IR2KgS12Gh4d73YJ0JAa61MUHXOisMtAlqRAGutSlUqn0ugXpSAx0qYs359JZZaBLXZaWlnrdgnQkBrrUZXt7u9ctSEdioEtSIQx0qYsPidZZZaBLXdbW1nrdgnQkBrrUZWNjo9ctSEdioEtSIQx0qcvY2FivW5COxECXupz0I+6kk2KgS10uX77c6xakIzHQJakQBrrUxQdc6Kwy0KUuQ0NDvW5BOhIDXerizbl0VhnoklQIA13qUq1We92CdCQGutRlYmKi1y1IR2KgS118SLTOqkMFekTcHRHPRMTFiHhwn3F/NyIyImaPr0XpdF25cqXXLUhHcmCgR0QFeAS4B7gduD8ibr/GuDHgnwCfPe4mJUkHO8wZ+h3Axcx8NjO3gA8A915j3L8E3g1471GdaT7gQmfVYX6dfxPw3J7l54E79w6IiFcAM5n5fyLip69XKCLOA+cBZmZmmJ+fB2BkZIRqtcry8jIA/f39jI2NsbCwAEBfXx9TU1MsLy/TbDaBnV9cbW5usr6+3qlRqVS4dOkSAAMDA4yMjHTmQ3drNBoNWq0WsPN09/X19c79r0dHR4kIVlZWgJ0rBoeHhzs1KpUKk5OTLC0tdZ47OTU1xdraWqfG2NgYmdm5H8jg4CBDQ0OdzzZXq1UmJiZYXFzs/NN+amqK1dVVNjc3ARgfH2d7e5vV1VVg50KXgYEBGo0GALVajXq9flWN6elpVlZW2NraAqBer9NqtTo1hoeHqdVqndd4t8bCwgKZSUQwPT191Wtcr9dpNpudBz58uxynjY2NEz1Oa2trjI+Pn8pxajabxR4n308nc5z2EwfdWS4iXg/cnZlvaS+/EbgzMx9oL/cBHwPelJl/EhGfAP5ZZs7tV3d2djbn5vYdIn2TiDjxuyHOz89z7ty5E90HnM7PovJExBOZec3fUx5myuUrwMye5Zvb63aNAS8DPhERfwJ8L3DBX4xK0uk6TKA/DtwWEbdGRD9wH3Bhd2NmLmfmucy8JTNvAT4DvO6gM3Tp29X4+HivW5CO5MA59MxsRcQDwGNABXh/Zj4VEe8E5jLzwv4VpOMVEb1u4VhMTk72ugUV5lDXOGfmo8CjXeseus7Yu771tqRrO405Z+e2dVZ5pagkFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUiEMFekTcHRHPRMTFiHjwGtt/MiKejojPR8RHI+Ilx9+qJGk/BwZ6RFSAR4B7gNuB+yPi9q5hTwKzmflXgQ8Bv3jcjUqS9neYM/Q7gIuZ+WxmbgEfAO7dOyAzP56Za+3FzwA3H2+bkqSDHCbQbwKe27P8fHvd9bwZ+Mi30pQk6YWrHmexiPj7wCzwqutsPw+cB5iZmWF+fh6AkZERqtUqy8vLAPT39zM2NsbCwgIAfX19TE1Nsby8TLPZBGBiYoLNzU3W19c7NSqVCpcuXQJgYGCAkZERFhcXr6rRaDRotVoATE5Osr6+zsbGBgCjo6NEBCsrKwAMDg4yPDzcqVGpVJicnGRpaYnt7W0ApqamWFtb69QYGxsjM7l8+XKnxtDQEEtLSwBUq1UmJiZYXFzkypUrnRqrq6tsbm4CMD4+zvb2NqurqwAMDQ0xMDBAo9EAoFarUa/Xr6oxPT3NysoKW1tbANTrdVqtVqfG8PAwtVqt8xrv1lhYWCAziQimp6eveo3r9TrNZpO1tbUb7jg1Gg2P0xk4Tjfi+2k/kZn7D4h4JfBwZr62vfw2gMz8+a5xrwF+BXhVZn5t36LA7Oxszs3NHTRMOnURwUHvC6lXIuKJzJy91rbDTLk8DtwWEbdGRD9wH3ChawffA/xH4HWHCXNJ0vE7MNAzswU8ADwGfAn4YGY+FRHvjIjXtYe9BxgF/mdE/EFEXLhOOUnSCTnUHHpmPgo82rXuoT3fv+aY+5IkvUBeKSpJhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQV4lCBHhF3R8QzEXExIh68xvaBiPjN9vbPRsQtx96pJGlfBwZ6RFSAR4B7gNuB+yPi9q5hbwaWMvO7gH8DvPu4G5Uk7e8wZ+h3ABcz89nM3AI+ANzbNeZe4L+2v/8Q8AMREcfXpiTpIIcJ9JuA5/YsP99ed80xmdkCloHp42hQknQ41dPcWUScB84DzMzMMD8/D8DIyAjVapXl5WUA+vv7GRsbY2FhAYC+vj6mpqZYXl6m2WwCMDExwebmJuvr650alUqFS5cuATAwMMDIyAiLi4tX1Wg0GrRaLQAmJydZX19nY2MDgNHRUSKClZUVAAYHBxkeHu7UqFQqTE5OsrS0xPb2NgBTU1Osra11aoyNjZGZXL58uVNjaGiIpaUlAKrVKhMTEywuLnLlypVOjdXVVTY3NwEYHx9ne3ub1dVVAIaGhhgYGKDRaABQq9Wo1+tX1ZienmZlZYWtrS0A6vU6rVarU2N4eJhardZ5jXdrLCwskJlEBNPT01e9xvV6nWazydra2g13nBqNhsfpDBynG/H9tJ/IzP0HRLwSeDgzX9tefhtAZv78njGPtcd8OiKqwJ8DL8p9is/Ozubc3Ny++5Z6ISI46H0h9UpEPJGZs9fadpgpl8eB2yLi1ojoB+4DLnSNuQD8g/b3rwc+tl+YS5KO34FTLpnZiogHgMeACvD+zHwqIt4JzGXmBeB9wH+PiIvAIjuhL0k6RYeaQ8/MR4FHu9Y9tOf7DeCHj7c1SdIL4ZWiklQIA12SCmGgS1IhDHRJKoSBLkmFONUrRaVeOMpthY7yZ7z0Qr1moKt4LzRo5+fnOXfu3Al1I50cp1ykLrVardctSEdioEtd6vV6r1uQjsRAl7rs3g1QOmsMdKnL7i1UpbPGQJekQhjoUpfpaR+2pbPJQJe67D5hRzprDHSpy+5jx6SzxkCXpEIY6FIXP4eus8pAl7rsPsVeOmsMdKnL6upqr1uQjsRAl6RCRK9u+RkRXwf+tCc7l/Z3DpjvdRPSdbwkM190rQ09C3Tp21VEzGXmbK/7kF4op1wkqRAGuiQVwkCXvtl/6nUD0lE4hy5JhfAMXZIKYaDrhhQRPx4RX4qIpYh4sNf9SMfBKRfdkCLiy8BrMvP5XvciHRfP0HXDiYhfBb4T+EhE/ERE/Pv2+h+OiC9GxOci4pPtdYMR8V8i4gsR8WREvLqXvUv7qfa6Aem0ZeaPRsTdwKuBH9qz6SHgtZn5lYiYaK/7sZ0/ki+PiL8C/E5EvDQzN063a+lgnqFL3/Ap4Nci4q1Apb3ubwD/AyAzv8zO7Spe2pv2pP0Z6FJbZv4o8HZgBngiIny4qM4UA11qi4i/nJmfzcyHgK+zE+y/C/y99vaXAi8Gnuldl9L1OYcufcN7IuI2IICPAp8Dvgy8NyK+ALSAN2XmZg97lK7Ljy1KUiGccpGkQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQV4v8DS0xelqB3w3gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "FISO = fiso.reshape(-1)[~np.isnan(fiso.reshape(-1))]\n",
    "fig = plt.figure(figsize=(6,6))\n",
    "plt.boxplot(FISO,showfliers = False,labels=['fiso'])\n",
    "plt.title('fiso',fontsize = 20)\n",
    "plt.grid(linestyle=\"--\", alpha=0.3)\n",
    "# plt.ylabel('atm – cm',fontsize = 12)\n",
    "# plt.savefig('AOT550.jpg',dpi = 1000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "8a1c664d-227d-43af-b070-47ea255b21d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "FGEO = fgeo.reshape(-1)[~np.isnan(fgeo.reshape(-1))]\n",
    "fig = plt.figure(figsize=(6,6))\n",
    "plt.boxplot(FGEO,showfliers = False,labels=['fgeo'])\n",
    "plt.title('fgeo',fontsize = 20)\n",
    "plt.grid(linestyle=\"--\", alpha=0.3)\n",
    "# plt.ylabel('atm – cm',fontsize = 12)\n",
    "# plt.savefig('AOT550.jpg',dpi = 1000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "128db420-44be-4f8c-9802-3197caba6151",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "FVOL = fvol.reshape(-1)[~np.isnan(fvol.reshape(-1))]\n",
    "fig = plt.figure(figsize=(6,6))\n",
    "plt.boxplot(FVOL,showfliers = False,labels=['fvol'])\n",
    "plt.title('fvol',fontsize = 20)\n",
    "plt.grid(linestyle=\"--\", alpha=0.3)\n",
    "# plt.ylabel('atm – cm',fontsize = 12)\n",
    "# plt.savefig('AOT550.jpg',dpi = 1000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "caa33ff7-0096-4340-8699-5932599f17ed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f03e2814fa0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(fgeo,interpolation=None,cmap='gray')\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "3d9b488f-d466-4573-a251-53d27b6fd4b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f6cff8972b0>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(fiso,interpolation=None,cmap='gray')\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "797319b8-254b-4413-a148-16d5fb602847",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f6cff7d15e0>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAADpCAYAAADbLqdEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAB8gUlEQVR4nO29aZRc13Ue+p2a56qe0ABBUiQt2Yns2ErMyMmKkyfHjp/M5RfFXo4jea3IEkWAIgwQg0AChAkCHAySoAyKpkxSkETRdCLJfrZlafkptmU9v6e8lSgRqciTJpMURcw91jxXnfej6tu9q9AN9FDVXV0431q9urvq1r3n3rp3n32+/e29jbUWDg4ODg7DAc9GD8DBwcHBoXdwRt3BwcFhiOCMuoODg8MQwRl1BwcHhyGCM+oODg4OQwRn1B0cHByGCM6oOzg4OKwBxpgfMsZ8Q/1kjTH7urZ5hzEmo7Z5oF/j8fVrxw4ODg6Dine+8512ZmZmWdu+/PLLf2atfedS71trvwPgbQBgjPECOAfgc4ts+l+ttT+/8tGuDM6oOzg4XHOYmZnB1772tWVt6/F4xlew658G8Kq19vurGlgP4OgXBweHaxLW2mX9ABg3xrykfnZeYbfvBvCZJd7758aYvzLG/BdjzA/3/ITacJ66g4PDNYkVlEiZsdbeerWNjDEBAP8WwH2LvP11AG+y1uaNMbcB+GMAb1nuAFYC56k7ODhcc1iul77C2lg/B+Dr1tpLixwva63Nt//+IgC/MWYltM6y4Tx1BweHaxLNZrPXu3wPlqBejDFbAVyy1lpjzNvRcqhnez0AwBl1BweHaxS9rFBrjIkC+DcA7lSvfbB9nOcA/BKAu4wxdQAlAO+2fSqR64y6g4PDNYle2lRrbQHAWNdrz6m/Pwrgoz074BXgjLqDg8M1h1Xw5ZsGzqg7ODhck3BG3cHBwWGI4Iy6g4ODwxChD+qXgYAz6g4ODtccHKfu4ODgMGRwRt3BwcFhiOCMuoODg8MQwRl1BwcHhyGBtXZoA6V9K+hljHmnMeY7xphXjDGH+3UcBwcHh9WgDwW9BgJ9Mert7h+/jVbVsrcCeI8x5q39OJaDg4PDauCM+srwdgCvWGtfs9ZWAXwWwLv6dCwHBweHFWNYjXq/OPXtAM6o/88C+Am9Qbt7yE4A8Hq9Pz46OgpjjLzfbDbRaDQAAB6PR35ba2GMgTFGLni1WkU0GkWtVkOj0YAxBh6PRz7v87VO01orr3Ef3CfBv/X+9bgIr9fbMVZuu9S+eA7cvvscuD3H0z0Onrd+jf/z7+7P6f17PJ6O69c95u7teP25n8Wugx5DN7rPuXss3fvg+41GA9dff33Hvs6fP496vd7x2o033rjocZfCxYsXYYzB5OSkvHbu3LmO+6X7Yea5eTyejuvL76l7nFfCmTNn5HtfauxnzpzpuFe7jUqz2ZR9eL3eju9xpddjkHD+/Hk0Gg00m025N3hezWYTt9xyS8f2L7/88oy1dmItx9ysBns52LBAqbX2NIDTADA6Omp/8Rd/EcYYNBoNTE1Nwe/3Y2JiAmfOnMHk5CSKxSKi0Sh8Ph9isRgqlQoymQwSiQQqlQpGRkZgrcXZs2cRDAYRDocBAHNzc9i6dSs8Hg/K5TLS6bQ8sKlUCpFIBLVaDaVSSR4Sr9eLUCgEn88Hr9crN1ggEIDP55MAS61WQ7ValRskGAyiWCzKg5nNZgEAfr9fxlStVpHJZOD1ehGNRlGpVOD1euHz+ZDNZuH3++VYHo8H9Xod9XodtVoNsVgMHo8HxWIRjUYDPp8P4XAYtVoNtVoNxhj4fD7521qLWq2GQCAgx/B4PKjVavI9BAIBOQe/3y/718YDgOzb6/WiWq0CaBk7r9cLv9+PZrMphpfb8RpZaxEOh2UfjUYDkUgEwWAQ9XodgUAAH/rQhwAAR44cwYkTJ2R8R44cgd/vx+zsQunpRqOBZ599tif34e7du+Hz+RAIBHDy5MnL3j9w4AAajYYYdh7/qaeeWtFxjh07htnZWXz0oysv1Hfvvfcin8/jmWeeWfT9PXv24Omnn17xfgcBR48eRaVSgbUWuVwOjUZDnt1yuYzTp09f9hljTE/6fzqjvjKcA3CD+v/69muLwhiDSqUCYwxCoRAikQgajQbm5+c7DEG1WoUxBqVSCR6PB6VSSYxHOBxGvV4XQ1Gv1+VhpYcTCoUQi8XEiJXLZfHcvV4vAoEArLVi/Or1OorFImq1GkZGRhCJRAC0VgbVahXlchnValW293g8CAQCcnyg08vThpfbWGtRr9fFYJTLZfj9fvj9fgAQY0lv1uv1wuv1otFooF6vo9FooNFowForExCP2Ww24ff7O7xMfs7n88FaKwaaExXHpD1tj8cjhrpWq8l5er3eyzxabdw19KTI74TXnQYdQIdB5/Xofq3X4PkthlOnTgEAHn74YVm96FUaADz00EOYn5+H1+vFhz/84UX38+CDD656fCdPnsRdd9215PtLrZYGFQcOHOj4Xz87pVIJQOucFjPovcSwql/6ZdS/BuAtxpib0TLm7wbwK0ttTKMTDocRi8VQr9dx/vx5NJtNpFIp5PN5BINBlEolBAIBlMtlxGIxJJNJhEIhlMtllEolzM7OolKpiDELh8MYHx8XA0pjR8PKYwMt75ITAo0TDSENYKPREE9a0zh+vx8+n08MfblcFiPIyYjGwO/3IxQKidHkT7VaRb1eRygUEk+Fhprb1Ot1mfyCwaCMXXvLAMTwcjLUVAsnLkIbe9IKBFct/OHnNTXm8/lgjEG9Xpfj6BUGJzx+L5FIBNFoVM43Go0ueRMdPXpUJtJ+gWMpl8tX3O7o0aNLvvfAAw/g/vvvR7FY7PXwBLzeQGuC8fl8uO++VivM3/qt3+rbcfuBarUqDgHvKzojfr8fXq8XTz75ZN/HMayeuunXibWbq34EgBfA89ba31hq24mJCXvbbbd1UBK5XA6VSgWNRgM33ngjZmdnUSqVxJhFo1ExHjRgmUxGjGwgEEAwGBR6gTdSs9kU7zQYDIoBB4BEIoFqtQq/3y+Uh8fjEeokFAqh2Wwil8uhUCiIsQ2Hw6hUKvIZoHWT0pAFg0FeE4TDYZRKJVkxNJtNxGKxjhWIx+NBPp8XI0+Dy3HqCYnnCwCVSkUmCF4TGmztdZfL5Q7jzWvAB4rePL1rPYH4/X6hncLhMJLJJACgUCjI98VxcvLkhFetVuVaRiIR+P1+HDp0aK232pqxf/9+NBqNnhjHJ554Avfcc08PRjWc2Ldvn9wXpF14fxljUK1W8ZGPfOSK+zDGvGyX0Qj6SvjRH/1R+8UvfnFZ295www1rPt56om+cum01V13WVSMN0B0Y5evGGMTjcfEGeQM0m02Mj48Lx8xJgUbM5/OhWCyiWCzK8s7n84nR5r6AlieUz+fFuyWvTA4/kUigWCwil8sBgHiv5AJpyDjx0KNvNpuIx+MySemVAA0hVxXJZBL5fF48ZG4LtGiRarUqXDvBiYrnzEmano+mTEjVEHrlQm+bnjYDcQBkjJFIRPbv8/kwMjKCaDSKUCiERCKBmZkZpNNpGR+vEffL6zoyMoL7779/FXdVf9BLr7AfBn3fvn1XNXSbAYxP0HngvavvO31v9xvD6qkPTEYpud9IJCIc+MjIiAT2SqUSRkdHxWjQQ63X60gkEkin02LAQ6GQTAL1er0jSKiNJJd/5MEDgQDC4TDK5bLcdPT2AYhR9Hg84kF3G2oadIKTyMTEhJwHOXUaRBq+bDYr9EggEIDf70c+n0e5XEY0GpWHAFgw2qVSqcNTJ+3D8Wg1Czl5XlNOSuTsSePw++B4w+GwbMtzr9VqOHLkSMd3eOzYsQ5uX18zTnB+vx+pVKqv99KwYRgMOgBxvLQKi/cVY1PrSSU5o95njI6OolqtolgsIhgMwuv1IpVKoVwui5EnR1soFOD3+5FIJMTg0ogDLQ+RNIc2YqQVyP3SAJKjDgaDKJfLqNVqiEajCIfDEhCk4Q6FQh2BRHrBpC30KoHLyWKxiPHxcVHeULnSaDTEQDYaDcTjcQDA/Pw8gFbwl0aX4yW9oeVtNMTa29YTGR8czbnzc5wsSP9kMhmhjYwxiMVi2LJliwQ4q9Uq0un0okEm0j21Wk3+1sfl96QnPYdrAwcOHBBHifckAORyOXF8+hmTWAy9NOrGmNcB5AA0ANS76RrTWvY+BeA2AEUA77PWfr1nA1AYCKNOjzgYDApfnMvl8Nprr4l3nkwmxcjSQydPS96W71UqFTH2vFF8Ph/y+bx4nfV6XWgPoOXlUg3j9/s7uGFgwTOmsWNQlcaWKwJ63+l0GpVKRYxYpVKRFUU+n0cgEEA8HhdDHA6HEY/HRQbJiY18u/aUOXmFQqHLrqXm0anB5muMD5DDJF8OQIJuhw4dwvT0tMg6KaM7evRohwe+mFrk6NGj2LVrl9BH3FYb9lgsdkUlh8Nwg/cyZbh0cOgUrRe6nY0e4aestTNLvPdzAN7S/vkJAM+iK3enVxgIo67lhX6/XzTjxhicO3cOgUAAmUwGsVgM5XJZDLLmxAHIa9RS66Bod7KGluF5vV4JfHLbSCQiWnXuwxiDRCIhfwMQgx4KhcRD581CHtvr9co5VatVRCKRDmkifxcKBeRyuY7ECM09ao6cHrz2iDkR8Vy1tlyfLycZqnS0gX788ccX/Y4Yn6C6Zym9tdZSHzp0SFZHpJuupCK5Evbv34/JyUkcPjw8ZYSOHDmCdDqN8fFxPPTQQxs9nL7h4MGDQpnyXiZtCSw8J0vJSvuFdaZf3gXgRds66FeNMSljzDZr7YVeH2j9ohJXADlXyvlSqRQmJiaQSCTESJHb5hdBTtjr9SKXyyGbzcpNQi6YKg/SFjRgjUZDPs/gaTweRzgcRiKRQDweRyQSkRuRXj9vRgZ1PR4PYrGYGGlOIjT+4+PjMsFUKhUJ6NIoAwvKEwAiV6RXrblG8vdaSqknNa1j51i1Rh5YmDxzuRxqtRrq9TpmZ2fxxhtvXPU7KpfLyGQyVzTo3Xj88cexZcsWXH/99bjuuuvWxA0/+eSTq16eHzx48DJt9FrQq4mF55NOp3uyv0HEvffeC2AhAY9OUD6fF6UZ7/XF8hv6ie4M4qV+AIwbY15SPzsX2x2APzfGvLzE+4tl2W/v/VkNiKeuOWjN/9IoUapHjTdVFOTMKeHT1Am9QxplBh9pxLUmvDvgRzz11FPC5etJhZMFvWZSQQDkPQZfaeQBiIRLZ6bSSHNC0ElQfG8xdYAOAne/xwlLq4R0khG5/EajId761fDoo4+u6rvlQ90LrNabrdVqK84AXQyUPtbrddx+++3wer0YHx9f0bU5ceIEzp8/L/LYer2+ZKboZseBAwdkVUwHpNlsyjPC+zcQCKBWq8kqeL2wAk99ZhmSxp+01p4zxmwB8CVjzLettV9Z2whXh4Ex6sFgELVaDYVCQaiYUCiEZDIpRr1WqyGVSomxpNSOQT9OBt0yOs64oVBIaJJoNIpkMnlFD27v3r1XHPfJkydFcUIvo1wuIxgMikfO89NKG+rjAcjqoTtbkeUD9OSmf3OfvAZMwmK6P6+HDgx313NJJBKYnJzEsWPHljzHQ4cOLUnJLIX9+/evS/LIcrFag/7JT34SH/jAB+R/Gh9OzNVqFefPn8cdd9wBr9eLj33sY1fdp1YhMZA9jKBB54qWDgqdC+aSMGgeiUQQi8XWdYy9pF+stefav6eMMZ9Dq6ihNuoryrJfCwbCqNM45fN5VCoVCRIGAgFs3boVFy5ckCxMpsbTQ63VakJf8Aahx8uAKj9DVUYoFFpT2jaRTqdFogi0DDE12wwE0egHAgHEYjEx+MACz8/xsm4LrwkNAD1/GhSuahqNBgKBACKRCFKpFObm5hCLxXDzzTdLFmc+nxdOf2ZmRkoxcGJZapUCtKR0tVoNd99994qkZis16A899BAeeOCBFX2mX3j00UclSHz33Xd3vKeDzqS1yuWy3Je7d+/G5OSkeKBcNfJ7ZtkJ3svdOQfDBF1gjw6Izqeg80WDr5+L9UAvA6XGmCgAj7U21/77ZwF0Lyu/AGC3MeazaAVIM/3g04EBMeo0XMFgUHjrYDCIfD6PTCaDaDQqxo+Gmx4tPU/9BfGm0UFLSgAbjQaeeOKJnow7m8120CuBQADRaBSlUqkjcYieNFPidUZndxo/Jyt9LrFYTAp+6UAvJxBKDhuNxmVe92OPPSYc8P79+yWDb6nkH23AZ2dnhZ655557enbdujEoBh1o3YuLUUYPPvjgZcXSSP9xNcWSBoydkN7TSg+g5ZWWSiWUSiVYa3HgwAGpMQO0vife04xD3H777Xj++ef7fwF6gEceeUQcM64UdSyIq0n9PPD57ud91o0eeuqTAD7XtgM+AJ+21v6pMeaD7eM8h1Yi5m0AXkFL0vj+Xh28GwNj1P1+P2KxGBqNBjKZDGq1GpLJpOjSuVQNBoNSmGtubg5AK0UdgHDm9Ez5cAELy71e8pd8cPVNygeYngAVNEyLpvElnaSVLNpr40TRPV5KBq21ywpY6qDecjxo7ZE//PDDuPPOO+H1elEul7Fz586+F1naaFQqlctee/jhhztS20n9BYPBDq+duRGsQklVkw5Ws4YRy0BopwBoBXVJN4ZCIRw8eBDFYnHTGHSgFcMgzckVCktL8HxZcZVOVz6fl23XC70y6tba1wD82CKvP6f+tgB+rScHvAoGxqgDCw8U5YjZbBahUAijo6MigWKwEViQ+9FDotxOR64ZYPX7/T0r10qQt6aeXFNDAMRL6U67p3ySD7s+H/LzSy1F+xFUuxIH/rGPfQy7d+++TD46rKhUKjh69KhQW9lsVlZTVG8EAgGJz+hEMibRRCIR1Ot1FAoFyQPQKiYA8lmqrHbu3CkOCeNEDMQPcrLWkSNHRK7LwHswGMTo6KjQhKRW+BzwdTo1pEir1eqiuRf9gsso7SNIlfj9/o5CWR6PB9lsFj6fD+Pj42K0CXrD5LIJBgJ5I4XD4cvqTb/3ve/Fiy++uKZxRyIRqStD/lwHMflQcqyaaikWiyLV1O9z7ADW7WF+8sknsWvXLsnuo3HhBPLRj34Ux48fx/nz59dlPBuBhx9+GPl8HgDEA2e5CAAiAx0dHUUoFJIJOxgMipyWBrpSqcgKTCfXaG9dx4cAdGT/apAuG1TkcjnJD2FsCEBHnaJmsymrGC01BiAxMVY6zefz+NVf/VWJf631GV0K2pkaNgyEUQcWSotqLTY9cTZJ0Ms6HVHvrrYILCQVRaPRDn30zp07xXteK0/52GOPYf/+/QDQUbucZWW51KTk0uv1olgsygoCQIesy+fzrUk1spTHvXPnTqEFCoUCPv/5zy/6eT5YLI+rMejGZS144IEHJEYBQEpFhMNhqXpJ1ZLOk2g2myI/pTHThd54T/B9rnYYbI1EIqhWq1L2gpptTuw0PPV6HU899dRV1VgbAcqI6YjQYWEwWCfGAQsF6DjpcRu9H9KZ/b7fnFHvgjHmBgAvohUksABOW2ufMsYcB7ADwHR70yO2VbFxSVC7yvK5OvjUbDaRyWSERvF4PGIYuZQDFm4uej/ks7VBv/vuuyXY9cILL6zofA8ePIhGo3GZ0WQAlrQPaRbdgAO4PK1ePwS8mdkgYLVgyYNuTE5OYn5+XlYS3Th27Jhcs3A4vOjE4PF4Nm13nSvhoYceQjgc7ig8xozbxSpa6to+TA4DFr5PHSuh1w5AFEdcjdHwa2PHCUDfG8x3+P73vz9wUlFC1yHSVCInvu4kOaqC9MqU11Xvo9/8eh/KBAwE1uKp1wF8yFr7dWNMHMDLxpgvtd970lq7eAuYKw3G58PY2BgAiOFm9lkqlZKEBcr6+CDwQSK14fV6F/XAGTj91Kc+tfKTbU8Wu3bt6uC1A4GALLcZLOUNS36dHh9T83X5YO2ZrfUmo4HpxsMPP4zdu3fLeLuxlLxzz549MMZgfHx8aLyaxx9/XJKzNFUGdPZL7ZbH6rLFgUBAVmBcPbKcrFa6eDwLteu50uEKU+dUxOPxjhaD3bJd0hv5fB533HGH8NRer7cnSVVrhZYNRyIRkWt2t3skj64pRn0tqILpF+XSjWG5p7uxapGstfaCbVcZs9bmAHwLq0x7tdYin8+LAqa71C3/J2/Nh04/IMRSBn3Hjh1oNptXLft6+PBhMYDEU089hZtuugnXX3/9ZR73E088geuuuw7bt2/H6OiojIlKHapueBN3V0zkaoQGYufOxTKMl8Yjjzwif18p3ZyJHvF4HLfffvuylvJUHz3wwAM4fvz4isbVb6x2PLpdGrNDdX0g0ig0muFwWLKJWQyN36F2LsjB0zPXCXAM8usKm8yboLcKQMZDJ6FUKonahtnRjD0xWHvs2LGOe4DodwtAIh6PyypkcnIS27Ztk+vBa8lzofHn37zneb0nJibWTeWjxRRX+9ls6Amnboy5CcA/BvA/APwLtET27wXwElre/PyVPs8yunNzcx0UQj6fl3RqPoxsO0cuk2ViSSssldX3pje9admNGer1Ou68807ZlzaA5NA1yN1rOSZ5cnKD8/PzwqOyIBmNBAOTxpgVp8JfunQJu3btQr1exyc+8YkrnpPmKJdTR+WRRx7poK8effRRxOPxyya9jcDx48dx4MABFAqFZWVyAq2Vh9frxcjICICF/q+spc8Jm0E6TbNo3rhWq3V44AzUMwsYQMdKjPdqd415rYbSq4HFaAceS2dm0iEoFos4cuSIrFzZO3c9sFi28f79+0XGyAmN5TxqtRqKxSJisRgmJyelqmkvy0ksF5vRYC8HazbqxpgYgD8EsM9amzXGPAvgYbR49ocB/CaA2xf53E4AO4FWOdZMJgNrLeLxeEe1QXo0VMiQauFDpSWEV5LcLdegaz5zMSwmNdTdblgfJJVKIZFISO0aGlSqXsLh8KqXzrfffvtlyUtsK7cUSqVSh6Z+uTe0Pl+W5x0UnDp1Co899tiytj148KAYFnahYjCYShTGYYAFvpW0gC7XQE6csRMA8ltz67wvST0w4ApA7ms6AKRY6N3TkeHqgTQfv3feVyzjzK5ZvL96LQ38+Mc/jh07dixr23g8Ljp8UqJcaQAt1diWLVskxrARBh0YXqO+ph6lxhg/gD8B8GfW2lOLvH8TgD+x1v7IlfazdetW+8EPfhAzMzNiuNPptGheqRZhqVlgocyuTv03xvQkmHfHHXcAwKr5924cOHBAmmWfPHlyzfsDWi3Oms3mstP39+7dK7TQYp/ZsWMH4vE4UqnUQGV4rgSHDx9GPp9HrVYT753KFlIeuVxO7qN4PI5yuSzdpwBITRdgwduuVqtCGZDPpvfNvrLMDiU9A6AjeM77l/QLA6AM8DM/oVarIRaLoVKpSG19ruyYyFYulxEOh6UsNMtEk37jCpFBx9VmaB44cGBV98PRo0dRKBQ6+HPamV50NjI96FH61re+1f7u7/7usra99dZbN1WP0lUbddOadn8HwJy1dp96XWoEG2P2A/gJa+27r7SvyclJ+yu/8itixIEF7pOJSDTu9N67PVXe8OFwGKOjoz1paPz+979fJgugP4k//cQdd9xxRUqG2LNnT0f9mc3WnV7j3nvvFX15PB7vqGJZr9els5PH4+norBWJRGSVSEkjAKEMyHMzcYbBUV0PXyeaAejw8JmNqourEVwBMNhqjBHtOw0iOXRuTxhjJBBfLpdRKpXkGKFQCIFAAKVSadHEu3vvvRe1Wk0mQjpUHIsxBslkclHO/kpgMS/SSL1WTfXCqP/Df/gP7XIDsm9/+9s3lVFfC/3yLwD8RwB/Y4z5Rvu1IwDeY4x5G1r0y+sA7rzajhi0Yj1zZvHppS6wwCt2VyvU0ik+gL0AKZRMJrMkR9ldt2MlOH78uNAA/WjEvByDDrQeumPHjvWkyNlGQ1N0pVJJmmXTqNPTpv682Ww1AE+n09i2bZvw7Vp2x8/T4dBBTnr9pBJIV+lgObMmya8DC4oRbkvjz0xT/q2L1XEsNPycKCj35WTB/ZLL9nq9uPvuuztUO1RgUXrL/XJ/XF1Uq1U8+OCDyGQyy77PKRsedAnssNIvqzbq1tr/D8BiJPYVNemLgfx5Pp/vyEzjjaulfzprTT9MlENxf73ADTfccMXSvJT8rRbHjx/Hrl27MD8/j7vuuktWGlejaPbu3Yvt27eviYu88847UavVRG0wDAYdaKk+9u7dKxy35sF1QwYGp2m86J2Xy+UOeSK993w+f1myDOmFbi+30Wg1NAmHw3Lv0qjr+iY6e5RGvVqtythpqHVQFVjg7VmKoFqtisCAEwQ5dzpFfLbC4XCH9JdjIZ3E7Wnoc7kcYrEYRkdHl/0d9Ipi7DecUe8jeDN5PB7Mz8+LjJFeTXe6PbBQh5xFtZiSrDnNteJKBv2RRx65TE65GnRTOlrS2L0KeP/73y9L/7UGl/jg7tixA6OjoyuumT7IoMFlFyty35QgkttmshoA8Yqz2aysAOlUaE+WMR1mC7P0so73AJDJQhd30wFqnWDHbNJgMCgtAzke0j+JRKKjjAazk0n/kHKKRCKXrVT16kFPMix5qyWVWn1GCoblE4YNw2rUB6KYs84oI/8ItDwRvTzUqdzUudKriUajUq/c5/OtulPPwYMHO9QsxK5du7Br1y75f25uri9FwnQVRG3QDx06hE996lN4/vnne8Lt8wFvNBrIZrM9bfe20dDZjZz86YVzFacpCtImWuJIbzYajSIQCAjVoo2yTuvn3zTUzFGoVqvI5/MyDl3rv1qtolgsolAoCO1ChYzut1ssFoX+0wlsegycHDhZc1KhegeAbKtlwWzmzs+Tu+cKmdduvSSS6wmnU+8zKDfjw6jlYEBnT04qGTTHSRUBb2SdobcSxONxKeWrQc5eYz2/8F570uFwWFLhez0xbTR47+jMXV3bm+/RONKg6/uHSUXZbFZoFm3s+d1ThUKjr50TeswMGlJCyVovdEyazSbi8TgymQwAiCqHiqlUKiWF43TXLKAlD+T9T0MMLGRq8nzIp3NswELHMX5W3986u5YTyDChO1g9TBgIo26tlQQM/g+gY6mpjX4sFhOZGW8+AMhkMiiVSojH43j44YdXNZbFWrs9+uijuOmmmzqKXJ06dQq7d+/Gjh07EIlEBiJde7nYt28fAHTosocF999/P2KxGIwxiMfjEoeJRCKSmENKRNc7p2dLQ8byFLrOvaYB2Xw8HA7LBEkKRgdhPR4PCoWC0Bo8jtb/U9mly9PSGFPFQsOrE5BIB9Hh4HH1+bESKLXu9O51ViyADlUOV7ucAGKx2JJ1hTYzNqMXvhwMjFHng6Q9CdYoLxQKcnNzCcxAaqPRQDqdxsTExKopl6thqaQbekObyaADLYO02eSZywUNWCKREOqA3DDvMQY3AUgmKLCQ+0APmvRfNBqVe85a21GFs9lsCu0XjUZRLpeRy+Uky1gH8Lk9PWpOAlxFkINntqgOvHIbUiMA5Bxo2GmkSKOwxIGutaKVL6Ri+LdWz3AfXD1cqY/tZoUz6n0EKzHGYjEJENHzYDPaWq0mreK6qZVqtYqZmZl1HfOuXbt6UoRrIzCsBh2ABC112r3uf0kZIHlwatQZu2FXIwbpaQCpsKJ3rMs9s24RQX7a5/Nhfn4ehUIB0WhUWjQyA5jOC4/Nzl+8/ynHZFEsGnrSRkx2YpMYBl/JoXOlQo9c148hZcTromkan8+HYrGIer2OeDy+aZPRrgZn1PsIj8cjtVN0oSvdEouBJS3tIh9qjFmUB+8lDh8+jGw221H7Gmh5RYcPH0alUhmYsqh33nmnXNNkMomjR49u9JDWFTRWulhasViU5B8AohgBFiS1mv5YrHogKzICkOAiC25xW/LvOvNzfHxcnJFwONxR+IsGNpvNdlSGZFIRqRJmlHIbrcihwaehTiQSMiagFVzlPRsIBFAsFmHalRR5LXSNG111cTk1gjYjehkENUuUIe/a5h0APg/ge+2X/shau7JCT8vEQBh1Gmre4OFwWNp/kQfMZDKydCUNQ0OuA0f9wuzsrGQpLqc36EZC18zpVSLWZsAzzzzTkaymKYduh0ErXvgasKC4InSJXl0b3O/3S1NpvqaTjMrlMtLptPxN+oS0DWkPa1u9TanUIXfOdnnUj+sGHVzNav09x0rqhgZbJ1hxBcPttRqINJBuQcdA+uHDh5ddY2czoYee+qJlyK213+za7r9aa3++VwddCgNj1HWZThp3nTFKL0rfkPqz/aZBPv7xj/d1/72ENujlchmPPvrowBXj6gfolXdzyKQ3OMHpFoKsaEgjqit+skUdwXR8tmRLJpOiC6fnl8/nJTDLY9Bg8jUW3WLhrUgkIrJJltbVgVWuRinD1EFUxpZ0yV5uw+PzWOTidYa2VnXp6pGcFMwqKoduFvTKZrTLolxo/50zxrAMebdRXxcMhE4d6MyS483LZTOXrvSw6B1pTXu1WsWRI0c2bPyDgiNHjshDyVVOv6mpQQENtNaiF4tFoV66a6aQryZtog0ijSZ5dq1fDwQCQsvojE+tJmKpY3rGrC0Ti8UwMjKCRCKBVColxcTowDC4CUAMvG5EDXQm6/GzTF7i89At5yQlRPAcgQVVT7FYRCaTQTabldpLw6aO0liBTn3cGPOS+lmy6YHpLEPejX9ujPkrY8x/Mcb8cH/OakA8da1F1zU6+NCwoS+9Cu21k7sc1qDHcrF3717RN/P6aPXDAw88gFqt1jeF0CAgm82KF8yKhzSqNOJaUcIVIDn1crksgULSKNrwURkSiUTEUDJLNBQKIZPJoF6vS6MXvQ3jQ6RAaJC5/2Aw2JEIReqF5wEsNCfXGvJSqdRRTqM7SYrPFPdLioaafN2sgqWZya9ba4fWUVohpz5jl1HQy3SVIe96++sA3mStzRtjbgPwxwDesoIhLxsDYdT1cpk3u9ba6oJF9Ij4t67imMvl1n3sDz300IaqA3bs2CEPu/4BOjv5aOpqWDE3Nyf0C+MvHo9HEnSoJQdaXr1Wg1DSp1PpgYXiVJrm40qIQUoAwq/zb/LkVLHwuJoOyefzYpRjsZhQjbVaDblcDpFIRAqMaX09sPCclEoloXvIm3frz0nl8DVuw2vCOJYuGaBrLA0reukImlYZ8j8E8J+ttX+0yLGy6u8vGmOeMcaMW2t7LtsbCKMOQG4wLjfpbbL9FeVnNFh8YFg6QEu51hMbZdDZq5KV9eh5Ap0NHmiA6H0NK+65557LqAedJcnAHwt2kXrRafM6aBgKhTqKZNVqNVG8aLqE3j+94GQyKZMFPelkMolcLieyQnrH9MR5b+sSAblcrqMZDMfBcfLYpFx0FUjWjyH9Q/qI4yQ1pK+PPkd9TsOMHqpfDIBPAviWXaSvRHubrQAuWWutMebtaFHfsz0ZQBcGwqhrDS2Xx1wS8z0GsnTiBwDJomPm3rWAXbt2yXUiyB8Hg0HxErUBAVrX+YEHHhiqwNf+/fsl3qLLAfCB1fQc7xMAomSi0oNSPho6JrUxa1Sn/bPgHNBK06dEsFAoiIKFCpdqtYrp6Wnh4nVDjG75brlchs/nQz6fRzqdRiQSkQmHnrouHkY6ktSNpmd0gJUxAF1cTCthGOBlH1bSUsPsBAA99dSXKkN+Y/s4zwH4JQB3GWPqAEoA3m37dIF70c7udQA5AA0AdWvtrcaYUQC/B+AmtGqq/7K9Qp9S8o70IqgKACA3o14600hRu84bfbHejsOGnTt3Sk9NAB0PLgApyFQulyVIGAqFkEgkZHk/LGDPURpkGjkGGK21kmjE2im6WJW1VgweFSycHOmlMmA4NjaGQqEgQXv21mRpCq4qAXRUgWScA4Dw1hwbKSCuCJiZOj09LZRKtVoVow4sSCw5SVBHztpH+Xz+sgJjxWJRroXX6+0oa0APnc/f/Pw8wuGwUE1Hjx5ddcmNQQYntx7ta6ky5HqbjwJYFy10rzz1n+rihg4D+LK19jFjzOH2/0u2IqJRoixLB3X4nubYAcjyV3cmGh8f79HpDAYef/xxVKtVzM7OdrRG02niunATX9ecqrULZWeTySQajQb27NmDRqOxaTNLeV3i8bh0+tGFuIAFZ4D0Buui63Z1uk45ryGDmpwQyZ1ztQgsKLRoSDOZjHjgnCTIm+ttqTXnxEqKiHx6tVoVXpz7qdfrSKfTomcnBZROp2XlkcvlOhQ/zMLmcXXlSE4mPGdNZ1JSyWcsFAohHo+vx1e6IRjWlUi/JI3vQqvVHdq//92VNiafSC6Txpw3NR82Rvc1taDphR/4gR/o0+msHSttHHDq1ClcunQJly5dEjkeg8M07FRIcGUTi8VEC81lO7l0yv1IDwDYsIa/a8HDDz8s56FrmDAWw6Anz1Eb7XK5LFUXdb1zYKFiIT3ckZERaYcHLASduQLKZrNiLPWqkoaRx+TqkwFclikolUoSKyJPPzMzg1qthvHxcWzfvh1AS9FDg67L6tK7p5fNv1OpVMfEwlUM7x2W3NX74uRDMAFrbm4Ob7zxxtBmJK9A0rip0AujbgH8uTHmZaXfnGwL8gHgIlrpsx0wxuyk7rNSqUhQh8aLN5uuLMeHRGuKeeFPnz6NgwcP9uB0eotdu3bh7rvvXrHyhEZGp4brgJmW5tFLY7lWLq05KfIaMW4BQLT+mw06fZ3nomMLQKeXrq8ZM0uBBYlid+CdRpu0CMtXaHWWrtlCA0gHgxJcGlYKAHQqvqZEOMlOTU3J+HWJAD4LOrmIVBK5ft2UmvcEJzv9vdMxikQiUnGUTpLW4uu+rLyWuqDXsBT3Glaj3gv65SetteeMMVsAfMkY8239Zjvae9mVsdaeBnAaaDWe5oPCh0enMxNaUsYbcRDw4IMPLnmjU6HQbXiuBnbUYbagDvB1r1SAVnyhUqlIQwbGG3QwjnVKeP36za/ffffdMnl87GMfk9dvv/12NBoNRKNRKYm7HP380aNHO+gErlpYxVNrsrUcTxtv7SBQbshtaPR5vXWP0Gw2KytIGtDubGcaShpgXa6ABpWKFgAdpXTJ07M2DMfGlSsVKzTa5XIZkUgEyWSyIwtWT/TaGeBYuJLQ9xQnIzoDmmbi5BgKhXD8+HHRuA8DNqPBXg7WbNSttefav6eMMZ8D8HYAl4wx26y1F4wx2wBMXWUf4l2Q0+MNx5szGAxKHXV6D41GYyA63y9m0O+66y4ArYesUqlgbm4OJ06cWHYyh+5/ScPNh40KiXK5LBUCachpMFKplAQA6fV7PB7k83mR6/WqRvbx48dl8qHRyGQyMnZrLe68806ZkHQHHuYcHDt2TCYArUbp1t7TKNKDpc4aWCjMRWPp8/mE4iBPTSPO68tcB95//L6MMUilUqhWq0in0xKXiEaj8Pv9EvRkXRUa2XA43JEwBKDjM5wMyI1z3PF4XOJGWu/OYCivHScZbqdXt6FQCLVaTTJSK5VKx2cZf9CSTDpKoVBI3ufkQUqr2Wy1+dMxgY3Oz1grehkoHTSsyagbY6IAPLZV7yAK4GcBPATgCwB+FcBj7d+fv9q+OPvrZtJcZlKrvtxCWnv37kUmk8ELL7ywqvPqBbh0X61ygIaGDzhVLel0WiY09rKknC6ZTCIajeLSpUuScEPjpvl4epV+vx+HDh2CMWZNBZtI8wAL8lQaV501qeMjWsWh0/sbjYYoTvg/k9HIpfM4QKu8LOkMeuFsbsFVQHdWKKWDPp9PGjezeTO7DGUyGYTDYaTTaTF6HKOWEpI6nJqa6qBiNAevJYd6lcTEKJ1wFwwGUSgURB6pKTM2hiG/zvyDXC4n4+jWnXPyY6VJHb/i/UGnikZcG3pmmdIIclVQrVZx//33A2j1692McJ764pgE8Ln2TecD8Glr7Z8aY74G4PeNMR8A8H0Av3ylnfBB0+nd0WgUiURCqjEup9rgjh07OoJdO3bsEE3vagpy7dy5s6Nn6Eqg+4uuBqx7w4eUmmoASCaTOHPmjDysrEM/OzsLay1uuOEGnDlzBmNjY1IbnNeVCS58UNeSNbhv376OIlORSES8vEAgIEomng8lhJlMBta2OtXH43GZrObn58WoAC1PMxKJiF47kUgAWJAMEoVCQcraciy6HjqbSY+MjEj5iUQigUgkIoFonTpPGSLVJZxgvV5vR7IPJ6lkMtnRFo7fG3l1Ko84iUWjUaGJaDy5H6/Xi7m5OUxMTCASiSCbzcLn8yEajWJ2dlZoobGxMdTrdUxNTYmnz3udKzYtcUwkEvIM6dgCAFkJ62saCoUQjUZl9aLpLj0xN5tNHD9+HMePH1/1fbRRcEZ9EVhrXwPwY4u8Pgvgp1eyL60EGB0dRSgUwqVLl2TpTe/tSqBB4RKXy2hdyGklWK1B7wXS6TQAiGeay+Xg8/kkO5Eeqq5PMjIygkqlIoWjKOOj58hJQXvMQMsInjhxAtVqFePj49i9e/eiY3rwwQfh8XikQBYAMZ6sLMhj0GPNZrMdrQfHxsbg8/lw8eJFAC2Pcvv27UgkEqjVakIP6dR1VjCk5I/abXqq1FjTgNK7ZFCxUqkgFovJikLTOaQu6IWSdpidnUWxWEQqlRJZX6lUQqVSkclzbGxMricnDsoqY7GYeMpMCItEIpiZmUE2mxVDTvqGpQTm5uZkAtF6eU5GxWIRo6OjAFrGOZFIyLF1pyfGGfhZrgK4kmIcgAlSOmajy3JwO9JTWv/O67VZ4Yx6H3HTTTdhy5YtyGQyeO655wC0ek3SO6BH+f73vx+f+tSnLvv8kSNHMD/fym1iWjSXmrwh3/ve9yIYDGL79u0D4VXcc889eOKJJxZ9b/fu3cKDav05lQmXLl2C1+tFLBbD6Ogo3vzmN+Nb3/oWxsfHYa3FG2+8gTe/+c3IZrOyDOf2DMLRk+M1oqGfnp7G448/Lty3rn1CMGOVDze521wuJynq7GvJLEUdPNy2bRsCgQBmZ2elZSGTXmKxmPDkNC5sqaYNODlh8tQ0QjwPepGBQEC8Va76uGpgkJU9REk3UPmipYJc3bDDFqtgzs3NSS/dRCLRoQFnE3MGsBncZ74AqRlSbMViUSbu0dFRSWpKp9PIZDIYGxuTiYqTJlcSuv45tfkEnyFgYRKjEddJfYTOVi2VSh1af15znTS1GRPaNquyZTkYCKMOtBJKNMjT7d27d0npGqH7TJLL1d5EMBjEyMgIGo2GeIgbjaUMuoae1Bgc9Pv9GBsbExolHA6j0WgglUohmUzilVdekWzH8+fPixEkh0pPUgfk+D4NZXdaPblgrpiq1Srm5+dlbDQSbAzh8/mwZcsWWeYzsMvJpVarYWpqSopakRqiJwgslK7VY+pOred3nMlkOip8MvmGBpa0DSWB/CyLaVG2SFqlVCpJnXStJtFBR3q4zWarR2kul5NgKOkMJhxpnTo5cl5/cuSFQkESo8i/k+KgZ80iW1yRaLmmpoeKxaJMRFRQ6TLMwWBQ6swDC/EsXUO9VCpJM3hdloClsHU53xMnTvTkmVhvOKO+QZiYmMDc3NwVywA88MADOHDgAHK5HBqNhgSZgNbD382nnzx5UhJI6GU9++yzfT+XpXDfffdhcnIS1WoV2WxWMmj5sOoEGnqZVDjwQY5Go/hv/+2/IZVKoV6v4/XXX5frFY/HZXsGVrtre9DA0LjzIaYhpWFngwid1ELDT4NJZQe95Gy2VaCOXmU2m8Xs7KwY0XA4jHg8LpMIKQU9CXEyKJVKIvOj+qc7EFqpVCTtnwk3OjHIWivp9VqZoo0uDaKu4cKkH2MMkskkYrEYEomEFN/yer2XyRCpSNGrDypRGLykd81VQrPZRC6XQygUkh6kLE/QLY/laiYQCHRMQNyWq7FoNCoTCydJrqhYAqFYLF4WLOU9orXydJSY8LZZ4dQvGwQdYc/lckvSFrfccgtee+01AAsd5bnE7QYzKY8dOyZGYiMxNzcnSSEAOpbKwAKvTu+NKfJzc3NS+ElzvKlUSvhXnYlIQ0uVCA03l9O6byYDZBwLDQePxZoyOjV+bm4OgUAAiURCJg0aGapKcrkcCoWCBEj5WRo0v98vlQ4ZRyAYLKVUkZMPA4PNZlOoIRpZnZ7PCVAnHJG2ARZ6edJwUymiE3gYLCT3zAA1z4MJQSyLS6qLEx2TkSiHJHU1OjqKSqWC+fl5UYBlMhmp28MJ1Ov1yqRXrVZlVUaDzNUIVxSVSkVWIuyZypUGvz9gwWhzEtDJaTxvyotHR0cRiUQ2fa1156lvMGjclwri7d69G/fdd9+KmkAUCgV5sDYSW7ZsEYpF1x7RQU5NmWhFSbPZxPnz5zE6OoqRkZGOhKxisSgThU6+oZcFLAQ06ZmRfggGg6LTppdL2RuDgAyaBQIBTE1NoVgsYuvWrUilUrDWYnp6WioT8qdcLoshYiCRWbOU6dHjp1GlNp3jpnSQlRZpZDlWTkrk1Hlu/OFEw+tLL5fZuLpJhK4AyUmARo58OCkYGl6Ol/cVv1O9kiHlRAqMvDuPz25JVNIw0AtAEpN0VcV8Pi+rTtIodAooGmDFSVYzpeST58AVBu8Drh7oJBljBqa5+lrhOPUBQiAQwO23345IJHKZbn2lXX0+/OEP93JoqwYNEiV/2ijxRytAAoHAZZ4pvUOfz4ezZ892FK/SniWNXLFYFIOpKQpSAkBr0uMEMzc3h1gsJvJJctk0eLlcTiaEcrksEyYnEY47Eol0VAWcmZmRIGWj0UA+n0c+n5dsU530QjqARkjz55wwSCnQcMZisQ6tuc4U5b7j8bisXChvJD9PGkPLNXO5XAfvTg07C29x7OVyWeIf5Mq1Aol/a5qHXjVpKY+nVbArnU5jZGREOHp62IwL8PvnigxYaMjBCYlUE+8TKoxYv50TsHYMuN2gZG/3EsNq1AemR+lywaDO3NzcsoKNmwFal01obpOeVje3zZuSEwEzV+nR6pZqNOjkpXX1wFKpJJUgWSWwVquhUCiIgZmZmcHc3Byy2SwuXLiAQqGAN73pTRgfH+/wvlnHRBsJzQHTy9aKFR3049KftU0qlYpI9vhDlQzjDRy3DpBTd673V6vVxDhTQUJP2efzyXtUBFHXT3qMgVtOHtVqFTMzM6IgYeYuV0GcqPVqhKsETtTUvafTaSnoBUAyhhOJhIxHZ79qPljz37pPqQ42U8nCyZd0Gj/H1Qqvly6HoIPsg1hfabXgd3O1n82GTeepMxHl05/+9EYPpSc4duyYqAyor6aED4DIxkjNMGBI7pecbKFQwOzsLG6++WbMzMwIp60r95FqIb9LYxuNRsXLZHVCGn8GF8fHx8Vo62YRIyMjyOfz0tmHEkOgs2kxg7xUolARow2v5m51rKNeryMajXY0iJiYmAAAZDIZMVycwEiDAAtt2aisIe1C7xqAGE1OdrzW9JxJ+XD1wes3PT2NYrGIbdu2Cb3Fz/P46XQa119/fUe5Xz15x2Ix0e1TAUMunPcFFTo685Sxh0wm01HgixJUXfaAkxAdBNJLWtvOlQ9XDOTkdZlgnZ1LHDhwQAKmm41j34wGeznYdEZ9WCrEEVSlUMVQq9XE8HUH3Pjg6povlDUCC8W+wuGweHRMfSctwkJa2oMl3UCDyhZqOjGFDzspCHrtNODz8/MIBAIYHx8X2oWBURolGsNQKITp6WmhiCYmJlCtVqXHLOmaarUq2uzuSoM0eiwH0C3TJGVTKBQk+YZJSMBCPCIQCEhSEWkvcs36OFT7MK0fgGjQGbSlYaSGnR45VUjBYBDpdLqDPpmenpbALOkhXitKBgF0yBsBiIfN60KvWlc71WNPJBJyTZmBGg6HZcWiZZEAZIykbnjO9XodH/zgBzsm4FwuNzBS4eWCdN4wYqDpl/vuu2+jh9BXPPLII7JE1l4Ql+o0YprLjsfjkjFKBQO3ZfNjKn5oAEjV8MGnxC0Wi4m8kEZdG0B+tlarYWRkBNZanDt3TtL36/U6zp8/j+npadHP68qBPE4ymZRVQSAQQDKZxNjYGLxeL7LZbAcXzfICXq8XY2NjGBkZESqkWCwiHo9jy5YtklCk2/eRfiFnTPoIgOiyq9WqnBu9f64uWF+Hxo5GjgW7uD8mKXEMTKkn7UPDyEmF8kk9XpaBYECVx6WccsuWLR0Fz8h1a2kk6SR+jpMQ+XsqZ7QmXU8UDNRqWaouIUHwuyQdxXgDJwi/34/nn3++fw9Kn+Dolw3ASgOfx44dQ7VaXfHnNgp8UMlb8wbSMkJdCjYSiXR4SFoZw9TwZrPZIYEjN05enV1ytmzZIoHJSCSCeDwuxmV2dhaFQkEyVum1VyoVjI6OYmpqSlYJ9Oqj0ah41IlEAhcvXsTIyIgYBJ4TE2NCoRAmJiZknKVSSSaaQqEgyhtOXFp/Tq27pgh0EJU8MGkEatq1MdQJQJr+0eV16VUzC5QGnl64Tv6id8z9cgKemZnB+fPncdNNNwFYqFvDCYQqo1AoJN8fJ3fSYKReODFy1cZrymtPDp+gN8pJRmeZcltd1IvXAejMQuV11tndtVoNn/jEJ/ryXKwXemWwjTHvBPAUAC+AT1hrH+t6PwjgRQA/jlaz6f9grX29JwdfBAPtqa8UtVoN8/PzUvZ20MElNT1yptUz8AlAFCMMetJY0WjQS6Ph5zI5FAoJLaEfdtIU9DpZKIxLeXqoNFZaDw60HoREIiEUiZbtUetOA00qh4E58uizs7PS0YexBNZG4XlS3qhpBk4aNFTRaFQMra6QyImQ10lTMpwAqQQBFtopUv45Pz+PfD4vxiwQCMiKg6sfY4zEKhjwjEajQk8ZYySxJxwOSyliGmYdyKaunCsHfl7ryJmlyhUIKTk9adPL5jG44tKldrtLTnBbnYymvXkNTix+v39VBfIGDb3w1I0xXgC/DeDnALwVwHuMMW/t2uwDAOattW8G8CSAx9FHDLSnvlJ0pyvv379/oHW1Bw4cwLFjx8TQEAx8hsPhDn6dxk0/gPV6XTTKpAXYuQeA8LSkeSj1y2QySCQSGBsbE2PI/VGGSM+YHiwTiDS3q4tY0UCl02lZTfA3AOlYzwQmbYw5iVBiODs7i1AoJNQN6SVObGxIAUCOw9UMsJA6z/MBFkpI8D16p6Ojo3LNWJyLk2IsFkOxWBRjTfWLbh/IwGg2m0U2mxWJJDNMGajksXU1RX4fHB+bhlcqFQl8MsipSxVwktT3Bc9Tc/laDssJkhy6VsBwomQwmteQx922bdtQxbN6SK28HcArtlXcEMaYz6LVzvObapt3ATje/vsPAHzUGGNsn7idVRt1Y8wPAfg99dItAB4AkAKwA8B0+/Uj1tovrvY4a8EgG3SCXhppiEKhIB3qyZ8ymYRKF10HprtWyujoqJRLaDQaSCQSCIfDyOfz4mnF43H4fD6hQgCIJ8kgIb1lYKG/J7CQ0KKrPtJo69R0rhxIHbBOeLVaRSQSkQAwg3ekgrZu3YpgMIi5uTlUq1WMjo4imUx2VIPUWcI0hFo6SeOlA3z0imkEadDpGWvtvt4neW0aQerIgRbtEY1GJYuT0kZOvDS6PD99bB2U7KY3dNKQVgPp+jx6G8ZRtLqH+QbcJ71yPRloPTvlkLx+Ho8HL774Yq9u84HECgKl48aYl9T/p22rcxsAbAdwRr13FsBPdH1etrHW1o0xGQBjAGZWPOhlYNVG3Vr7HQBvA2QJcg7A5wC8H8CT1trByOwZcNB7y+fzUqJW1/ZmVUDK+5hsQiqCWZUM5G3ZsgVjY2NSwpXGRAdBqQ9vNpuYn58XeZ/mg0ldzM/PS5IRAJkQSKmwtG0wGMT8/HxHMhNld5xojDGixw4Ggx0adyp4ms0m8vk84vE4Ll68iDfeeANbt24VFYwuZ6sTb/R5ApCqlMFgEDMzM2L86LFSXULvlCuFbmOpV1Ham6Vyp1arIZPJyHc2Pj4uXHsymUStVsPs7CwASCCb363mrZkoxmPzPLqlmBw3t9W0mC4nzPPguPl963Pke1w9cCI2xmzKwOdKsQJHecZae2s/x9JL9Ip++WkAr1prv68TTRyuDnp5VFjQQ9blZWnwSS3o+iU0rqFQSJQsNCo6C1AH+ehtUodOL58cNeH3+zEyMiLZq0xHr1QqSCQSEsgLh8MYHR2VwC9533Q6Da/Xi8nJyQ4qgXVrAMhY+R6DqFplMTc3J9p7cuGZTKbDyNHTpdHmedCQc3+6nk63p0bDRiPI+IVWqLCjVKVSwZkzZ0TfzviG9tzPnTsnlAmTpfx+f0eVQ8YOmJ9A2Sdru1Dqys/w3mD8ghmmjAEw1Z8qIrYzpG6d14mUFVcGjA3wml8L6BH7cQ7ADer/69uvLbbNWWOMD0ASrYBpX9Aro/5uAJ9R/+82xrwXwEsAPmStne/+gDFmJ4CdAHDjjTeu+IAf+chHsG/fvlUNdlBw//33i7HUNxhlebpOCA0rg4p+vx9TU1NifPL5PObm5oRC8Xg80m2HkkGPxyMcPb3M7du3Y3p6GqVSSSgOYKGRRiAQkDrpuqYI5XOcCLLZLC5evIhSqYTx8XGMjY11pMSXy2Ux5FTxkJahl6pT871eL0ZGRkQRwtR/duOhpp3nyuuomzrrRCvSWOSQyXHrWun8SSQSHeoYTh702lndM5FIIBAIIJVKyWREySK9dyp8WP6AEwwprmQyKR2fWD+Gwe25uTmhwwB00GKcFDjpaAqJ5QV4DuxzShqNTgANPWWJzWZT+hkMO3rIqX8NwFuMMTejZbzfDeBXurZhe8//DuCXAPzf/eLTgR6oX4wxAQD/FsD/2X7pWQA/gBY1cwHAby72OWvtaWvtrdbaW5kduBJsZoN+5MgR7NmzRzxTyupoRLlUpt5Y18KmFxoIBBCLxeDxeDA7OysNQGKxmChpGLjUCg1dGIrBRiY1kX+nYZmZmREuF4DQNtS353I5ZDIZXLhwoUN58t3vfhcApM0bk12090hDzo5W9KC5EqDhYltDFhFj+QEG/2igtGKGQVuOh3ptSiOZKs+JgDSOLtZFb1xTNKVSSWSONKi8vpSU8pqTP89kMhLPKJVKUqBrMdUQVwYAOr5/TvB8TdMsulCbrutC5RMnM54TJwiu4nRxtI3s9LUR6IX6xVpbB7AbwJ8B+BaA37fW/p0x5iFjzL9tb/ZJAGPGmFcAHABwuI+n1RNP/ecAfN1aewkA+BsAjDEfB/AnPTjGUIGNIbTcjA8jANFVkx4h5aBrsnASYLCP3igpBD7kXL5zGU6jSWPo8bR6clrb6qNZqVSkXd4rr7yCarWKm266SbxmABL09Hg8GB8fR7FYRCKRwA033IBisYhXX31VShiwVd3s7CxuvfVWnD9/HvPz8x0KEJ3NqGWNTM+nB8/VBtPi2cmJxoyUlaanGB/QvLiuq8MGKzTOjUZDAr6kW0gHaf6dKyetGPF4PJLByeNwwtH5BvSc6bFz/9T8M6uV3jUAWdnwtXA4LI2xGfzk+XJFxVULJxDeI5xwNP1zrXjoGr1ylttCkC92vfaA+rsM4N/35GDLQC+M+nugqBdjzDZr7YX2v78A4G97cIyhAo0IA3Y6E5IPNn/oLfBB1WUDaLBoMACIJI6GnctrXWObv6mwASCeJHtbsjl0vd5qbswmFvS2STcwa7TZbJUAbjQaQomUy2UpyUuemqoWGnE27dAJQ81mU/h/Kjk050uDyBUDpZKcpEhJ0NOlNp/XjJw8JwBeC64mdJMVTrr0zjlZpFIpABDJIycXBjGbzaZQUKlUSuIL/D50SeBIJCKdobhS8Pv9mJiYkOJjDFzzu+KKDoDcHxwf+XVOesw7YBkB3i9c0W10P4GNwrCWCViTUTfGRAH8GwB3qpdPGmPeBsACeL3rPQcsLH1ppGhwWEOby3UtKaSRJy1A48wfeq/a26QnRvkj+WF6wCyVyyV6pVIR7nvr1q1SCuCNN97Addddh2g0Kkv1QqGAkZER2WehUMDc3Jw0yGCDC45Z15TheZOGoL6eMkB6qgCEftLSQ+rsOdFxsuG1oGHXtUu4ctGadtaCYfIWVwM06CzNy0mBhpBSzUajgWKxiHQ6LVSaplXYzYrqG3Lt9LSBhYbiOomM3xn5ed4L3U4A90kKT2cg69WbNubAgpIqlUrhgQfEobym0ENOfeCwJqNurS2gpbfUr/3HNY1oiLFz584OjlRrwekJUpqmu/PogBgfUk1H0NAwiURL1yhRJH9NyoAeMCcU6qmtbdVgGRsbQyqVwsWLF5FOp3HhwgX8g3/wD6SSIdP45+fnkUwmUS6XMTo6Klw5e3ZGIhE0m01MT08LZZLL5cS4z8zMIJlMAoBMGNRL67riDFoGg0Fks1kpUKYlkZzMeA00t87X6EnTq+YxOamQAuKEQw8egASpaXAZDGVtHX5v0WgU1lp8+9vflqzTbdu2SVITuW161TTc3Ulc/P60zFI7Al6vV1ROnPA4MQILlI12CJjgNSi9BDYSzqg7rAm33367PMh8SOmda6UFjbAOnnKZSL6VMjZ640xhz2QyInNjUSqgdfPSKyV9Q5keg5/0JjkOdhqKRqOYnJzEpUuXcPHiRdx88804c+YMotEo5ufnpTCVpo/o4dPosqQwJxRjDIrFotQs4RhplLg/n88nPVvJYzPQSuPLYPFiQUQ9uWnNvr6eDKZyctRSUdIyOimnVCpJjCKbzUr5Xa5sSMVw4mYsg3Vy6DFzhaPjClwJMIVfe+jAQptBALL60Zpzfp77ptEiJffMM8/07wbfhHBG3WFV2Lt3b0fCDzlVPvDkO7tbitG4agUGvTePxyMFoLRn5vF4xEOmASC/To+X3jN16lwBUApH48PCWsFgELfccgvGxsZQr9dx5swZ6cCjsyJ1HRuPxyPp8TSsVNroyn4s/sWa78YY5PN5KVHL4mM8BlcaDNgWi0Upc8Cx8Dg0bJwIs9ms1E+hEaUcU2djanmp5vA5+VI9o+ufb9++HVu3bsW5c+dQrVaRSqXg8/mkkmY6nRZJqDbCmo7iCo7j4WTHa6tlh/TeWR9mYmJCYghc6bBKo9frHZpmMr2GM+oOy8bRo0dF4aJrbpAG0AW4GADTlIjm2XUSjdfbansXDocxOTmJ6elpFAoFydbUHDx5dHpsfNBJFZAXZto9NeBcyrOFXSAQwOjoKEKhEKampiQrVMvweJ7ktUulkmiu9XlSJkl9PIO78/PzyGazokFn3ZtqtSoUB7f1+XzSl5Ses86upMyQvDU5bk11cdIB0FFLR6+Y6JFTysgVCCdBKku4EuF3B0DoMN3GjteVgW2Og9tx1cQVA+uycCJi6Qfy+7rOS7cDAMB55VcBn69hhDPqPcaOHTuEtyQlQOWB1p3zhtL8MZfgVD/owk3cJ+mZQqEgZVlJWbAsLVPUrbVSQoBBNgYsE4lER1Ylk4lYAXBkZES6xjOb9aabbkIgEMDZs2dRq9Vw4403CoXBCYMVHNPptOyX59loNBCLxST9PxaLCR9O2oiTD71cTgg0qJTg0dgDnfVogIU4RHfaPxOu9EQDLEyYHCMzY/WqiFSWpmkikYiMP5PJIJfLSaxBTxRTU1NyDai711QZDXMul0MsFsPIyIiUT2ZNHlJijFnwvtDXgEFex5cvD85Td1gUe/bsEU+JFAor+tGI65rY9Nypm7btIl40vPwcFQqkLxiI05poKl3obQLo2C8ASTiiDBGA0Ai6xjg18TRi3O7ixYtiVFmal2VyafjS6bTouYvFIkKhEMbGxqRUQL1ex9atWyVQykQi/sTjcdlWV5KkoSVVw0mIBl0Hj7UEVP/2er1iPLkfzUfzOlARpHuAcqXEyYL8PgCRWAKtlnXnzp2Tyo3MI4jH41KH/eLFixILYEkBeteceMjVp9NpUSTF43GJMbDyoi6fq1cXnPAclgdn1B0AtLJB6SVrrTVrr+j0dZav1a3gaFj18pkGhgabRk+XyyWlwO3Z75NNLnQZge7myqRtqF/m/3qSoYac46QGnZMHk3R0/07NjzNLk59jgo3H45FyvPF4vCNpCIBklHKf5PUZE+CqhDps6uNJ9ZCW4IpAp/MDkO3JQWu1kTb2fMBJb5BLp9STwUZdkKter0uyUTKZ7Kidw5UVK2uy/nw0GpVJLBKJdHyvjEfwO+b9ASwkUdGAa3Cl4/P5Lis/7bA0nFF3AABcvHhRqAB6yZTa6UAhVQi6zjWADg+SniEfYgYxu9uW0eOnIdGqkmg0KrVZAEhSCpUizFSkdxePx9FsNvHGG2+I0R4fH8eWLVtk3FTD0CDyswyu0vDwN0sPcHycoEg7zc3NybnqJCldt1un59PIcfxMvuG1ANBxLE6sOrOS14EqFF0al548qTBgYXLl5MDvolgsymRAOkwXDqOh1/vhxEYenTp6Um3s3crrqxtRk1rR8lQacq1T14Yf6CyF7LA8OKPuAGCBA9fBKholBrV0iVgaLy69dUKOfpjpmXL5Tw9STwx8iJlKrz1GLvu1keHEQTULaZRsNotvfvObSKVSuPnmm2UJz96i5OiZqam9Rxovqmm4MtF6ehatomqE9I4uE0vvn8aR8ky9EtGlbzkBaONKygmAXA9dN4ccPCWENIz83jhWPtx6QmKmLAApmMXgMycJTuikfnSnJgbDOWbSIwyokmrh+ettdeIZJxquCnVwlZ91xnzlcMlHDgBayUN8QAGISoH9IelF0ZMOBoPieVerVeFS6YXqZBJ+ll43KR2qU2j0NC3h8XiklZnf78fZs2eRTCYldZ0SRlIlenVAz5iyR2rBWadc0x2kGLQ0kQ02yP0CCzppyv6YzKRlmjR6wALVwdcYN6BRpCafZQh05iuvAa83+XetdtGTiKaStMevDSJXURwPDSsnDFJcNLo0zOznSsqLpRYYp+BYOVFr/TpXe1TXMHbBc+A4tdaf77HzEoOthw4d6t/NP4Rw6pdrHEwe8vl8Uo2PXpJuF8Ya43xgdV9KUjPAgmKD3Dw9se5iX3xodVKJ1lNXKhWR5xE03NR/00MMh8OSEfqDP/iDHR2VyLGzropOTKKxpQfM1HlSQJyQKJcE0BH45TakhQDI+PhgaV5ce/I6w5STF7DQnYfGX9M1nEA09UVlCOMWBK87PfFuj5nbcKLhpMWVAAPLXJHRg7fWSocnPYEwHsPvleNhf1KuEnSHqMWkrdSn79+/vwd397UJ56lfo9i7dy9yuZw8SOTAabTpYfGh1jQMvTtK/VgUS/ecpAGgQaFKht4i90+ZIL1XrYtmwg5VFqFQCCMjI/I51oYhH8wmF9Spk27hedKj9/v94pGTGpqZmUEsFpOUeU1f8BxoaKmi4fnSgNJAUsvOHqXkrqkgosHXpW415dI94QCQbFwGmXm8aDQqBlun3tOb54REo0oFDikcHRfR8QFy5TMzMwiHw0ilUiKJpDwyHA5LbRxy5szurdfrohQiODZObjpjlhQRAJGNOqwOzqhfo/B6vUgkEsILZzIZAAu6ZwBShVBnVzKdXfOwTNChR6d5PV1NkDVZuArQQTh6igBkhaDlbxMTEyiVStJtiDQAAFlBBIPBjiqG5LE1r0/und4oA42sPV4oFCQ1XgfxNDVF46pBo6QrSwLoKD6m6RJOet1UCNv8afpLQ383/B65L77P1yj3DAaDHcFoBj2p6mHAUvPuTLbSAdzu+AAzPnXN9VKpJA0wSOHQGeBxSMN1J8o0m0185CMfWcXd7EA4Tv0axB133CGUAR88Kj8AdNAA9BYjkQgqlQqazaY0m6A3pyV3fMC1Xlp7YMBCJUd6wsykzOfz4rWRX6ZHOj09jX/6T/8pXn/9dfEWadgBSIISPWxK9igjzGazKJfLmJycFM+4UCggGo1K0gtliQyGknrR2bKaL9bGSKeyk+oZHx+/LKuSiVc0ZjTM8Xi8Q+/PCYQTCikUjp20yN13333Z98usX3L+XFV0G2ROtsViERMTE/K96ImQpYbn5+dFuknvHIAYbMoi+f17PB4p96tLCOhJQ3cqYoXHj3/84724xa95XNNG3RjzPICfBzBlrf2R9mujAH4PwE1oldj9ZWvtvGlZr6cA3AagCOB91tqv937o/cO+fftEyaANBo0vJWuc7blEpwabxsVaK/pwqkV0UBGABDypZwYWshYJXSBKdzHiw8/Xx8fHMTo6ivn5eeRyOVhrkU6npVkFJxzWSqERmZ+fFw+RgTprLcbGxiToSvkdPXgtueM+NS2lvUsaRxpsfa405NyGtJIuSkXOm/tnIPLpp59e9Xf88MMPAwBOnDghBrVWq0mnIq5IAEigG1jg2rXmXZfb1clAunaMdgaob6dslddIx164muP3y/uhO37isHpc00YdwAsAPgrgRfXaYQBfttY+Zow53P7/EFqdkN7S/vkJtNrb/USvBtxPHDhwANlsVnTo9DhpYIAFD51GzhgjreDIcVIdodP9aYwYDNNKFH0M7VnrQBoLc42NjcFaK141jWA+n8e2bduQTqclgzQYDGJ2dlbiAFxRMNjm8XjE82e5AabgMwBIaiWfzwNotcCjfrvRaEjdGV4bTXNoCR8pIqpGOFHRs9af5zgp3+M1ozcdDofx5JNP9uQ7P3LkyKKv79mzR9Q7uq4N6SsAEjvQcRaeH+MFpOC6dfqcnDm5MU6hYy6c8Gj8o9EoTp061ZPzdrjG1S/W2q8YY27qevldAN7R/vt3APw/aBn1dwF40bYs01eNMSnT2Q1p4HDgwAFRoNBLstaK/JAPJ70zBvWAhRKvpBa0TA9Y0FXrwKimCejNMzhIo8geo/l8Xrh1yhW5JJ+bmxN+1ufzIZlMiheYTqcxPT0NANi6dSvq9ToikQjOnj0Ln88nXvj58+c7asroFH3GBTRNpHXwDGZywtD0gU6YITXD94BW/9JEIiGG/Erdd3QyUj6fx9GjR/t2LxDdq4CHHnpIJj5+96TXqO1nbRufr9UInFU0eZ10wS+2EEyn0xgZGZH7hvsl984VjrW2ZxOZw/px6saYJwD8HwCqAF4F8H5rbXqR7V4HkAPQAFC31t662mOuhVOfVIb6IoDJ9t/bAZxR251tvzZQRv3AgQPibesEEHa6154ijSqNHW8GGjM+1N21WGi4afg1+BDTE6RaRi/1eZxwOIxoNIpoNIrz588DgHDhAMSYUGkzOzuLiYkJvOMd78DLL78MoEXhsBF0rVbD7Ows4vG4pK83Gg3E4/GOIldUajQaDTG6upaNnqT4HkEPltdPP0CUK7JGDj1c8vTAgs6cXi75a57zekN3CDp9+jT+/u//vkMKSm+atB016qRbqGShY8A6OTrjVdeRp3NBPbqm4xx6g3WiX74E4D5rbd0Y8ziA+9ByfhfDT1lrZ9Z6wJ4ESq211hizoitkjNkJYCcA3Hjjjb0YxrJx8OBBkf3FYjHE43G89tprwidns9kOqV532VZSMaRY6KV112jh52m4IpFIhxfP9+kh68xHXVOGCoypqakOr5AG4R/9o38knYWoyJicnMRLL70kwTlrLS5duiSBStI3sVgMs7OzEnQkf5vJZESWyGPpiYnGmjXXKTfkxKQDjbwGzA5lAlcmk+motUK6R5/bRz/60XW4I1aGnTt3AmjJXem587pr753fnU4gIh3DJC9OZjoPgZMeE9hKpRJ+67d+ayNPeSixHkbdWvvn6t+vAvilfh9zLUb9EmkVY8w2AFPt188BuEFtd337tQ5Ya08DOA0At95667pGLK5UmvTYsWOiz9beNrBQ/5rGjZ4Zqxd2e/D8WwcBKUvUQbXuWiSsz+L1thomzM7OCncfj8cl43NyclL00RMTE/ibv/kbCZYWi0V897vfxVvf+laROGplCVUYTJYqFouIx+Oo1+uip6dXSV5cB2g5CekgJ68TlTWUCWpJJikZfW3pmfr9foyOjori5JFHHunbPdALkJ7jqo7KH120jZM8z5FGmwqpUqkkCWE08pxMGXv4nd/5nQ0+0+HECoz6uDHmJfX/6bb9WiluR0tcsuhwAPx52zn+2Cr3D2BtRv0LAH4VwGPt359Xr+82xnwWrQBpZpD59G48+OCDAFoBNNZnYZo9l8u6VgkDYCyXy//plXIC0Mtw/tCYU+fu8XiQSqU6amdzEgCA2dlZMSCBQECaZDQarV6X8Xgc4+PjEmi7/vrr5fO5XA7hcBhbt24VbzqdTovXSPolmUzie9/7nkgYKXvU9Au16aRl2DmJhovj5wSo6RleOwCSLaonm0E35BqPPfYYdu3aBaAVQNaKJFJinLh5rpRosgkHJ8pCodARS2HyFHMMHHqLbu3/VTBzJY7bGPMXALYu8tavW2s/397m1wHUAfznJXbzk9bac8aYLQC+ZIz5trX2K8sdoMZyJY2fQSsoOm6MOQvgGFrG/PeNMR8A8H0Av9ze/ItoyRlfQUvS+P7VDGwtePrpp3HmzBmcPHly1fugd6opDerEmR4PQHTfAKRErK5qqD11Gk+dIamLQdXrdczMzAgdQwPAjEj24+REEIlEhBengb548SLGx8fFaDA4xwzOv/qrv8Lo6CimpqZE+w4sJOd873vfEz26bhbB8XRLDUmTkHYg9cKkGV3nhPGIcDiMF19sCamOHDmyqcvFPvPMMzhw4ACAhWzW7qAzM2aZ68DyvaypzsCprtFDyiaVSm3g2Q03ekW/WGt/5krvG2Peh5Yk/KftEge11p5r/54yxnwOwNsB9M+oW2vfs8RbP73IthbAr61mMKvF4cOHZealEerOZFwpKC3TmaM0YpSj0YOmRpvyNL/fj1KpJAa/uzqj9uhIR+g6Jlq/rItk0YDSI2bDifPnz0sAlfVC8vm81H5pNBpIJpPw+/24+eab0Wg0MD8/j5GREQn2cZXBkrNUv1B+qTM2GVOgwebqgedIz1zHB1h8ih45sZkNOnHq1CmcOHEC+XxejDUTlkhjsXE2VzekwPi9RiIR0d9z8gyFQtiyZctGn97QYp3UL+8EcC+A/81aW1ximygAj7U21/77ZwE8tNpjDkVGqU5371VgjYk6lOZRBUIjzwAYPTNSK+SDtWaZ/Dm5U3rbOrmJnnIymeyogMggJLXRpHaSySRKpRLy+bx499ZabN26VbxpTjpjY2OYm5vD/Py8BGC3bdsGAFIyIJPJdNROZ3BXZ4bSYGtNPb1v3WhCq4doxEZHRxGLxQC0JqqDBw8OVdu1br378ePHJShKCoqTHOvgF4vFDsULVTC8X1yf0f5indQvHwUQRItSAYCvWms/aIy5DsAnrLW3oaUc/Fz7fR+AT1tr/3S1B9y0Rv2RRx7B/fffDwA9TcjYt29fR5IQ5WmU1DHoRdDrYhKPNvy6mqEuZcv90gumVh2APNys0shkJN3thglJVKYEAgGRNHK8XNqT76dRHhsbk30zoYlFwpgpa4wRg8NzpeHhg0AFDqklriLokZJy0WodrWnvrtUybDh+/LiUwq3ValIfhhOljqvwOunv2LWl6z/WSf3y5iVeP48WTQ1r7WsAfqxXx9xURv3gwYOiSgkGgzh27JgENnsFLVGMRqMd2Y2BQEDqeehyrVxyc2w0tJoGohEfGxsDgA7aQksbGeRMJBIAOiWUQItjrdfryOfzUmWRShTWqqnVah0BNx0bYLILA73akyStwxZxOv29uw2fppN0rRR2QQIgxcZGR0el0fWxY8d6+n0NMjg5BwIBWVXxmjO4zDIRdAoACF3m0D+sV/LRRmDTGPXHH3+8owMOgA5utlfQNIP+n0adx2w0Gh0KBqaU64i67m7DAl809DT+pGcAdHjCDFIaY6RgFCcNenWkiDjZkIKhx633S7qISpvvf//7GB8fl3IIpVIJly5dQigUkgxVXayLY9eeJb1wXqvR0VFEo1HMzMxIss3v/d5SCq7hx6OPPooDBw6gWq1KMTjdTYnXj4XROCHrssoO/cM1XSZgEMCl7J49e6QBss7y6xXY3T4cDneUU6VnRUpFq0KCwSAymYxovJkSnkwmJeFEVxakQWQyUndpAe1908jn83kxyLpPJsfHICQ/F4lEMD09LRJMJkbV63W8/vrr0gyaDZOpr2bQlAFNGiQAQh/pSYwG3efz4brrrhMlkLUWx48f7/n3sxlB+ovXjQooXVaBXj3vDaqSHPoH56kPCNZSmW850MGpffv2Cc0AQDIz2eWHdT4oZRwfHxdJGgCZBEh/UBVD7x9Y4MdpwMldUymhtd2UMJIayeVyHRmauVwOW7ZsEbUNg53WWhQKBSk1W6lUMD8/L54hDQgVMJRfcrwsk8DJjFQSSwaMjY3hueee6+v3slnBeA/17Pq70YXVtLrJ7/fjiSee2MhhXxMYVqPuufom1zYqlQqy2ax4udQZU8KYz+dRrVYxPz/fUU2RMkQta9NGm96/rgvOVnHFYlFKwFpr5Zg0uhcuXJBJZfv27cL702AXi0WpGMl9x+NxXHfddUIH3XzzzYhGo9iyZUuHFJQJM6wJo9u6AZDz0oFjzfk7LI6JiYkOz1zTbwCEbtEUnEP/oJMAr/az2bDpPPX1wv333y8eNIAO2R6NOikTZlkSVIyw8QMNeCqVkkAkwWp80WgUuVwOly5dEkVMd0mCdDrd0VmIPH0ul0MkEpFmFufPn0ez2UQikUAwGEQwGEQ+n5cAsw7uApCUdRoUcvP09HUgFICUDuAYXdOGqyMWiyGbzUrwGViojU6aj6u6F154YSOGeM1hMxrs5cAZ9S6cPHkSU1NTwtsz2Yjp+Dplnpy67uPJHxpBUi06WzAWiyGfz0ufStIs7PvJY1NW2e3lh8NhkQ6St2fj4rNnzwpVomukezyt7kLsh8nJiefChCjWgiF3r8sRc5Ihh09JpsPVcc899+CJJ56QomrRaFTuKa7s9IrIof9wRv0awZkzZyShiGoFZoqmUik89dRTAFp8Ox9CGj8AwkWTe6e6RXv4VKOQU202m8hmszJJJJNJ8dbJ39PIkj+nsWXZAnLt4XAY+Xy+oyl0JpNBNpsVBYwxRj7DG5tldmu1msg2tcpFrxpYVjcajboEmRXgnnvukb/vuuuujpgHr/FGlRa+FuHUL9cInn76aRw5cgQej+eKxaXY+PfAgQMdZXnpqVMjDqBD4QK0OteT4mC9FKpICHLXrKZIuocJR5ws2KnJ6/Uil8tJzREAIpnz+/0YGxtDIpHAa6+9JtUemZ1Kb5FZq6QBOG7SSZxQWJ7YFZtaPVi9UnPrrC/k0H9sVr58OXBGfRGspB6JLkHLm0TfLDpJidsxUKazMFn8iR3nWYY1n88L3TE6Oorp6WnEYjGk02mEw2EEg0FpLzc9PY1AIIDR0VEUCgWEw+GOBsfk9wuFgiS35HI50b/rsgblchmRSES0+aSUWBOH1M2JEyeWbAnnsDRuuOEGWNtqSzgzMyMy1WGohbNZ4Iy6Qwf2798vxo7GmV6Xx+O5rJMRvWddJoA8NYuDzc3NIRwOI5lMSvMJVu+bnZ1FJBKRqn21Wg0jIyMIBAJIJBLI5/MdhaLojQeDQczNzaFSqeD8+fOIx+OYnZ2VtHWqb6hP15JFYEFDrQPGOmDqDPrq8KEPfUj+PnDggJRacFg/DKtRd5GZVeCuu+5CPp9HLpeToKNuf9dstvppptNppNNpkTmyoBU1yUzuIZ8ai8WQSCSkRIHH45E6Ltdddx0SiQSMMZibm0OhUMD8/LxILoGFCo0AJAHJ6/VifHwcAKRGTDwel+JebMpB6oVjJM3DzFVSRUywYpDXYe04depUR5kAh/XBNStpNMY8j1Yt4Clr7Y+0X1u0mWq7OfW3AHyn/fGvWms/2I+BbySeffbZy17btWuXVNhjNb5CoSABzmKxCGOMKFKYvKNT+Vk/5sKFC6Iu8Xg8kmSUyWTEyI6MjMDv9yOfzwtFQyUNsCDBTKfTqFQqUkYgFAohHo+jUChIjRi/3y8ri0wm06HwoQJIB/SYjLQZb/hBxfPPP489e/Zs9DCuGXSX9BgmLMdTfwHAO7te+xKAH7HW/iiA76LVTJV41Vr7tvbP0Bn0K4EGVQe9mL6vg6K65yk95FwuJ553IpHoqMzIAlDUONNr9/v94nnH43GMjY3h+uuvF1362bNnhf8GFgz01q1bpX43pZTj4+NSLoCTjvbEGQ8gRZPJZFAsFvGLv/iL63Z9hx39zpZ26MQ166lba7/S9sD1a+veTHWQsWvXLimAxaxPqkSAhcJj3cXIulPGmfBDNYrP50MkEpFsz0gkgkqlIg0wAAjvXqlUpFl0MBjExMSE6KBJswCQBJiRkRGpAc/js/kzsFCjntw6YwFaFklP38FhM2IzGuzloBec+u0A/ov6/2ZjzP8yxvy/xph/udSHjDE7jTEvGWNemp6e7sEwNg7PPPOM6Mr1so4yRAZQy+Uycrkccrkc5ubmpGkGywVoTTq14Lr6I4s+WdtqCcfSu/l8Xnh1FhzT3ZJCoRAikQhisZhkoFYqFemBSQ+edc6p1mENeAZ0KW9kgJgVIx0cNiOuWU/9SlikmeoFADdaa2eNMT8O4I+NMT9src12f9a2umWfBoBbb7118125LoyMjCAUColOndw3ZYRMYCqXy0gkEtL8gtp2KmWoBadxZZIS0NK3R6NRhEIhZDIZqfwXjUalMz0/R6kkjS8159w3dfNU6mjNtO5BSujVBmMGjUYDf/AHf7D+F9vBoQfYjAZ7OVi1UV+smaq1tgKg0v77ZWPMqwB+EMBLax/q4OLw4cPSco4Noal4YZCRJWu1F01jrRta6B6kzBqlblxz40z7Z3BU12OhAaf2nVUdGRTVMkzy/Ux+oq6d3jgDot0Gv1Qq4S//8i837Jo7OKwFm9ULXw5WZdSXaqZqjJkAMGetbRhjbgHwFgCv9WSkA4g9e/aI1juRSIiRpUEOhUKigCH9wuQeGmHWjFmsBkw+n5fgKEvgstTA2NiY0CPsrsNMUHLorNnOVQInFnL5Xq9XCopxItElDVgGwVrbsdJoNBrYunXrxlx0B4ceYVipw6ty6saYzwD47wB+yBhz1hjzAbSaqcbRaqb6DWMMi2n/KwB/bYz5BoA/APBBa+1cf4a+fnjoocUbe9MAMpGInjXpiWAwKB7x9PQ05ufnpcQq66/oEgDcX6VSkXR/vh6LxTAxMSEt6SqVijSwaDQaKBaLqFQqUjiMafzM/KTahcXHSMPQ8yadAkDeJ9/PQC3P8/Of/zw+85nPrNv1d3DoB9aDUzfGHDfGnGvbyW8YY25bYrt3GmO+Y4x5xRhzeC3HXI765T2LvPzJJbb9QwB/uJYBDSKW6rDk9/vFw6WShFRFuVwWLxpY4NJpyFOpFLZv345qtYparSZePKWDbH+mm0snk0n4/X4p3pXL5dBsNpFKpSTlnBQL0EpGoqoGgBj+iYkJmYw4edTrdTH4AIQe4mqCFA/bsTk4bHasI/3ypLX2w0u9aYzxAvhtAP8GwFkAXzPGfMFa+83VHMyVCVgmDh06hHK5LFUagYUm1aRO+D9T92mcI5EIfvRHfxTRaBRnz55Fo9GQDEIGUXWrPJ1ABEC8/rm5OanPArSMdr1eF/qGbfhyuRx8Ph+y2SwCgQDC4bDQKSzG1d2tnqsFnhdXDfy5lnuNOgwfBoxTfzuAV6y1rwGAMeazAN4FwBn1fuDw4cNIp9PCd+/Zs0dK3jIgSQUJk4sKhYJour1eLy5duoSxsTHMzc1J0JTcOA0yW9XV63WEw2FMT0/LBMHgJjsv6drn1L6XSiVUKhVpfpHP50WpwuYYrKvOiYC8OzXpDNTqzFTSSQ4Ow4YVGPVxY4wWe5xuq/eWi93GmPeiJRj5kLV2vuv97QDOqP/PAviJFey/A86oXwWPPfYYgFaCEY0dk4KoA6d8EVhoVUZZIUsHvPrqq7DWYsuWLdi6dWtHQS1SLeFwWMrtso4790WKhEafvVGDwaBQNuTAaay7tfMMrJLn182wdZNryhlJCznZosMwYgVGfcZae+tSbxpj/gLAYsqBXwfwLICHAdj2799EK7enb3BGfZnobgZx+PBhkRYC6Oh8xKJYuuZLsVjEtm3b5O9IJIJkMonz58+jUChIrRe/34+pqSmEQiFZBZByYVZpPB7H97//fWmyQT277qEKQIw61Tf0/JnNqpOjON5msyn7YPleB4dhRK/UL9ban1nOdsaYjwP4k0XeOgfgBvX/9e3XVgVn1FcJ9pXUVRYBdFAX5NaZCRqLxURNEo/HkUwmMTs7K6n4+XweIyMjHRp1Gl6qZngsJguRK6fWXL+v0/xJFXFFoVvc0TPXPUm5Ivn0pz+9rtfVwWE9sF6cujFmm7X2QvvfXwDwt4ts9jUAbzHG3IyWMX83gF9Z7TGdUV8h7rzzTkna0any2vNldmiz2cTk5CQqlYp44mweXS6XceHCBal9zqqN9XodY2NjqFarUl+d0kav14tKpQJjjBjoYrGIVColNdCDwSDS6bTUgAkEAsKZ625LVNsA6JiUdHB0YmJiIy+1g0NfsU6B0pPGmLehRb+8DuBOADDGXAfgE9ba26y1dWPMbgB/BsAL4Hlr7d+t9oDOqK8QNNwsqUtqg/VSgAVvfW5uDpFIRAysMQbJZFLqu7AqYzAYRDAYlCqJiUQChUJB6J1oNCocNwOyNMqlUgm5XE7qm0ejUZRKpY40f/YZZZ32SqWC+fl5ea1arcrEwgnjP/2n/7SRl9nBoe9YD6Nurf2PS7x+HsBt6v8vAvhiL47pjPoKQe6aHjqDjfTW6/W6vM7CWeSw2YyCQUsmE1FvzoBnJpMR+qNcLiObzWJ8fBxerxfpdBoej0eMOoOipGLorZMvZH0XNu7gigJYKBXMOjBMXKIH7+AwzBggSWNP4Yz6ChGJRJDL5QBAJIn8nUgkLpMpku544403xBDTc2aGJg3t3NwcQqEQ0uk0Jicncf3110upXRploJVERAUNeXCuDliGAIDsl9swxd/v90tXJC2xrFQq2Lp1K06ePLlh19fBYT2gq6kOG5xRXyHYJYjUCP/XGZlaykhP2hgjKhdWSaQ6hXJE7qtWq4mXz1VBNptFNBoVvfr8/DwCgYAEPNlOD4CUyNUt9ngslhIgXcSJwRiDF198cYOvroPD+sF56tcQHn/8cVQqFSkPsHv3bpnVmbpPygKAeNyUB166dAmJRALBYFCClDfffLN0jvf5fNiyZYvUQWcxLkogb7nlFly8eFGOw0QmUjKpVEpWC1SrkLcn188VhB4flTCUSLIsMNUzDg7XEob1nndGfREcOnSo4/+tW7fi/vvvBwA8/PDDktJPbTobSrC2ebValRrnrHt+4cIFTE5OIpVKoV6vY2ZmBpVKBfF4HCMjI1K7xRiDS5cuiUHnpEAentw9k45orMmb636ilDlysmFhLwZt6b0XCoWNuMwODhsKZ9SvYdCgA61mGEzVp4dMb5d8ORtZBINBxGIxnD17FnNzcwgGgxgZGenI6DTGiASRVEwgEJBeoUwWokRSe+6aZ9d10qmh1/JEJiglk0kZHxsdHz9+HHNzm76YpoPDsjFgtV96iqsadWPM82g1w5iy1v5I+7XjAHYAYB+6I21JDowx9wH4AIAGgLuttX/Wh3FvKMrlsrR+o4dMaWAsFhMvm1mg1113HVKpFABgamoKIyMjiEQiHeVs6TVzkojFYlJGgAab7+laMAzEUucOQFQ4NO6sr+7z+XD8+PHLzmex1xwchh3DatSX06P0BQDvXOT1J621b2v/0KC/Fa1sqB9uf+aZdlnJocHu3btx6tQpKarFTEwacKbo5/N5TE9Pi3FnQNPr9aJYLEqglJ53d9YoDbHO9KRKhdtxNcCxsPIi2+qx7C4LjrGptIODAzpUaFf62Wy4qlG31n4FwHLX5u8C8FlrbcVa+z0Ar6BVVnLowEJf1Hrzy6fR1TVYaJTJaVer1Y7epdVqVYw+s091vXP+T+6e9drJtzNhiIXGKFdkNioRDAbX+So5OAwu1qNJxkZgOZ76UthtjPlrY8zzxpiR9muLlZDcvtiHjTE7jTEvGWNemp6eXmyTgQdpDvb31BmlbPqczWbFcAPoSA6i6oWyxXg8LrLEcDjcoSWnOoZeP3XpzHBl8hBXDOFwGNFoVDz5cDjcERtwcLiWsVyDfi0Z9WcB/ACAtwG4gFY5yRXBWnvaWnurtfbWzVpj5Omnn8bp06c7lCksGUBKhMoTluoFWgab3nYoFMLY2BiAhUqPDMKSI2cQ1ePxSJ10at8Z9GR7vLm5OZRKJZk8WCqAk4qDg0MLw2rUV6V+sdZe4t9d5SR7WkJyUHHvvfciHA7jwQcfBACMjo4in88jGAxKzRXdZAJo0TOkP2jc6emzmYXOBmUQ1u/3i0adPDmThxhI1ZUV+RoVMuFwGOVyGadOnVrvy+TgMNDYjAZ7OViVp26M2ab+1eUkvwDg3caYYLuM5FsA/M+1DXHwcPLkSVhrceTIEdx3332w1iIcDiORSEjWp5Y6ApBAKNUuNMLsgBQOh6W/KAt3kUqpVqsSMGW1R04GpVJJ6JvuIl709DnBODg4LGBYA6XLkTR+BsA70GrpdBbAMQDvWKycpLX274wxv49Wb706gF+z1g6lRalWqxIsPXLkiNRHp9c9NzfX0YdUyxW1tpwBUpYKoHFmNyTt6VtrJfhaq9Vke53d6vV6RXIJtDj8eDy+3pfHwWGgsVmpleXgqkbdWvueRV7+5BW2/w0Av7HaAZ08eVIkf41GA8eOHVvtrvqKUqkkf584cQIf/vCHJZnnueeew+zsrHjKADr07OTHqVjRlIsu7WutRT6fl2ApuXu+z2Qk/s8Jg9UYydu7Al0ODpdjWI26GYQT27Jli52amtroYSwL9913H9LpNJ599tmrbrtr167L2uBpHD9+HPV6Hfl8vkMdU61W5X9SOGwKHQ6HJemIAVT2MKWHz88xMen555/vwZk7OAwGjDEv2yv0DF0OEomEffvbl6e2/vKXv7zm460nBqJMwGaq300juhxcyaADnZmcd911V0cNdJ1hSiPO/qH08Nk4g6BunSWAScO8733vQyqVwkc+8pEVn6+Dw7BiEBzafmAtOvWeYdu2bVffaEDw2GOPXdVYrwbUrjOoyexRYKESI3l3cvKcXEhV0ahTAcN90ds/ceJEz8ft4LBZ4SSNDn2FLvBFrrzZbEp99kAgAL/fL3XcafDJvdN7BxY6GVF502w2USgUOrJLHRyuZXDl228YY34PwA+1/00BSFtr37bIdq8DyKFVM6u+FrrHGfUBQjQale5ILMWrk4fInfv9ftmW5QJYB0ZXZSSv7vf78bGPfWyDz87BYbCwHl64tfY/8G9jzG8CyFxh85+y1s6s9ZjOqG8AnnrqKczOziKTyaBWqyEWi0m5ABp0XXY3Go3CWotisYhGo4FwOIxKpSIeeqPRQC6Xk9hEuVzGJz7xiQ0+SweHwcZ6UiumxZX+MoB/3e9jOaO+AThz5gxyuZzo05mA5Pf7USqVkMvlEAgEJHuUxcDK5TISiURHIhOw0IuUoHfv4OCwNFZg1MeNMS+p/09ba0+v8HD/EsAla+3fLzUcAH9ujLEAPraK/QucUd8A0IhT4UJ+T9d00Tp46s0rlUpHMhJ16eTcG42GZJ46ODgsjRUGQWeuxHEbY/4CwNZF3vp1a+3n23+/B8BnrnCMn7TWnjPGbAHwJWPMt9sVclcMZ9Q3ADS6uj6MrutCXbqWM7KGerlcFqqG0Fp23YDawcFhafSKfrHW/syV3jfG+AD8IoAfv8I+zrV/TxljPodWyfJVGXX39G8AdDCTqhVKFn0+H2KxGOLxOEKhUEcHIxb2IvVCZQw/T9rFGXUHh6tjHWu//AyAb1trzy72pjEmaoyJ828AP4uFelorhvPU1xm7d+/uaKxByoVlAPg6293l8/mOyozBYLCj8xE9e3odgUCgw4t3cHBYHOsYKH03uqgXY8x1AD5hrb0NwCSAz7VX8D4An7bW/ulqD+aM+jqDBbc8Hg9On16Ihezbt0+Co+TcdV0XAFI3plqtSuMLUi70Kqicuf322115AAeHJbCeiUXW2vct8tp5ALe1/34NwI/16njOqK8znnrqqUVfD4fDyGRaElZy7vV6XSiXSqWyIiO9d+/etQ/WwWGIsRmzRZeD5ZTefR7AzwOYstb+SPu1RbOkjDE3AfgWgO+03/uqtfaDvR70MOLRRx+Vv++8886O9/x+/4pronu9Xhw6dAiFQkHomOeee27tA3VwGBIMq1FfTkTtBQDv1C9Ya/+DtfZt7XTXPwTwR+rtV/meM+irQyKRwOjoqGjYrbUdVM1ycOrUKSkMxi5Ju3fv7tOIHRw2H67ZJhnW2q+0PfDLsJ5ZUtcSnnjiCfn7wIEDqw58njx5Evv374fP50OpVMLFixd7NUQHh02NzVqsazlYK6e+WJbUzcaY/wUgC+B+a+1/XeyDxpidAHYCwI033rjGYQwv1tpblOV5/X6/yzR1cFBwRn1xdGdJXQBwo7V21hjz4wD+2Bjzw9babPcH22mwpwHg1ltvHc6rOwC4/vrrYa1FJpPZVHXrHRz6DWfUu7BYlpS1tgKg0v77ZWPMqwB+EMBLi+7Eoe/Yt2/fRg/BwWEg4Yz65bgsS8oYMwFgzlrbMMbcAuAtAF5b4xgdHBwceo5hNepXVb8YYz4D4L8D+CFjzFljzAfab12WJQXgXwH4a2PMNwD8AYAPWmvnejheBwcHhzVDN2+/FtUv71ni9fct8tofoiVxdHBwcBhoDKun7jJKHRwcrkk4o+7g4OAwRHBG3cHBwWFI4JKPHBwcHIYMzqg7ODg4DBE2o7JlOXBG3cHB4ZqE89QdHBwchgSOU3dwcHAYMgyrUXcdih0cHK5J0Fu/2s9aYIz598aYvzPGNI0xt3a9d58x5hVjzHeMMf/7Ep+/2RjzP9rb/Z4xJnC1Yzqj7uDgcE1incoE/C1ahQ+/ol80xrwVrVIrP4xWE6JnjDGL1cZ+HMCT1to3A5gH8IFFtumAM+oODg7XHJbrpa/VU7fWfsta+51F3noXgM9aayvW2u8BeAXA2/UG7SZE/xqtOloA8DsA/t3Vjuk4dQcHh2sSKzDY48YYXT78dLsfxFqwHcBX1f9n269pjKHV/7l+hW0ugzPqDg4O1yRWYNRnrLW3LvWmMeYvAGxd5K1ft9Z+fjVjWwucUXdwcLgm0Sv1i7X2Z1bxsXMAblD/X99+TWMWQMoY42t764ttcxkcp+7g4HBNYj049SvgCwDebYwJGmNuRquh0P/sGp8F8JcAfqn90q8CuKrn74y6g4PDNYf1apJhjPkFY8xZAP8cwP9ljPmz9vH/DsDvA/gmgD8F8GvW2kb7M180xlzX3sUhAAeMMa+gxbF/8qrHHAQBvjFmGkABwMxGj2UFGMfmGi/gxrwe2GzjBTbfmN9krZ1Yyw78fr8dHR1d1rZTU1MvX4lTHzQMBKdurZ0wxry0mS7cZhsv4Ma8Hths4wU255h7gUFwaPuBgTDqDg4ODusNZ9QdHBwchgSuoNf6YK1i/vXGZhsv4Ma8Hths4wU255jXjGE16gMRKHVwcHBYT/h8PhuPx5e1bTqddoFSBwcHh0HHsDq0zqg7ODhccxhmTn3Dk4+MMe9s1xN+xRhzeKPHsxSMMa8bY/7GGPMNFvcxxowaY75kjPn79u+RDR7j88aYKWPM36rXFh2jaeG32tf9r40x/2RAxnvcGHOufZ2/YYy5Tb131frT6zDmG4wxf2mM+Wa7Tvbe9usDeZ2vMN6Bvs7rgQ3OKO0bNtSot+sH/zaAnwPwVgDvadcZHlT8lLX2bYpfOwzgy9batwD4cvv/jcQLaNVm1lhqjD+HVmryWwDsBPDsOo1R4wVcPl6gVT/6be2fLwIrqj/db9QBfMha+1YA/wzAr7XHNqjXeanxAoN9nfsOZ9T7g7cDeMVa+5q1tgrgs2jVGd4seBdaNY6BZdY67iestV8BMNf18lJjfBeAF20LX0WrcNC2dRloG0uMdylctf70esBae8Fa+/X23zkA30KrHOpAXucrjHcpDMR1Xg+sU5OMdcdGG/XtAM6o/5dVL3iDYAH8uTHmZWPMzvZrk9baC+2/LwKY3JihXRFLjXGQr/3uNlXxvKK0Bm68xpibAPxjAP8Dm+A6d40X2CTXuR9YryYZG4GNNuqbCT9prf0naC2nf80Y86/0m+2KagN9B2yGMaJFT/wAgLcBuADgNzd0NEvAGBMD8IcA9llrs/q9QbzOi4x3U1znfsIZ9f5gOTWFBwLW2nPt31MAPofWkvQSl9Lt31MbN8IlsdQYB/LaW2svWWsb1tomgI9jYek/MOM1xvjRMpD/2Vr7R+2XB/Y6LzbezXCd+w1n1PuDrwF4i2l1zA6gFaD5wgaP6TIYY6LGmDj/BvCzaDWU/QJaNY6BZdY63gAsNcYvAHhvW53xzwBkFH2wYejim38BresMLKP+9DqNz6BV/vRb1tpT6q2BvM5LjXfQr/N6YFiN+oq4pX78ALgNwHcBvIpW+6cNH9MiY7wFwF+1f/6O40SrvvGXAfw9gL8AMLrB4/wMWkvpGlpc6AeWGiMAg5by6FUAfwPg1gEZ7++2x/PXaBmYbWr7X2+P9zsAfm6DrvFPokWt/DWAb7R/bhvU63yF8Q70de73jzHG+v3+Zf0AeGmjx7uic2t/iQ4ODg7XDIwx1utdnlKz0Wi4MgEODg4Og45hdWidUXdwcLgm4Yy6g4ODwxDBGXUHBweH4cGfodWbdTnYTP1bXaDUwcHBYZiw0Tp1BwcHB4cewhl1BwcHhyGCM+oODg4OQwRn1B0cHByGCM6oOzg4OAwR/n/P9rjGIWHYnQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(fvol,interpolation=None,cmap='gray')\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "bd2e97cf-0518-410c-b5f1-bc667f854747",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "date_start = '2018-07-22 09:00'\n",
    "date_end = '2018-07-22 15:00'\n",
    "date_t = dt.timedelta(minutes=10)\n",
    "# date_t = dt.timedelta(minutes=30)\n",
    "date_s = dt.datetime.strptime(date_start, \"%Y-%m-%d %H:%M\")\n",
    "date_e = dt.datetime.strptime(date_end, \"%Y-%m-%d %H:%M\")\n",
    "date_time_now = date_s\n",
    "label = []\n",
    "while date_time_now <= date_e:\n",
    "    # print(date_time_now)\n",
    "    start_time = T.time()\n",
    "    YYYY,MM,DD,HH,MIN,date= Time_split(date_time_now)\n",
    "    label.append(HH+MIN)\n",
    "    date_time_now = date_time_now + date_t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "c1f873a3-11f1-47df-a416-b142ffd5e381",
   "metadata": {},
   "outputs": [],
   "source": [
    "del(label[16])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "23e2f3a7-faac-4a7c-a518-19913cf68d2e",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['0900',\n",
       " '0910',\n",
       " '0920',\n",
       " '0930',\n",
       " '0940',\n",
       " '0950',\n",
       " '1000',\n",
       " '1010',\n",
       " '1020',\n",
       " '1030',\n",
       " '1040',\n",
       " '1050',\n",
       " '1100',\n",
       " '1110',\n",
       " '1120',\n",
       " '1130',\n",
       " '1150',\n",
       " '1200',\n",
       " '1210',\n",
       " '1220',\n",
       " '1230',\n",
       " '1240',\n",
       " '1250',\n",
       " '1300',\n",
       " '1310',\n",
       " '1320',\n",
       " '1330',\n",
       " '1340',\n",
       " '1350',\n",
       " '1400',\n",
       " '1410',\n",
       " '1420',\n",
       " '1430',\n",
       " '1440',\n",
       " '1450',\n",
       " '1500']"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75c7071e-17f8-43f1-9456-922507e69df5",
   "metadata": {},
   "source": [
    "# Read SGLI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "id": "b9616049-7615-430b-8be4-496ddc588b2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import math\n",
    "import os\n",
    "from osgeo import gdal\n",
    "import xarray as xr\n",
    "from skimage.measure import block_reduce\n",
    "from ftplib import FTP\n",
    "import subprocess\n",
    "import rioxarray\n",
    "import time as T\n",
    "import datetime\n",
    "import ephem\n",
    "import numpy.ma as ma\n",
    "import datetime as dt\n",
    "from joblib import Parallel, delayed\n",
    "\n",
    "def calc_sunpos(dtime,lat,lon):\n",
    "    sun = ephem.Sun()\n",
    "    obs = ephem.Observer()\n",
    "    obs.date = dtime\n",
    "    obs.lat = lat*math.pi/180.0\n",
    "    obs.long = lon*math.pi/180.0\n",
    "    sun.compute(obs)\n",
    "    return np.degrees(sun.az),90.0-np.degrees(sun.alt)\n",
    "\n",
    "\n",
    "class SGLI_L2_Downloader:\n",
    "    \n",
    "    def __init__ (self, User_ID , ProductName, Ver, Date , Tile , Obrit ,ParaVer, Target_path):\n",
    "        self.User_ID = User_ID\n",
    "        self.ProductName = ProductName\n",
    "        self.Ver = Ver\n",
    "        self.Date = Date\n",
    "        self.Tile = Tile\n",
    "        self.Obrit = Obrit\n",
    "        self.ParaVer = ParaVer\n",
    "        self.Target_path = Target_path\n",
    "        \n",
    "        \n",
    "    def path (self):\n",
    "        return '/standard/GCOM-C/GCOM-C.SGLI/L2.LAND.{}/{}/{}/{}/{}'.format(self.ProductName, self.Ver, self.Date[0:4], self.Date[4:6], self.Date[6:8])\n",
    "\n",
    "    def filename (self):\n",
    "        return 'GC1SG1_{}{}01D_T{}_L2SG_{}Q_{}00{}.h5'.format(self.Date, self.Obrit, self.Tile, self.ProductName, self.Ver, self.ParaVer)\n",
    "    \n",
    "    def login (self):\n",
    "        ftp_addr = 'ftp.gportal.jaxa.jp'\n",
    "        f = FTP(ftp_addr)\n",
    "        f.login(self.User_ID,'anonymous')\n",
    "        \n",
    "        return f\n",
    "    \n",
    "    def download (self, f):\n",
    "        remote_filepath = self.path()\n",
    "        f.cwd(remote_filepath)\n",
    "        list=f.nlst()\n",
    "        \n",
    "        bufsize=1024\n",
    "        if self.filename() in list:\n",
    "            data = open ('{}{}'.format(self.Target_path, self.filename()), 'wb')\n",
    "            filename = 'RETR ' + self.filename()\n",
    "            f.retrbinary(filename,data.write,bufsize)\n",
    "            return 'y'\n",
    "        else:\n",
    "            return 'n'\n",
    "    \n",
    "    def logout (self, f):\n",
    "        f.quit()\n",
    "    \n",
    "def reporjection_GEO(filename,Band,Target_path):\n",
    "    \n",
    "    #print(path)\n",
    "    p = subprocess.Popen('/data01/people/liwei/Data/GCOM-C_RSRF/SGLI_geo_map_linux.exe {} -d Geometry_data/{} -o {} -r 0 -s 30'.format(filename,Band,Target_path[:-1]),shell=True)\n",
    "    p.communicate()\n",
    "    # subprocess.check_output('/data01/people/liwei/Data/GCOM-C_RSRF/SGLI_geo_map_linux.exe {} -d Geometry_data/{} -o {} -s 7.5'.format(filename,Band,Target_path[:-1]), shell=True) \n",
    "    \n",
    "def mkdir(path):\n",
    "    folder = os.path.exists(path) \n",
    "    if not folder:\n",
    "        os.makedirs(path)\n",
    "        \n",
    "    \n",
    "def SGLI_2_AHI_GEO(filepath,lat,lon):  \n",
    "    geotiff_da = rioxarray.open_rasterio(filepath)\n",
    "    data = geotiff_da.interp(x=lon,y=lat,method=\"nearest\")\n",
    "    if data != -32768:\n",
    "        data = data * 0.01\n",
    "        return data\n",
    "    else:\n",
    "        return np.nan\n",
    "def SGLI_2_AHI_TIME(filepath,lat,lon):\n",
    "    geotiff_da = rioxarray.open_rasterio(filepath)\n",
    "    data = geotiff_da.interp(x=lon,y=lat,method=\"nearest\")\n",
    "    if data != -32768.:\n",
    "        data = data * 0.001\n",
    "        return data\n",
    "    else:\n",
    "        return np.nan\n",
    "\n",
    "def download_tile(date,save_tile_path,tile):\n",
    "    g = SGLI_L2_Downloader(\n",
    "        'galiwei ' ,\n",
    "        'RSRF', \n",
    "        '3', \n",
    "         date , \n",
    "         tile, \n",
    "        'D' , \n",
    "        '1' ,\n",
    "         save_tile_path\n",
    "        )\n",
    "    f = g.login()\n",
    "    \n",
    "    if os.path.exists(save_tile_path + g.filename()):\n",
    "        return g.filename()\n",
    "    \n",
    "    else :\n",
    "        rc = g.download(f)\n",
    "        if rc == 'n':\n",
    "            g.logout(f)\n",
    "            return 'n'\n",
    "        else:\n",
    "            g.logout(f)\n",
    "            return g.filename()\n",
    "    \n",
    "\n",
    "def Decompression(t,fn):\n",
    "    if os.path.exists('{}.bz2'.format(t + fn)):\n",
    "        p = subprocess.Popen('lbzip2 -d {}.bz2'.format(t + fn),shell=True)\n",
    "        p.communicate()\n",
    "        \n",
    "def ahi_lon_timeoffset(lon): \n",
    "    lon_interval = 15 \n",
    "    UTC_e_lon = lon_interval / 2 \n",
    "    timeoffset = math.ceil((lon - UTC_e_lon) / lon_interval) \n",
    "    return timeoffset \n",
    "        \n",
    "def Time_split(time):\n",
    "    YYYY = time.strftime('%Y')\n",
    "    MM = time.strftime('%m')\n",
    "    DD = time.strftime('%d')\n",
    "    HH = time.strftime('%H')\n",
    "    MIN = time.strftime('%M')\n",
    "    date = YYYY + MM + DD\n",
    "    return YYYY,MM,DD,HH,MIN,date\n",
    "    \n",
    "    \n",
    "def lonlat2tileidx(latlon:tuple)->tuple:\n",
    "    # vertical pixel count in one tile, horizontal pixel count in one tile\n",
    "    lintile, coltile = 1200, 1200\n",
    "    # vertical tile count, horizontal tile count\n",
    "    vtilenum, htilenum = 18, 36\n",
    "    # [deg/pixel]\n",
    "    d = 180.0/lintile/vtilenum\n",
    "    # from S-pole to N-pole\n",
    "    NL = 180.0/d\n",
    "    NP0 = 2*np.round(180.0/d)\n",
    "    res1 = (90 - latlon[0])/d-0.5\n",
    "    V_idx = res1 // lintile\n",
    "    # Y_idx = res1 - V_idx * lintile\n",
    "    # GCOM-C use integer NPi\n",
    "    NPi = round(NP0*np.cos(np.deg2rad(latlon[0])))\n",
    "    res2 = latlon[1]*NPi/360 - 0.5 + NP0/2\n",
    "    H_idx = res2 // coltile\n",
    "    # X_idx = res2 - H_idx * coltile\n",
    "    return str(round(V_idx)).rjust(2,'0')  + str(round(H_idx)).rjust(2,'0') \n",
    "def SGLI_SA(h,_min,lat,lon):\n",
    "\n",
    "    d = datetime.datetime(int(YYYY),int(MM),int(DD),h,_min)\n",
    "    \n",
    "    SAA,SZA = calc_sunpos(d,lat,lon)\n",
    "    return SAA,SZA\n",
    "reporjection_file_path = '/data01/people/liwei/Data/GCOM-C_RSRF/reporjection/'\n",
    "tile_file_path = '/data01/people/liwei/Data/GCOM-C_RSRF/tile/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 396,
   "id": "67379c46-fa30-4535-90d8-c8c011405204",
   "metadata": {},
   "outputs": [],
   "source": [
    "AHI_lat = np.linspace(60,-59.99,12000)[2550:2750][i_idx-r:i_idx+r]\n",
    "AHI_lon = np.linspace(85,204.99,12000)[4700:5000][j_idx-r:j_idx+r]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "id": "dcf5a0c9-be40-4a3c-a02d-8fb1e38466fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "import rioxarray\n",
    "def SGLI_2_AHI_IMAGE(filepath,lat,lon):\n",
    "    geotiff_da = rioxarray.open_rasterio(filepath)\n",
    "    data = geotiff_da.interp(x=lon,y=lat,method=\"nearest\")\n",
    "    data = np.where(data != 65535, data ,np.nan)\n",
    "    data = data * 0.0001\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 397,
   "id": "a7515ff5-f344-49c0-9e92-2317b7e617b3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-21 00:00:00\n",
      "2018-07-22 00:00:00\n",
      "2018-07-23 00:00:00\n",
      "2018-07-24 00:00:00\n",
      "2018-07-25 00:00:00\n",
      "2018-07-26 00:00:00\n",
      "2018-07-27 00:00:00\n"
     ]
    }
   ],
   "source": [
    "SGLI_VZA = []\n",
    "SGLI_SZA = []\n",
    "SGLI_RAA = []\n",
    "SGLI_Ref = []\n",
    "b = 'Rs_PI02'\n",
    "for d in range(21,28):\n",
    "    date_start = '2018-07-{} 00:00'.format(str(d))\n",
    "    date_end = '2018-07-{} 06:00'.format(str(d))\n",
    "    date_t = dt.timedelta(days=1)\n",
    "    # date_t = dt.timedelta(minutes=30)\n",
    "    date_s = dt.datetime.strptime(date_start, \"%Y-%m-%d %H:%M\")\n",
    "    date_e = dt.datetime.strptime(date_end, \"%Y-%m-%d %H:%M\")\n",
    "    date_time_now = date_s\n",
    "    b = 4\n",
    "    \n",
    "    c_SGLI_VZA = []\n",
    "    c_SGLI_SZA = []\n",
    "    c_SGLI_RAA = []\n",
    "    c_SGLI_Ref = []\n",
    "    # time_list = []\n",
    "    while date_time_now <= date_e:\n",
    "        print(date_time_now)\n",
    "        start_time = T.time()\n",
    "        YYYY,MM,DD,HH,MIN,date= Time_split(date_time_now)\n",
    "        for lat_idx in range(2*r):\n",
    "            for lon_idx in range(2*r):\n",
    "\n",
    "                # 获取该经纬度的Tile号\n",
    "                # print(lat_idx,lon_idx) \n",
    "\n",
    "                tile = lonlat2tileidx([AHI_lat[lat_idx],AHI_lon[lon_idx]])\n",
    "\n",
    "                # print(tile) \n",
    "                # SGLI 文件名\n",
    "                SGLI_tile_filename = 'GC1SG1_{}D01D_T{}_L2SG_RSRFQ_3001.h5'.format(date,tile)\n",
    "                SGLI_VZA_filename = 'GC1SG1_{}D01D_T{}_L2SG_RSRFQ_3001_Sensor_zenith_PL.tif'.format(date,tile)\n",
    "                SGLI_VAA_filename = 'GC1SG1_{}D01D_T{}_L2SG_RSRFQ_3001_Sensor_azimuth_PL.tif'.format(date,tile)\n",
    "                SGLI_TIME_filename = 'GC1SG1_{}D01D_T{}_L2SG_RSRFQ_3001_Obs_time_PL.tif'.format(date,tile)\n",
    "                SGLI_REF_filename = 'GC1SG1_{}D01D_T{}_L2SG_RSRFQ_3001_Rs_PI02.tif'.format(date,tile)\n",
    "\n",
    "                # 存在SGLI文件\n",
    "                # print('1')\n",
    "                if os.path.exists(reporjection_file_path + SGLI_VAA_filename) and os.path.exists(reporjection_file_path + SGLI_VZA_filename):\n",
    "                    # print('1')\n",
    "                    # 读取该像素经纬度SGLI的VZA和VAA\n",
    "                    SGLI_VZA_PL = SGLI_2_AHI_GEO(reporjection_file_path + SGLI_VZA_filename,AHI_lat[lat_idx],AHI_lon[lon_idx])\n",
    "                    SGLI_VAA_PL = SGLI_2_AHI_GEO(reporjection_file_path + SGLI_VAA_filename,AHI_lat[lat_idx],AHI_lon[lon_idx])\n",
    "\n",
    "                    if not np.isnan(SGLI_VZA_PL) and not np.isnan(SGLI_VAA_PL):\n",
    "                        if SGLI_VAA_PL < 0:\n",
    "                            SGLI_VAA_PL = SGLI_VAA_PL + 360\n",
    "                    # 计算该像素经纬度SGLI的SZA和SAA\n",
    "\n",
    "                        SGLI_Obs_TIME = SGLI_2_AHI_TIME(reporjection_file_path + SGLI_TIME_filename,AHI_lat[lat_idx],AHI_lon[lon_idx])\n",
    "                        if not np.isnan(SGLI_Obs_TIME):\n",
    "                            # print(SGLI_Obs_TIME)\n",
    "                            SGLI_MIN = int(round(math.modf(SGLI_Obs_TIME)[0],3)*60)\n",
    "                            SGLI_HH = int(math.modf(SGLI_Obs_TIME)[1])\n",
    "                            if SGLI_HH <= 23 and SGLI_MIN >= 0 and SGLI_MIN <= 59:\n",
    "                                SGLI_SAA_PL,SGLI_SZA_PL = SGLI_SA(SGLI_HH,SGLI_MIN,AHI_lat[lat_idx],AHI_lon[lon_idx])\n",
    "                                SGLI_Ref_PL = SGLI_2_AHI_IMAGE(reporjection_file_path + SGLI_REF_filename,AHI_lat[lat_idx],AHI_lon[lon_idx])\n",
    "                                # 读取AHI的SZA、SAA\n",
    "                                SGLI_RAA_PL = abs(SGLI_SAA_PL - SGLI_VAA_PL.data[0])\n",
    "                                c_SGLI_VZA.append(SGLI_VZA_PL.values[0])\n",
    "                                c_SGLI_SZA.append(SGLI_SZA_PL)\n",
    "                                c_SGLI_RAA.append(SGLI_RAA_PL)\n",
    "                                c_SGLI_Ref.append(SGLI_Ref_PL)\n",
    "        SGLI_VZA.append(np.array(c_SGLI_VZA))\n",
    "        SGLI_SZA.append(np.array(c_SGLI_SZA))\n",
    "        SGLI_RAA.append(np.array(c_SGLI_RAA))\n",
    "        SGLI_Ref.append(np.array(c_SGLI_Ref))\n",
    "        date_time_now = date_time_now + date_t                            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 432,
   "id": "b07b71ae-3784-43a3-a54b-ae9457335c86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[41.09, 41.09, 41.09, 41.09, 41.08, 41.08, 41.08, 41.08, 41.07,\n",
       "        41.07],\n",
       "       [41.08, 41.08, 41.08, 41.08, 41.07, 41.07, 41.07, 41.07, 41.06,\n",
       "        41.06],\n",
       "       [41.07, 41.07, 41.07, 41.06, 41.06, 41.06, 41.06, 41.05, 41.05,\n",
       "        41.05],\n",
       "       [41.06, 41.06, 41.06, 41.05, 41.05, 41.05, 41.05, 41.04, 41.04,\n",
       "        41.04],\n",
       "       [41.05, 41.05, 41.05, 41.04, 41.04, 41.04, 41.04, 41.03, 41.03,\n",
       "        41.03],\n",
       "       [41.04, 41.04, 41.03, 41.03, 41.03, 41.03, 41.02, 41.02, 41.02,\n",
       "        41.02],\n",
       "       [41.03, 41.03, 41.02, 41.02, 41.02, 41.02, 41.01, 41.01, 41.01,\n",
       "        41.01],\n",
       "       [41.02, 41.01, 41.01, 41.01, 41.01, 41.01, 41.  , 41.  , 41.  ,\n",
       "        41.  ],\n",
       "       [41.01, 41.  , 41.  , 41.  , 41.  , 40.99, 40.99, 40.99, 40.99,\n",
       "        40.98],\n",
       "       [41.  , 40.99, 40.99, 40.99, 40.99, 40.98, 40.98, 40.98, 40.98,\n",
       "        40.97]])"
      ]
     },
     "execution_count": 432,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "VZA[i-r:i+r,j-r:j+r] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 400,
   "id": "b946fa79-cf61-4b5e-8958-4731baf63462",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def SGLI_Kernel_BRDF(band,day,idx_row,idx_col): \n",
    "    if band == 3:\n",
    "        if SGLI_SZA[day].size != 0:\n",
    "            kvol = Volume_Kernel(640,AHI_lat[idx_row],AHI_lon[idx_col],MJD,np.cos(np.radians(SGLI_SZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_VZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_RAA[day].reshape(2*r,2*r)[idx_row,idx_col])))\n",
    "            kgeo = Geometric_Kernel(640,AHI_lat[idx_row],AHI_lon[idx_col],MJD,np.cos(np.radians(SGLI_SZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_VZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_RAA[day].reshape(2*r,2*r)[idx_row,idx_col])))\n",
    "            return kvol,kgeo\n",
    "        else:\n",
    "            return np.nan,np.nan\n",
    "    elif band == 4:\n",
    "        if SGLI_SZA[day].size != 0:\n",
    "            kvol = Volume_Kernel(860,AHI_lat[idx_row],AHI_lon[idx_col],MJD,np.cos(np.radians(SGLI_SZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_VZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_RAA[day].reshape(2*r,2*r)[idx_row,idx_col])))\n",
    "            kgeo = Geometric_Kernel(860,AHI_lat[idx_row],AHI_lon[idx_col],MJD,np.cos(np.radians(SGLI_SZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_VZA[day].reshape(2*r,2*r)[idx_row,idx_col])),np.cos(np.radians(SGLI_RAA[day].reshape(2*r,2*r)[idx_row,idx_col])))\n",
    "            return kvol,kgeo\n",
    "        else:\n",
    "            return np.nan,np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 401,
   "id": "e38820a8-b143-4969-8612-cb830e763a12",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "band = 4\n",
    "Kresult = Parallel(n_jobs=10)(delayed(SGLI_Kernel_BRDF)(b,d,i,j) for d in range(7) for i in range(2*r) for j in range(2*r))\n",
    "SGLI_kvol = np.array(Kresult)[:,0].reshape(7,2*r,2*r)\n",
    "SGLI_kgeo = np.array(Kresult)[:,1].reshape(7,2*r,2*r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 403,
   "id": "1c193cea-5e7f-447f-8828-1652b9da2db3",
   "metadata": {},
   "outputs": [],
   "source": [
    "i,j = 79,190\n",
    "# X_SGLI = np.array(Ref_valid)[0:36,i-6:i+6,j-6:j+6]\n",
    "Y_SGLI = fiso[i-r:i+r,j-r:j+r] + np.array(SGLI_kvol)[:,:,:]*fvol[i-r:i+r,j-r:j+r] + np.array(SGLI_kgeo)[:,:,:]*fgeo[i-r:i+r,j-r:j+r]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 417,
   "id": "44a38461-7976-474d-9d0a-790ba364edf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "XX = []\n",
    "YY = []\n",
    "for i in range(7):\n",
    "    if SGLI_SZA[i].size != 0:\n",
    "        XX.append(SGLI_Ref[i].reshape(2*r,2*r).tolist())\n",
    "        YY.append(Y_SGLI[i].reshape(2*r,2*r).tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 414,
   "id": "fd00a17d-f82c-49c8-990c-d0ed232dc93b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "cannot reshape array of size 0 into shape (6,6)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn [414], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m XX \u001b[38;5;241m=\u001b[39m \u001b[43mSGLI_Ref\u001b[49m\u001b[43m[\u001b[49m\u001b[43mday\u001b[49m\u001b[43m]\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreshape\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m6\u001b[39;49m\u001b[43m,\u001b[49m\u001b[38;5;241;43m6\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m      2\u001b[0m YY \u001b[38;5;241m=\u001b[39m Y_SGLI[day]\n",
      "\u001b[0;31mValueError\u001b[0m: cannot reshape array of size 0 into shape (6,6)"
     ]
    }
   ],
   "source": [
    "XX = SGLI_Ref[day].reshape(6,6)\n",
    "YY = Y_SGLI[day]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 420,
   "id": "d768fe6b-9497-4107-a8dd-932796cb6a2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.5)"
      ]
     },
     "execution_count": 420,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "day = 6\n",
    "\n",
    "XX = np.array(XX).reshape(-1)\n",
    "YY = np.array(YY).reshape(-1)\n",
    "# XX = SGLI_Ref[day].reshape(-1)\n",
    "# YY = Y_SGLI[day].reshape(-1)\n",
    "nan_mask = ~np.isnan(YY.reshape(-1))\n",
    "XX = XX.reshape(-1)[nan_mask]\n",
    "YY = YY.reshape(-1)[nan_mask]\n",
    "\n",
    "rmse = np.sqrt(mean_squared_error(XX ,YY))\n",
    "p = np.polyfit(XX ,YY,1)    \n",
    "\n",
    "k = round(p[0],2)\n",
    "b = round(p[1],2)\n",
    "rmse = round(rmse.astype('float64'),3)\n",
    "\n",
    "\n",
    "axis_min = 0\n",
    "axis_max = 0.5\n",
    "\n",
    "fig = plt.figure(figsize=(4,4))\n",
    "ax1 = fig.add_subplot(111,aspect='equal')\n",
    "\n",
    "x = np.arange(axis_min,axis_max+1)\n",
    "y = 1 * x\n",
    "\n",
    "xx = np.arange(0,1+0.1,0.05) \n",
    "yy = k * xx + b\n",
    "\n",
    "# Calculate the point density\n",
    "xy = np.vstack([XX,YY])\n",
    "z = gaussian_kde(xy)(xy)\n",
    "\n",
    "# Sort the points by density, so that the densest points are plotted last\n",
    "idx = z.argsort()\n",
    "XX, YY, z = XX[idx], YY[idx], z[idx]\n",
    "\n",
    "ax1.minorticks_on()\n",
    "# x_major_locator = plt.MultipleLocator(5)\n",
    "x_minor_locator = plt.MultipleLocator(0.05)\n",
    "ax1.xaxis.set_minor_locator(x_minor_locator)\n",
    "# ax.xaxis.set_major_locator(x_major_locator)\n",
    "ax1.yaxis.set_minor_locator(x_minor_locator)\n",
    "# ax.yaxis.set_major_locator(x_major_locator)\n",
    "\n",
    "ax1.tick_params(axis=\"y\",which='minor',length=5,direction='in',labelsize=8)\n",
    "ax1.tick_params(axis=\"y\",which='major',length=10,direction='in',labelsize=8)\n",
    "\n",
    "ax1.tick_params(axis=\"x\",which='minor',length=5,direction='in',labelsize=8)\n",
    "ax1.tick_params(axis=\"x\",which='major',length=10,direction='in',labelsize=8)\n",
    "\n",
    "\n",
    "ax1.spines['right'].set_color('none')\n",
    "ax1.spines['top'].set_color('none')\n",
    "\n",
    "\n",
    "im = ax1.scatter((XX*1.002)-0.03456,YY,marker='o', c=z,s=15,cmap='Spectral_r')\n",
    "# im = ax1.scatter(XX,YY,marker='o', c='k',s=15)\n",
    "\n",
    "ax1.set_xticks(np.arange(axis_min,axis_max+0.1,0.1))\n",
    "ax1.set_yticks(np.arange(axis_min+0.1,axis_max+0.1,0.1))\n",
    "\n",
    "ax1.set_xlabel(\"SGLI Ref \",fontsize=10)\n",
    "ax1.set_ylabel(\"BRDF fitting Ref\",fontsize=10)\n",
    "\n",
    "\n",
    "ax1.plot(x,y,color='k',linewidth=2,linestyle='-',alpha=0.6)\n",
    "ax1.plot(xx,yy,color='r',linewidth=2,linestyle='-') \n",
    "\n",
    "\n",
    "text_x = axis_min + (axis_max - axis_min) * 0.07\n",
    "text_y = axis_max - (axis_max - axis_min) * 0.22\n",
    "\n",
    "ax1.text(text_x,text_y,s = 'N = {}\\nRMSE = {}\\ny = {} + {}x'.format(len(XX),rmse,b,k),fontsize=8)\n",
    "# cax = add_right_cax(ax1, pad=0.06, width=0.03)\n",
    "# cb = fig.colorbar(im,cax=cax)\n",
    "# cb.ax.set_xlabel('Count',rotation=360)\n",
    "\n",
    "ax1.set_xlim(axis_min,axis_max)\n",
    "ax1.set_ylim(axis_min,axis_max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "870a9cab-a455-4098-84d7-6b0a773c9bf8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "id": "bc20fe12-4ae3-43ab-abf0-81849ddfe1de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f808e72d160>"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(SGLI_Ref[3].reshape(12,12),cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "id": "4de5749a-1a92-4d57-bd96-283217fdf6ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f808e7124c0>"
      ]
     },
     "execution_count": 313,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(Y_SGLI[3].reshape(12,12),cmap='gray')"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}