gp1_wrf_MANUSCRIPT_energy_fluxes_cape_skewt_AMAZON_REGION.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# import stuff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:22:27.703401Z",
     "start_time": "2018-08-16T14:22:27.685632Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy\n",
    "import xarray\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as mp\n",
    "import datetime\n",
    "import glob\n",
    "import netCDF4\n",
    "import pandas\n",
    "import wrf\n",
    "import string\n",
    "import seaborn\n",
    "import matplotlib.colors\n",
    "import cmocean\n",
    "import matplotlib.cm\n",
    "\n",
    "import metpy\n",
    "import metpy.plots\n",
    "import metpy.calc\n",
    "import metpy.units\n",
    "\n",
    "import cartopy\n",
    "\n",
    "DEGREE_SIGN = u'\\u00B0'\n",
    "mp.rcParams.update({'mathtext.default': 'regular'})\n",
    "\n",
    "% matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# set up wet and dry regions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:22:27.793160Z",
     "start_time": "2018-08-16T14:22:27.707845Z"
    }
   },
   "outputs": [],
   "source": [
    "dry_lon_lo, dry_lon_hi, dry_lat_lo, dry_lat_hi = -60,-50,-10,10\n",
    "wet_lon_lo, wet_lon_hi, wet_lat_lo, wet_lat_hi = -80,-75,-10,10\n",
    "\n",
    "dry_lon_lo_degE = 360+dry_lon_lo\n",
    "dry_lon_hi_degE = 360+dry_lon_hi\n",
    "\n",
    "wet_lon_lo_degE = 360+wet_lon_lo\n",
    "wet_lon_hi_degE = 360+wet_lon_hi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:22:27.892725Z",
     "start_time": "2018-08-16T14:22:27.795552Z"
    }
   },
   "outputs": [],
   "source": [
    "lon_lo = dry_lon_lo\n",
    "lon_hi = dry_lon_hi\n",
    "lat_lo = dry_lat_lo\n",
    "lat_hi = dry_lat_hi\n",
    "\n",
    "# lon_lo = wet_lon_lo\n",
    "# lon_hi = wet_lon_hi\n",
    "# lat_lo = wet_lat_lo\n",
    "# lat_hi = wet_lat_hi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# open control data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:22:29.266692Z",
     "start_time": "2018-08-16T14:22:27.894847Z"
    }
   },
   "outputs": [],
   "source": [
    "root_dir = '/beegfs/DATA/pritchard/blangenb/WRF_ARCHIVE/CHEYENNE/FMAASO/400/ens_mean_wrfout_files'\n",
    "file_list = sorted(glob.glob(root_dir+'/'+'wrfout*.nc'))\n",
    "ds_list = [netCDF4.Dataset(file) for file in file_list]"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T20:25:50.798686Z",
     "start_time": "2018-06-27T20:16:43.318449Z"
    }
   },
   "source": [
    "td_400 = wrf.getvar(ds_list, 'td', units='K', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_td_400.npy', td_400)\n",
    "\n",
    "cape3d_400 = wrf.getvar(ds_list, 'cape_3d', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_cape3d_400.npy', cape3d_400)\n",
    "\n",
    "cape2d_400 = wrf.getvar(ds_list, 'cape_2d', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_cape2d_400.npy', cape2d_400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:22:37.275147Z",
     "start_time": "2018-08-16T14:22:29.274769Z"
    }
   },
   "outputs": [],
   "source": [
    "cont_mfds = xarray.open_mfdataset(root_dir + '/' + 'wrfout*.nc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:08.938071Z",
     "start_time": "2018-08-16T14:22:37.283578Z"
    }
   },
   "outputs": [],
   "source": [
    "ptot_400 = numpy.load('DATA_ptot_400.npy')\n",
    "geopot_400 = numpy.load('DATA_geopot_400.npy')\n",
    "#cape3d_400 = numpy.load('DATA_cape3d_400.npy')\n",
    "#cape2d_400 = numpy.load('DATA_cape2d_400.npy')\n",
    "td_400 = numpy.load('DATA_td_400.npy')\n",
    "tk_400 = numpy.load('DATA_tk_400.npy')"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T23:47:44.809283Z",
     "start_time": "2018-06-29T23:47:44.784593Z"
    }
   },
   "source": [
    "cldfra_400 = cont_mfds['CLDFRA']\n",
    "\n",
    "fira_400 = cont_mfds['FIRA']\n",
    "fsa_400 = cont_mfds['FSA']\n",
    "glw_400 = cont_mfds['GLW']\n",
    "swdown_400 = cont_mfds['SWDOWN']\n",
    "\n",
    "alwdnb_400 = cont_mfds['ACLWDNB']\n",
    "aswdnb_400 = cont_mfds['ACSWDNB']\n",
    "alwupb_400 = cont_mfds['ACLWUPB']\n",
    "aswupb_400 = cont_mfds['ACSWUPB']\n",
    "\n",
    "alwdnbc_400 = cont_mfds['ACLWDNBC']\n",
    "aswdnbc_400 = cont_mfds['ACSWDNBC']\n",
    "alwupbc_400 = cont_mfds['ACLWUPBC']\n",
    "aswupbc_400 = cont_mfds['ACSWUPBC']\n",
    "\n",
    "alwdnt_400 = cont_mfds['ACLWDNT']\n",
    "aswdnt_400 = cont_mfds['ACSWDNT']\n",
    "alwupt_400 = cont_mfds['ACLWUPT']\n",
    "aswupt_400 = cont_mfds['ACSWUPT']\n",
    "\n",
    "alwdntc_400 = cont_mfds['ACLWDNTC']\n",
    "aswdntc_400 = cont_mfds['ACSWDNTC']\n",
    "alwuptc_400 = cont_mfds['ACLWUPTC']\n",
    "aswuptc_400 = cont_mfds['ACSWUPTC']"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T23:47:45.139322Z",
     "start_time": "2018-06-29T23:47:44.817437Z"
    }
   },
   "source": [
    "qvapor_400 = cont_mfds['QVAPOR']"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T23:47:53.150529Z",
     "start_time": "2018-06-29T23:47:45.147773Z"
    }
   },
   "source": [
    "net_atm_400_sw = (aswupb_400-aswupt_400)+(aswdnt_400-aswdnb_400)\n",
    "net_atm_400_lw = (alwupb_400-alwupt_400)+(alwdnt_400-alwdnb_400)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T23:48:02.910898Z",
     "start_time": "2018-06-29T23:47:55.383388Z"
    }
   },
   "source": [
    "net_atm_400_swc = (aswupbc_400-aswuptc_400)+(aswdntc_400-aswdnbc_400)\n",
    "net_atm_400_lwc = (alwupbc_400-alwuptc_400)+(alwdntc_400-alwdnbc_400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:09.015330Z",
     "start_time": "2018-08-16T14:25:08.945990Z"
    }
   },
   "outputs": [],
   "source": [
    "hgt = cont_mfds['HGT'].sel(Time=0).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:42.163145Z",
     "start_time": "2018-08-16T14:25:09.023349Z"
    }
   },
   "outputs": [],
   "source": [
    "time = cont_mfds['Times'].values\n",
    "xlat = cont_mfds['XLAT'].values[0,:,:]\n",
    "xlong = cont_mfds['XLONG'].values[0,:,:]\n",
    "landmask = cont_mfds['LANDMASK'].values[0,:,:]\n",
    "landmask_midpoints = landmask[1:,1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:42.178450Z",
     "start_time": "2018-08-16T14:25:42.171109Z"
    }
   },
   "outputs": [],
   "source": [
    "landmask_indices = numpy.copy(landmask)\n",
    "landmask_indices[landmask==1]=1\n",
    "landmask_indices[landmask!=1]=0\n",
    "landmask_indices = numpy.array(landmask_indices, dtype=bool)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:42.273319Z",
     "start_time": "2018-08-16T14:25:42.186038Z"
    }
   },
   "outputs": [],
   "source": [
    "lon_indices = (xlong>=lon_lo)&(xlong<=lon_hi)\n",
    "lat_indices = (xlat>=lat_lo)&(xlat<=lat_hi)\n",
    "latlon_indices = (lon_indices)&(lat_indices)&(landmask_indices)\n",
    "latlon_indices_high_andes = (lon_indices)&(lat_indices)&(landmask_indices)&(hgt>1500.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:42.693613Z",
     "start_time": "2018-08-16T14:25:42.281848Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x2b99b22b14e0>"
      ]
     },
     "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": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mp.contourf(latlon_indices);mp.colorbar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# open phys data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:43.806436Z",
     "start_time": "2018-08-16T14:25:42.701581Z"
    }
   },
   "outputs": [],
   "source": [
    "root_dir = '/beegfs/DATA/pritchard/blangenb/WRF_ARCHIVE/CHEYENNE/FMAASO/1000/ens_mean_wrfout_files'\n",
    "file_list = sorted(glob.glob(root_dir+'/'+'wrfout*.nc'))\n",
    "ds_list = [netCDF4.Dataset(file) for file in file_list]"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T20:36:34.057718Z",
     "start_time": "2018-06-27T20:25:57.865123Z"
    }
   },
   "source": [
    "td_1000 = wrf.getvar(ds_list, 'td', units='K', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_td_1000.npy', td_1000)\n",
    "\n",
    "cape3d_1000 = wrf.getvar(ds_list, 'cape_3d', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_cape3d_1000.npy', cape3d_1000)\n",
    "\n",
    "cape2d_1000 = wrf.getvar(ds_list, 'cape_2d', timeidx=wrf.ALL_TIMES).values\n",
    "numpy.save('DATA_cape2d_1000.npy', cape2d_1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:25:51.364634Z",
     "start_time": "2018-08-16T14:25:43.814768Z"
    }
   },
   "outputs": [],
   "source": [
    "phys_mfds = xarray.open_mfdataset(root_dir + '/' + 'wrfout*.nc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:07.952186Z",
     "start_time": "2018-08-16T14:25:51.373201Z"
    }
   },
   "outputs": [],
   "source": [
    "ptot_1000 = numpy.load('DATA_ptot_1000.npy')\n",
    "#cape3d_1000 = numpy.load('DATA_cape3d_1000.npy')\n",
    "#cape2d_1000 = numpy.load('DATA_cape2d_1000.npy')\n",
    "td_1000 = numpy.load('DATA_td_1000.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:46.798471Z",
     "start_time": "2018-08-16T14:27:07.960234Z"
    }
   },
   "outputs": [],
   "source": [
    "tk_1000 = numpy.load('DATA_tk_1000.npy')"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-07-01T05:59:48.565539Z",
     "start_time": "2018-07-01T05:59:48.541320Z"
    }
   },
   "source": [
    "cldfra_1000 = phys_mfds['CLDFRA']\n",
    "\n",
    "fira_1000 = phys_mfds['FIRA']\n",
    "fsa_1000 = phys_mfds['FSA']\n",
    "glw_1000 = phys_mfds['GLW']\n",
    "swdown_1000 = phys_mfds['SWDOWN']\n",
    "\n",
    "alwdnb_1000 = phys_mfds['ACLWDNB']\n",
    "aswdnb_1000 = phys_mfds['ACSWDNB']\n",
    "alwupb_1000 = phys_mfds['ACLWUPB']\n",
    "aswupb_1000 = phys_mfds['ACSWUPB']\n",
    "\n",
    "alwdnbc_1000 = phys_mfds['ACLWDNBC']\n",
    "aswdnbc_1000 = phys_mfds['ACSWDNBC']\n",
    "alwupbc_1000 = phys_mfds['ACLWUPBC']\n",
    "aswupbc_1000 = phys_mfds['ACSWUPBC']\n",
    "\n",
    "alwdnt_1000 = phys_mfds['ACLWDNT']\n",
    "aswdnt_1000 = phys_mfds['ACSWDNT']\n",
    "alwupt_1000 = phys_mfds['ACLWUPT']\n",
    "aswupt_1000 = phys_mfds['ACSWUPT']\n",
    "\n",
    "alwdntc_1000 = phys_mfds['ACLWDNTC']\n",
    "aswdntc_1000 = phys_mfds['ACSWDNTC']\n",
    "alwuptc_1000 = phys_mfds['ACLWUPTC']\n",
    "aswuptc_1000 = phys_mfds['ACSWUPTC']"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-07-01T05:59:50.581848Z",
     "start_time": "2018-07-01T05:59:50.575783Z"
    }
   },
   "source": [
    "qvapor_1000 = phys_mfds['QVAPOR']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## calculate from perspective of atmosphere"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T02:03:21.743237Z",
     "start_time": "2018-06-27T02:03:03.566249Z"
    }
   },
   "source": [
    "net_atm_1000_sw = (aswupb_1000-aswupt_1000)+(aswdnt_1000-aswdnb_1000)\n",
    "net_atm_1000_lw = (alwupb_1000-alwupt_1000)+(alwdnt_1000-alwdnb_1000)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T02:03:30.849201Z",
     "start_time": "2018-06-27T02:03:21.748664Z"
    }
   },
   "source": [
    "net_atm_1000_swc = (aswupbc_1000-aswuptc_1000)+(aswdntc_1000-aswdnbc_1000)\n",
    "net_atm_1000_lwc = (alwupbc_1000-alwuptc_1000)+(alwdntc_1000-alwdnbc_1000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# calculate area mean net bottom and top diffs"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T02:07:54.965118Z",
     "start_time": "2018-06-27T02:03:37.800079Z"
    }
   },
   "source": [
    "net_atm_diffs_1000_sw = numpy.zeros((net_atm_1000_sw.shape))*numpy.nan\n",
    "net_atm_diffs_1000_sw[:-1,:,:] = numpy.diff(net_atm_1000_sw, axis=0)\n",
    "\n",
    "net_atm_diffs_1000_lw = numpy.zeros((net_atm_1000_lw.shape))*numpy.nan\n",
    "net_atm_diffs_1000_lw[:-1,:,:] = numpy.diff(net_atm_1000_lw, axis=0)\n",
    "\n",
    "net_atm_diffs_400_sw = numpy.zeros((net_atm_400_sw.shape))*numpy.nan\n",
    "net_atm_diffs_400_sw[:-1,:,:] = numpy.diff(net_atm_400_sw, axis=0)\n",
    "\n",
    "net_atm_diffs_400_lw = numpy.zeros((net_atm_400_lw.shape))*numpy.nan\n",
    "net_atm_diffs_400_lw[:-1,:,:] = numpy.diff(net_atm_400_lw, axis=0)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T02:09:42.990640Z",
     "start_time": "2018-06-27T02:07:56.717945Z"
    }
   },
   "source": [
    "net_atm_diffs_1000_swc = numpy.zeros((net_atm_1000_swc.shape))*numpy.nan\n",
    "net_atm_diffs_1000_swc[:-1,:,:] = numpy.diff(net_atm_1000_swc, axis=0)\n",
    "\n",
    "net_atm_diffs_1000_lwc = numpy.zeros((net_atm_1000_lwc.shape))*numpy.nan\n",
    "net_atm_diffs_1000_lwc[:-1,:,:] = numpy.diff(net_atm_1000_lwc, axis=0)\n",
    "\n",
    "net_atm_diffs_400_swc = numpy.zeros((net_atm_400_swc.shape))*numpy.nan\n",
    "net_atm_diffs_400_swc[:-1,:,:] = numpy.diff(net_atm_400_swc, axis=0)\n",
    "\n",
    "net_atm_diffs_400_lwc = numpy.zeros((net_atm_400_lwc.shape))*numpy.nan\n",
    "net_atm_diffs_400_lwc[:-1,:,:] = numpy.diff(net_atm_400_lwc, axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# pressure averages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:46.868893Z",
     "start_time": "2018-08-16T14:27:46.806690Z"
    }
   },
   "outputs": [],
   "source": [
    "ptot_latlon_subset_timemean = numpy.mean(ptot_400[:,:,latlon_indices], axis=2).mean(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# td diffs"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T06:14:29.036936Z",
     "start_time": "2018-06-29T06:12:25.002153Z"
    }
   },
   "source": [
    "td_400_latlon_subset = td_400[:,:,latlon_indices]\n",
    "td_1000_latlon_subset = td_1000[:,:,latlon_indices]\n",
    "\n",
    "td_400_latlon_mean = numpy.nanmean(td_400_latlon_subset, axis=2)\n",
    "td_1000_latlon_mean = numpy.nanmean(td_1000_latlon_subset, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:47.116807Z",
     "start_time": "2018-08-16T14:27:46.879144Z"
    }
   },
   "outputs": [],
   "source": [
    "td_400_latlon_subset = td_400[:,:,latlon_indices]\n",
    "td_1000_latlon_subset = td_1000[:,:,latlon_indices]\n",
    "\n",
    "td_400_latlon_mean = numpy.nanmean(td_400_latlon_subset, axis=2)\n",
    "td_1000_latlon_mean = numpy.nanmean(td_1000_latlon_subset, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:47.267947Z",
     "start_time": "2018-08-16T14:27:47.125091Z"
    }
   },
   "outputs": [],
   "source": [
    "td_400_latlon_alltime = numpy.nanmean(td_400_latlon_subset, axis=(0,2))\n",
    "td_1000_latlon_alltime = numpy.nanmean(td_400_latlon_subset, axis=(0,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# tk diffs"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T06:15:57.941316Z",
     "start_time": "2018-06-29T06:14:41.695003Z"
    }
   },
   "source": [
    "tk_400_latlon_subset = tk_400[:,:,latlon_indices]\n",
    "tk_1000_latlon_subset = tk_1000[:,:,latlon_indices]\n",
    "\n",
    "tk_400_latlon_mean = numpy.nanmean(tk_400_latlon_subset, axis=2)\n",
    "tk_1000_latlon_mean = numpy.nanmean(tk_1000_latlon_subset, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:47.483040Z",
     "start_time": "2018-08-16T14:27:47.276051Z"
    }
   },
   "outputs": [],
   "source": [
    "tk_400_latlon_subset = tk_400[:,:,latlon_indices]\n",
    "tk_1000_latlon_subset = tk_1000[:,:,latlon_indices]\n",
    "\n",
    "tk_400_latlon_mean = numpy.nanmean(tk_400_latlon_subset, axis=2)\n",
    "tk_1000_latlon_mean = numpy.nanmean(tk_1000_latlon_subset, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:47.635251Z",
     "start_time": "2018-08-16T14:27:47.491406Z"
    }
   },
   "outputs": [],
   "source": [
    "tk_400_latlon_alltime = numpy.nanmean(tk_400_latlon_subset, axis=(0,2))\n",
    "tk_1000_latlon_alltime = numpy.nanmean(tk_1000_latlon_subset, axis=(0,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# qvapor diffs"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-29T06:19:58.070992Z",
     "start_time": "2018-06-29T06:16:10.604997Z"
    }
   },
   "source": [
    "qvapor_400_latlon_subset = qvapor_400.values[:,:,latlon_indices]\n",
    "qvapor_1000_latlon_subset = qvapor_1000.values[:,:,latlon_indices]\n",
    "\n",
    "qvapor_400_latlon_mean = numpy.nanmean(qvapor_400_latlon_subset, axis=(2))\n",
    "qvapor_1000_latlon_mean = numpy.nanmean(qvapor_1000_latlon_subset, axis=(2))\n",
    "\n",
    "qvapor_diffs_latlon_mean = qvapor_1000_latlon_mean-qvapor_400_latlon_mean"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-07-01T06:01:54.808524Z",
     "start_time": "2018-07-01T06:01:24.443012Z"
    }
   },
   "source": [
    "qvapor_400_latlon_subset = qvapor_400.values[:,:,latlon_indices_high_andes]\n",
    "qvapor_1000_latlon_subset = qvapor_1000.values[:,:,latlon_indices_high_andes]\n",
    "\n",
    "qvapor_400_latlon_mean = numpy.nanmean(qvapor_400_latlon_subset, axis=(2))\n",
    "qvapor_1000_latlon_mean = numpy.nanmean(qvapor_1000_latlon_subset, axis=(2))\n",
    "\n",
    "qvapor_diffs_latlon_mean = qvapor_1000_latlon_mean-qvapor_400_latlon_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# cape diffs"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-07-01T06:05:18.824065Z",
     "start_time": "2018-07-01T06:05:18.743317Z"
    }
   },
   "source": [
    "cape3d_cape_400_latlon_subset = cape3d_400[0,:,:,:,:][:,:,latlon_indices_high_andes]\n",
    "cape3d_cape_1000_latlon_subset = cape3d_1000[0,:,:,:,:][:,:,latlon_indices_high_andes]\n",
    "\n",
    "cape3d_cape_diffs_latlon_mean = numpy.nanmean((cape3d_cape_1000_latlon_subset-cape3d_cape_400_latlon_subset), axis=2)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-28T01:51:35.833827Z",
     "start_time": "2018-06-28T01:51:35.639060Z"
    }
   },
   "source": [
    "cape2d_diffs = cape2d_1000-cape2d_400\n",
    "\n",
    "cape2d_mcape_diffs = cape2d_diffs[0,:,:,:]\n",
    "cape2d_mcin_diffs = cape2d_diffs[1,:,:,:]\n",
    "cape2d_lcl_diffs = cape2d_diffs[2,:,:,:]\n",
    "cape2d_lfc_diffs = cape2d_diffs[3,:,:,:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot skew T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:48.090284Z",
     "start_time": "2018-08-16T14:27:47.643261Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Quantity(1309.189707336896, 'joule / kilogram')>,\n",
       " <Quantity(-32.064096602061085, 'joule / kilogram')>)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_latlon_alltime[:]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_latlon_alltime[:]-273.15)*metpy.units.units.degC)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# cape calculations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:48.865074Z",
     "start_time": "2018-08-16T14:27:48.098231Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(<Quantity(1868.337738514075, 'joule / kilogram')>, <Quantity(-9.686478570210037, 'joule / kilogram')>)\n",
      "(<Quantity(1474.6045808203353, 'joule / kilogram')>, <Quantity(-4.586800208100992, 'joule / kilogram')>)\n"
     ]
    }
   ],
   "source": [
    "ptot_lo = 100\n",
    "ptot_hi = 1013.25\n",
    "hour = 19\n",
    "\n",
    "ptot_indices = (ptot_latlon_subset_timemean/100.<=ptot_hi)&(ptot_latlon_subset_timemean/100.>=ptot_lo)\n",
    "\n",
    "cape_400 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "cape_1000 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "print(cape_400)\n",
    "print(cape_1000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3-hour afternoon mean"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-28T22:48:05.917221Z",
     "start_time": "2018-06-28T22:48:05.471278Z"
    }
   },
   "source": [
    "ptot_lo = 300\n",
    "ptot_hi = 1013.25\n",
    "hour = 19\n",
    "\n",
    "tk_400_afternoon = numpy.nanmean(tk_400_latlon_mean[(hour-1):(hour+2),:], axis=0)\n",
    "td_400_afternoon = numpy.nanmean(td_400_latlon_mean[(hour-1):(hour+2),:], axis=0)\n",
    "\n",
    "ptot_indices = (ptot_latlon_subset_timemean/100.<=ptot_hi)&(ptot_latlon_subset_timemean/100.>=ptot_lo)\n",
    "\n",
    "cape_400 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_afternoon[ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_afternoon[ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "cape_1000 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "print(cape_400)\n",
    "print(cape_1000)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-28T03:25:25.377744Z",
     "start_time": "2018-06-28T03:25:23.942102Z"
    }
   },
   "source": [
    "ptot_lo = 300\n",
    "ptot_hi = 1013.25\n",
    "\n",
    "ptot_indices = (ptot_latlon_subset_timemean/100.<=ptot_hi)&(ptot_latlon_subset_timemean/100.>=ptot_lo)\n",
    "\n",
    "cape_400 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_latlon_alltime[ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_latlon_alltime[ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "cape_1000 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_1000_latlon_alltime[ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_1000_latlon_alltime[ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "print(cape_400)\n",
    "print(cape_1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:48.880765Z",
     "start_time": "2018-08-16T14:27:48.873049Z"
    }
   },
   "outputs": [],
   "source": [
    "color_list = [\\\n",
    "matplotlib.cm.tab20c(4/20), \\\n",
    "matplotlib.cm.tab20c(5.5/20), \\\n",
    "matplotlib.cm.tab20c(0/20), \\\n",
    "matplotlib.cm.tab20c(1.5/20) ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:52:11.309407Z",
     "start_time": "2018-08-16T14:52:06.130110Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 320.4x306 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fontsize=12\n",
    "hour = 19\n",
    "\n",
    "skewt_plot = metpy.plots.SkewT()\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_1000_latlon_mean[hour,:]-273.15, lw=3, label='phys. T$_{env}$', c=color_list[0])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_400_latlon_mean[hour,:]-273.15, lw=3, label='cont. T$_{env}$', ls='--', c=color_list[1])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_1000_latlon_mean[hour,:]-273.15, lw=3, label='phys. T$_d$', c=color_list[2])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_400_latlon_mean[hour,:]-273.15, lw=3, label='cont. T$_d$', ls='--', c=color_list[3])\n",
    "\n",
    "pprofile_400 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_400_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_400_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "pprofile_1000 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_1000_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_1000_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_1000, label='phys. profile', c='0.1', lw=3, ls='-')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_400, label='cont. profile', c='0.5', lw=3, ls='--')\n",
    "\n",
    "\n",
    "# fill in control and phys CAPE\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_1000_latlon_mean[hour,:]-273.15,\\\n",
    "                      numpy.array(pprofile_1000),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.666,\\\n",
    "                      label='phys. CAPE')\n",
    "\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_400_latlon_mean[hour,:]-273.15,\\\n",
    "                      numpy.array(pprofile_400),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.333,\\\n",
    "                      label='cont. CAPE')\n",
    "\n",
    "skewt_plot.ax.set_ylim(1013.25,600)\n",
    "skewt_plot.ax.set_xlim(10,30)\n",
    "\n",
    "fig = mp.gcf()\n",
    "fig.set_size_inches(4.45,4.25)\n",
    "fig.tight_layout()\n",
    "\n",
    "skewt_plot.ax.tick_params(labelsize=fontsize)\n",
    "skewt_plot.ax.set_ylabel('hPa', fontsize=fontsize)\n",
    "skewt_plot.ax.set_xlabel('$\\degree$C', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.plot_dry_adiabats(color='0.5',linewidth=1,linestyle='-')\n",
    "skewt_plot.plot_moist_adiabats(color='0.5',linestyle=':')\n",
    "skewt_plot.plot_mixing_lines(color='0.5',linestyle='--',lw=1)\n",
    "\n",
    "handles, labels = skewt_plot.ax.get_legend_handles_labels()\n",
    "skewt_plot.ax.legend(handles, labels, fontsize=fontsize, scatterpoints=1, \\\n",
    "                     bbox_to_anchor=(0.6,1.0), \\\n",
    "                     bbox_transform=skewt_plot.ax.transAxes, loc='upper left', \\\n",
    "                     borderaxespad=0, labelspacing=0.25, fancybox=False, \\\n",
    "                     edgecolor='0', framealpha=0, borderpad=0.25, \\\n",
    "                     columnspacing=0.5, handletextpad=0.5, markerscale=1)\n",
    "\n",
    "skewt_plot.ax.text(s='skew-T diagram at 15:00 on day 1',\\\n",
    "                  x=0,y=1.02,\\\n",
    "                  transform=skewt_plot.ax.transAxes,\\\n",
    "                  ha='left',va='bottom', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.ax.annotate(\"\",xy=(11.6,700), \\\n",
    "                       xycoords='data',\\\n",
    "                       xytext=(13.6, 700), \\\n",
    "                       textcoords='data',\\\n",
    "                       arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=10))\n",
    "\n",
    "skewt_plot.ax.annotate(\"\",xy=(10.9,700), \\\n",
    "                       xycoords='data',\\\n",
    "                       xytext=(8.9, 700), \\\n",
    "                       textcoords='data',\\\n",
    "                       arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=10))\n",
    "\n",
    "ptot_lo = 300\n",
    "ptot_hi = 1013.25\n",
    "\n",
    "ptot_indices = (ptot_latlon_subset_timemean/100.<=ptot_hi)&(ptot_latlon_subset_timemean/100.>=ptot_lo)\n",
    "\n",
    "cape_400 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "cape_1000 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_1000_latlon_mean[hour,ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "CAPE_CIN_text = 'CAPE$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CAPE$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[1]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[1]))+' J kg$^{\\,-1}$'\n",
    "skewt_plot.ax.text(s=CAPE_CIN_text,x=.02,y=0.02,ha='left',va='bottom',\\\n",
    "                   transform=skewt_plot.ax.transAxes,fontsize=fontsize,linespacing=1)\n",
    "\n",
    "fig.savefig('figures/manuscript_both_cape_skewT_hour'+str(hour-4)+'_zoom_amazon_region.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:54:00.860823Z",
     "start_time": "2018-08-16T14:53:54.135565Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk8AAAHSCAYAAAD11y9uAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXl4lOW5/z/3TPZAICxhCUsgQNg3oSIKRREUENcKKNSlWKW1px7rqbXtqUv3Y2uXXzdqVcSiqICsKqUUVFzYQZZsZF/IQvY9sz2/P2bmzYSEZCYzySzM57q4mHnmXe43yTzv/T7fexGlFEGCBAkSJEiQIEGcQ+dtA4IECRIkSJAgQfyJoPMUJEiQIEGCBAniAkHnKUiQIEGCBAkSxAWCzlOQIEGCBAkSJIgLBJ2nIEGCBAkSJEgQFwg6T0GCBAkSJEiQIC7gE86TiBwSkTm21yIivxORahG5KCLrHLYbJSKfikijiBwRkQneszpIkCBBggQJcjXiVedJRG4RkfXADQ7D3wBuBiYCtwA/F5EZts/eAD4GBgFbgc09aG6QIEGCBAkSJIjXV55uAEKAOoexNcCLSqlCpdRZrA7SShEZAcwAfqqUqgFeAhJEZHxPGx0kSJAgQYIEuXrxqvOklPqJUuoRoNxheCpwyuH9eWAMMA1IUUo12/a1AKm2z4IECRIkSJAgQXqEEG8b0A4xQLXD+1rb2OXjjp+1QUQeBR4FCAkJuWb06NEYDAaam5sBCAsLIzw83MOm9xw1KgJ9iABgURBpaiBEJz1uh6WsDOrq0IXoiIgKo7nvQJRO7/5xLRZ0Om8vjHoGewskkZ7//ZhMJurr6+ndu7fHfp6B9LuBwLoepRRKqYC6HvDOd8dd0tPTy5RSA71tR5DuwRedpyog0uF9FFDZzrjjZ21QSr0MvAwQFhamVq1axcSJE0lNTdW2Wb58OTNnzmyzr8lk4sSJE1xzzTWEhPjej+jcJTM78yrR2+bHsFojdw1vYtSoUT1qR/F//sOxRx+Fvn0BKLnzTqbfdx+zZs3q8jGVUnz66adMmzaNmJh2/WK/oqmpicOHDzNv3jz0evedSlepqKjg3LlzzJs3zyM3oKKiIoqKitr93vgjZ86cITY2luHDh3vbFLexWCwcOnSIWbNmER0d7W1z3Ka+vp7jx48zb948v3QGRSTX2zYE6T588S8yFZjs8H4i8CWQBkwQER2A7f8k4IwzB33llVcoLS0lISFBG9uzZw8pKSltts3Ly+Ozzz4jN9c3//b/dERIKY7AoqC2Uc8MdaHVdfUExtpazj77rPY+/q67mHD33YwbN86t4+bn59O3b9+AcJwAkpOTGT9+vFccJ4B+/foxf/58RIT6+vpWDw+uYjabSUtLY+LEiR600HvU1NRQU1PDsGHDvG2KR8jNzSUuLi4gHCeA8+fPM2nSJL90nIIEPr74V/km8D8i0k9ErgVWA1uUUhlAOvA9EYkEfgKcU0oVOnvgX//611RVVREbGwtYVzm2bdtGdnZ2q+1Gjx7Ngw8+SGJioocuyXMcyoW9mToOZfRiy8m+jDGUMmXSxB5f1i558QdIrTVULax/fyb/+Mdcf/31xMTEoJTizJkzGAwGl45pNBrJzMxk/PjAyAGoqKigubmZQYMGedsUAI4ePcq//vUv6urqOt+4HTIzM4mPjyciIsLDlvU8SinOnTvHpEmT/FISuhyDwUBOTg5jx471tike4dKlS4gIAwYM8LYpQYK0iy86T/8ATgO5WMsR/LdS6oLtsweAe4EKYCHwoDMH7NWrl/b6N7/5DSNHjqR///6A9Wn67bff5uLFi632sX9+6dIl9u3bh9lsduOSPIPRDM9/0vJ+bpyJydE1PT7BVH24lXjdRRYsH0dCUn+mPPccYTaHFKC0tJT9+/dz7tw5l4574cIFRo0aRWhoqKdN7nGUUpw/f57Jkyf7zM15/vz5rFy5Uvs+2ONJnKGxsZGLFy/65ANFVyguLiYyMlJ7kPJ3UlNTGTt2rE+GGbiKxWIhOTmZSZMmeduUIA6IyPO20kLdeY5XRcRk+6dExOzw/oHuPLer+ITzpJRKUEodtr02K6UeV0r1VkoNV0q95bBdulLqWqVUpFJqvlLKKV0tISGByZMn24/B9773PSZNmkTv3r0B61Pbm2++SXl5eZt9CwsLyc7OpqGhwQNX6h6bzkJGhfV1dKhicdSpHp9gTLXVhB58AxEhJFTP8GmjGLJkSattBg0axH333ceMGTOucJS21NXVUVZWxsiRIz1tslewy4/2vzFfQK/XM3jwYAAyMjLYtWsXRqPRqX3t8mMgSCh2+XHChMCosWuXH+Pj471tikfIyckhLi6OqKgob5sSpIdRSq1VSoUopUKw1nTU3iul3vC2fY74/0zoBDqdjr/+9a+afNLU1MS6deuYOXOmJkE0NDSwceNGKioqWu07ffp0HnroIe0m6MrTuicxmhVvpxgI1VsAWD2mgqRhsT0+wVRt/hvRUdb4HZPRQsTa55F2bqhDhgxBRGhoaGDLli2UlZV1eFx7fIOvrNK4gz/Ij/X19TQ0NDj191xRUYHRaPQZ+dFdMjMzGTZsWEDJj760wukOBoOB3NzcgJEf/Q0RWSAiX4rI70WkVkROiMhUh00GiMgBEWkSkc9FZKCI3CoiBQ7H6CUizSKSKCI3ikiy7Vj/sdVrdMe+SBH5h4iUi0iWiKy0jSeISLGI/EhECkUkR0RuF5HeNltnOBxjh4j8jzt2gG9m23ULxcXFbNy4kQceeIDS0lLq6upYt24df/vb3/jiiy8wGo00NDRQXl5Ov379Wu1rL2lw+PBhysvLWbJkSY8/gZ8uMXPj+BoWKCitCWW2+SxjxtzQ+Y4eRpUXa6+rI+PpP6bjp/fGxkaqq6u1EhHtUVpail6v16RSfyc9Pd3n5cdp06YxZcoUdDodZrOZ+vr6doP07fLj9OnTA+LmbJcf58+f721TPEJRURFRUVH0tWW8+jspKSkBIz92ld2JiT32hL48M7O9L/VU4DWgH/AE8LaI2JO4bgOWAMeAncBjwItALxGZppT6EmuHkFSlVKaI/AdYBxwEfgH8Fljhhsm/AcKABGA6sFNEjgEWrJ1HBgNjgUeAPyqlRonIv4BlwCkRibDZ9103bACukpUno9FIdHQ0U6ZM4e2339aco9raWh5//HHmzJlDREQEK1eu7PCJR6fTee0GcrrMBIAIhIuJSeO9NME4Xn9I585B//79efjhhzVJ4fJgZYvFQkpKSsBkcNXV1VFeXu4X8qP9AeDgwYO89dZbNDU1tdkmPz+f2NhYn5If3SHQ5Mf09PSAkR+rq6upra0NGPnRj6kA/p9Sygj8DogHRts+26eUOqiUqgMOAAlKKQPwAVYHBdv/7zkc7xasjs0PgW921Six3nwfBr6vlKpVSh0CdgHLHTb7sVKqwTZurz+y1cG2RcAZpVReV+2w4/8ziBM0NTVpE0xSUhJvv/229qRWXV3Nf/3Xf3Hrrbd2ulT8la98RVt1am5uxmKxdLvtdorqW2JTwoxNDB06tMfO3RoH58lJCdOepp+Xl8crr7zSKrsxOzubQYMGBUx8gz/KjzNmzNAeIByxy49JSUlessyz2OVHe9yXv5ORkcGwYcP8utivnUCUH/2YAmXT822dPMqBONtnRQ7bmWlRr7YDS22vl9DiPC217XvS9m+BG3YNxFrbMc8mxTUB92N1zACalVK1ttdNgL0+zC5gmogMAG4H3nHDBo2rYm308mrikyZN4p133mHFihVUV1dTVVXFmjVr2Lp1a6sbRW1tLRaLhT59+mhjIoLJZGLLli0MGjSIRYsW9cg1GDARZns9bUgv700wjk/sLsZ/DRo0iGnTpmmOX3NzM3l5ecybN8+TFnoNf5Uf+/fvr9lcVlZGaWkpEydOJD09ndGjR/u0/OgsjvJjINDY2EhRUVFAyY/R0dEBJT92lStIaT2JVjFWRMKxSmH2eI0rTfofAq+LyE1Ak1LqrIjEAGOUUqtFRI+17NCrWOW+rlABGIH+Sql6m32JWHvjXl5AW0MpVS0iB7CuPi0Fnuvi+VtxVaw8tfdkNnnyZDZv3qzFeVRUVLBixQouXLBWRaipqWHjxo1s3LiRmpqaVvuGhISQmJjI6NGj2xy3O6huVkSFW0slKAULRvmIhOKi8xQeHs6NN95IeHg4ZrOZPXv2BEx8Q6CkVx87doxPPvmEiooKysvLGTHCrfhOnyEvLy/g5McJEyYEjPwYSNmP1dXVXa6l5iPEisgjtvigZ4EMpVRWRzvYZLyDWGW+bfZh4E0RWYTV1wgDLnXVKKWUCdgB/MgWOH4N8BlW564ztgA/xnotFzvb2Bn8/5vnBtOmTePNN9/U6t6UlZVpDpS9dEFlZWW7DtR1112n1bwpKyvr1iy8wxdN2NvW1Tfr6R/lxV+bOJ6769d88uRJPvnkE5+on+UJsrOzGTJkCJGRV3wA8gsWL17MypUruXDhQsDEoRmNRrKysgJOfgyU7MeMjAxGjBgRUPKjnz9EZQLzscp1C3E+wHs7MA14F8AmoT0KrMfal/Zb2Goz2uo5vdoF276NtbNICVYn7Ye2IPXO2Ik1yPzdLpyzXa5q5wlg5syZbNq0SYu5KS0tZdWqVUyePFl7qquoqOCNN96gtra2zf4VFRVs2rSJY8eOdZuNqRUt8U4h3lZaHReUu+gwKqVobm5m3bp1fhFY3Rl2+XHMmDHeNsVt9Ho9BoMBvV5PdnY2u3btwmQyedsstwhE+dFet87fscuPPd2Xs7sIEPnRpJR6QCkVrZSao5RKBVBKPa+UWmffSCn1a6XUQw77JWPNsjvhsM1mpVSiUipKKXWNUuqobXytUmptR0YopRYopV6/bKxMKfU1pVSMrT7kBtt4jlIqwmG7YqWUOLyvUkqFKaX+0pUfSHtc9c4TwOzZs/nnP/+prRoUFxfzve99jwULFmgOVHl5ORs3bmyzHBsbG8v8+fOZMmVKt9l3qanFeRoS5V3nyRgSTUlBDcX5NTRL11ZZLl68SO/evbVl+vLyct577z0aGxs9aWqPkZKSQlJSktf613kSe/bjpEmT0Ov1hISE+PV12bMfA0l+7NevX6uuCf7M+fPnmThxYlB+9HNEpDfWTLgN3ralp/D/v1gPMWfOHF5//XUt4+jixYt8//vf56abburQgRIRZs6cSWRkJEop0tLSPC7h6fQtT/6T+3v36bkhcjBHD+Zw7KMcqvQDXd7fZDKRnp7eqoBkdXW11gfO36iqqqK+vp4hQ4Z42xSP4Cg/zpw5k6VLlyIiWr0ufyOQ+tfZ5Ud3m2/7CuXl5ZjNZuLi4jrf2A8IJPmxCxzHKtn93duG9BRB58mBG264gQ0bNmh//Pn5+ZoDZZ98y8rKeOONN6ivr2+zf1paGnv27CEnJ8djNp3OLScq3OqMGc0we4iXVwEcbkKqC6Ua2ptgRo8ezcMPP6wtdbf3s/VFfLF/nTs0NzeTn5/fSn60X9fevXvZsmWLX8WolZSUEBoa6nfZj1ciEOVHP48N0ggU+VEp9ZFSyuXWCEqpJFvrNP97wuoiQefpMubPn88rr7xCWJi1MEBubi4/+MEPWLhwoXYjuXTpUrsOVFJSEnfddRcJCQkesUUpxYFzl8i8FEZtk44mQwhheu/epNtrxeIsDQ0NlJSUtDvB2KWhL7/8kg0bNrRpk+OLXLx4kV69erUqZeHPpKSkMG7cuHZlunnz5rFgwQK/kfACsfhqRUVFQMmP/fv3D8qPQfyW4G+6HW666Sb+/ve/a0942dnZ/OhHP+Lmm2/WHKjS0lI+/vjjVvuJCKNHj0ZEqK6u5tixY25JeLm5uRSZ+rMvJYZNR/sRYfKBm7QbK0/nz5/vNL06ISGBKVOm+HzAZXvyoz/Tmfw4YMAAbUUqOzub5OTknjTPZbKysgIi+9FOUH70XQJNfgziHEHn6QosXryY9evXazWIMjIy+PGPf6w5UImJiR0WyDx37hxHjhzpcr0Po9FIdnY2BaYWyWHGYO9PnOHGGsZNHcS4qYPobal0er+ysjIsFkunE0yfPn346le/ik6no7GxkbS0NHdN7hYyMjIYOXJkQMQ3uCo/nj59mhMnTvRohX1XaG5upqCgICCyH6FFfry856a/kpaWRmJiYlB+vAoQkddF5JlOtvmqiJhs/8wiohze/6enbHWVoPPUAbfeeit/+ctfNKkiPT2dZ599lttvv51Vq1Z1+OWfO3cua9as6XJRvrS0NEaNTuRsacuvaLoPdJUIN1SRNG0QSdMG0dvinLytlCI5Odnl9Opjx47xwQcf+Fygsl1+9JQ8620KCwtdkh+XL1/OPffcg06nw2Kx+JwT1ZH86G8EmvxYW1tLZWUlw4cP73xjPyA3Nzeg5EdvoJT6WCkVopQKAdYC2nul1EJv23clgs5TJ9x222386U9/0qSmlJQUnn766XZrPjkG04qIJj2dO3eOL774wulz2icYQ/Rwam0tkgZEQrwPFEduvTLhnCSZm5vLgAEDiI6Odulc119/PStWrPC5mCJn5Ed/wWQyceHCBZfSq0NCQoiKikIpxf79+9mzZ4/POFBVVVU0NDQETPZjVlYWQ4cODQj50bGAZKDIj9nZ2QEjP9oRkdtEJFVEakRkq4hE2cZvEZGzIlItIntEZLht/Hlb0cs9IlIlIh+JSH8ReRp4APiF7bU7Nt0nIuk2m/5uq36O7VzPisgJEam0faYTkbdE5I8O+98pIufcseFy/L8vRg9wxx13YDKZeOKJJ7Rl2vvvv5+3335bu7F/9tlnpKSksGbNmjYNVgsLC7U+eZ3dcB0nmJ2FBm5KMlBaG8Lw6FBEfODX5TjpOeE7GQwGsrOzueGGG1w+lV6v1zqs5+bmkpyczOLFi726ouCs/OgvXLhwgZEjR2oJEq4gIvTr14/m5mafcCTt352pU6cGxM25qamJgoKCgOn9WFJSQlhYWMDIj6mpqd0nPz5/1/M434PtHzy//dHL9n8Z+GYH+7zA89ufv3xQRBKAjcBiIA1rU90nRWQT8DZwD9ayBD8ANgP2if1+rH3jVgF7ge8opV4QkYlYC2e+6OS1tEFEZgG/xtpw+BLwlu38L9g2WQ3civWOdMr2+RbgN8ATtm081hDYjvdnPD/hnnvu4aWXXtLenzlzhtWrV1NbW8unn37K/v37KSws5M0332xTr2jRokXceeedmszRESUlJYSHh9OvXz/y640kDWpm3ph6xsb5SJduF9uzpKWlMWbMGLcnGHvDWqPR2PnG3USgVXduaGigtLTULflx1qxZXH/99YC12v7lbYx6ksLCQmJiYrR+lf5OoBVfTU1NDSj5saqqKmDkRwdWAbuUUids/eoeB44C9wG7lVIHlFI1wPPADBGxB5O8r5Q6aNtnPw7NhT3AWuDPSqlkpdQl4BfAnQ6f/0kpla2UygG+tJ37Q2CQiCSJiA64jaDz5D1WrlzJiy+2ONCnTp1izZo1rZ5yCwoK2jhQOp2OkJAQTCYT27dvv2IrF7PZ3GqCMagWRyEp1gdWneCylaeOnaeamhqqqqoYNmyY26e95pprWL16NREREVgsFq9UI8/JyWHgwIEuy4++iifTq5VSvP/+++zcubNb+zxeCbv8GEjZj42NjQwe7AOBjh4gMzMzKD/6ByMArQmwUipFKfVvYCSQ4TBuxNqvzr4EX+ZwjCY8q2qNxCr9NYlIE9aVLcel/zbnVko1Ae8Dy4DrgEKlVLoHbQrKdq6yevVqjEYjP/7xjwE4fvw4L730Ek899RT79+8HrMU133rrLVavXt1KDtHpdERERFxxAsnKyiI+Pp6IiAhqmhVR4dYYKqXg2iE+8qty0nnqjgKS9szHjz76iOzsbNasWdNj2W4Gg4GcnJyAkVAuXbqEUoqBA12vEt8eIsKtt96K2Wz2yg3FHfnR1whE+bGwsDBgvjuO6kC3YZXUnndj/0exNuV1lXJA6zgtItcBiVjlsmEO432AAUBel210nlLgCaXU32znjsHq5HXGVqzNiAfj4VUnCK48dYmHHnqIn/70p9r7I0eO8Ic//IGbb75ZG8vLy+PNN9/EYGiR23Q6HUuXLtVkH8fVE/sEM3r0aAAOF5nQ2ebN+mY9A6N85FflZMB4cXEx4eHhxMbGetyEpKQkJk2a1KNlAuzyo92B82csFgvJyckeT68eOHCgtlJy+vRpUlNTPXr8K+EJ+dGXCMqPvsvl6kAA8h6wQkSmiEh/4PdABLANuFdE5opIL6yO3QdKqapOjmfC6mS5w7vAYyIyXERigdeBh5zY7wNgJnCv7RgexUfuyP7H2rVrefbZZ7X3n3/+OX/+859ZuLAlszIvL4+33nqrlQNlf5KsqKjgtdde49w5awJAcnJyqwkmtaJFstMrH7phS+d/MvYGmd01wcTHxzNnzhwAKisrKSgo6Jbz2KmpqaG6utoj8qMvkJubS1xcXLfJjxaLhbS0tG7p89ge586dC5jqzoEmP1ZWVtLU1BQw8qM9+/HypKBAQSl1CvgRVskrHTgMvKaU+hL4LvBPrKtQE4BvO3HIfcC3ReRJez2nLtj0AdYg9i+AHKAOJ4LplVINwL+BMqVUVmfbu4oP3ZX9j8ceewyTycQvf/lLAD755BP0ej2PPfYYBw4cAKw3qs2bN3P//fe3Cpru06cP48ePZ/jw4VRWVtLc3NxqgilrMhFlU/cGR/nQr8kJ2S4zM1OTH7ubjz/+mJKSEtauXdstq0KBFt/QE/KjTqfjnnvuQSmFiGAwGAgNDe2Wn9+lS5cAPCY/epsLFy6QkJAQMPLj+fPng/Kjn6GUegV4pZ3xf2J1ni4ff/6y9792eP0urVd9OpyklVKvY11Zunz891hXwS4fX3DZ+1sve7+io/O5g/8/qnmZxx9/nO9///va+4MHD/LKK69w4403amM5OTls2bKl1VO4Xq9n4cKFxMTEcPbsWQYOHNhqghF9i4M+ZYAPVeLVdZxt19jYyMWLFzX5sbu55ZZbuPPOO7tNTisuLiYyMrJb5EdvkJqa2iPyY0hICKGhoVgsFnbs2MGHH37o8XN0l/zoLerr6wNKfiwoKKBPnz4BIz9erg4EuboJOk8e4L//+7958skntff79+9n48aNLFiwALA6Sl/5ylfaffoqKCigsrKSnTt3avJTXo2FyDBrSQOTGWYP9p0vqzmsN7kXyslNL6fW1DbwPSUlhfHjx/fYBBMZGcmgQdb4xjNnzvDZZ595TCqyy4+uFJD0ZWpqaqipqelR+VFEGDVqVLc4BN0tP/Y09jYfgbBKYzKZyMjIICkpydumeIT21IEgVzc+pAf5N0899RRGo5E///nPAOzduxe9Xs+aNWuIj48nMTGxzT5Go5GMjAwWL17M8OHDtYKQR4pa4p0aDCGEh/jOZGroFceZw4UAxA+ZzUiHzyoqKmhubtacmZ5EKUVxcTF1dXWaXOQuPSk/djfekh9FhNmzZ2vvi4uL6du3r9s/00DMfhQRBgxwN7bWN0hPTw8o+fHcuXNMmzYtIBzbIJ4huPLkIUSEZ555hnXr1mlj77//Pm+//TYjR45sd58LFy4watQoIiMjmTlzJiJCY2MjZy+2JDBE63xIsgPEUbZzWOHpjtIELtklwqJFi7j99tvR6XQ0Nze7VQvKLj+25/T6I0VFRV6XHw0GA++9955W0sMdUlNTGTt2bDD70Qepr6+nrKwsYOTH/Px8+vbtGzDyYxDPEHSePIiI8L//+7+sXbtWG9u5cydPPvlkq7531dXV7Ny5k9LS0jaO1dmzZ/n0fBPvn43hWG4ko2N87MnN0TFyqJaen59PbGxslxshewIR0W6mH3zwAVu2bOlyz7Xk5GTGjx8fEBlcZrOZ9PR0r8uPYWFhLFu2jK9+9atuHccuP9pXav2dnJwcBg0aRFRUlLdN8Qj27MdAWKUxGo1kZmYGjPwYxHP4/53BxxARXnjhBR566CFt7L333uOpp57CbDZTXV3Nxo0bOX36NHl5ea2cKoBrZs2mRDecvMowjudGc328j608OUyI9tgiX5xgZs2axaxZs7rk/FRUVGA0GgMmviEzM5Nhw4b5hPw4cuRIevfujVKKI0eOUF1d7dL+dgnFWyucnsZgMJCbm8uYMWO8bYpHKC0tRafTBYz8aFcHAkF+dBdbA+D13XyOV0XEZPunRMTs8P6BdrZ/XUSe6U6brkTQeeoGRISf//znrFmzRhvbsmUL3/nOdzhz5gyVlZUAZGdns3nz5lZ1oCoahTqj9dcSqrNw5tD7bRwsbxJqqGbqtfFMnRPPQJ21Kn56ejqjRo3qngaZXWT48OFanamCggIyMjI62cOKXX4MFAmlp7MfnaW2tpZjx46RnJzs0n52+bFv377dZFnPkpKSElDyY0pKSsB8d+rq6igrK7ti2EUQz6OUWquUClFKhQAfA9p7pdQb3rbPkaDz1E2ICL/61a+4//77tbFdu3bxyiuvtJIbsrKy2LRpE01NTQD0i4Q+tsLZRouOtDILJpPLdcW6jaiRCYwc15+RY/szOLKOutpaysvLfXqCOXLkCJ9++qlTEp4vyI+eJDk5mQkTJvic/BgTE8PXv/51rdipMxmSdvkxUKo719TUUFdXF5QffRR778dAWOF0FhFZICJfisjvRaRWRE6IyFSHTQaIyAFbn7nPRWSgiNwqIgUOx+glIs0ikigiN4pIsu1Y/xERZ9qqdGRfnIjsF5FGETlE6x53PYr/P+74MDqdjv/7v/8jIiKC1157DbCWMaioqOCJJ57gs88+A6w37I0bN7JmzRqio6OZOxw+zIDocDNRU24lPNxaL8dkMnl9+Th6xhyM282EhukJC9Nxdv/7TJq/yKcnmNtvv52mpiZ0Oh1KKZRS7ToTdvnxhhtu8IKVnscuP3oj+9EZ+vTpA1gDjHfv3s2CBQs6lEozMjIYNmxYj7bl6S4Crfhqc3Mzubm5AZP96E35ceQfO+h75WFyn6C9P76pwGtAP+AJ4G0RmWz77DZgCXAM2Ak8BrwI9BKRabZK5DcDqUqpTBH5D7AOOAj8Avgt4E7hyt8ARVidpuuwVkL/xI3jdRnfehwNQHQ6HT/96U+Iy4N/AAAgAElEQVT5zne+o42dPHmSP/3pT60CZ4uLi3n99depqalh9jAjq2ZV8sC1lRQ0WTPG/v3vf7Nlyxavr0JJSCh1xhZ5LjrrHP379/eiRZ0TGhqqrSR9/PHH7N69u91VqPT0dEaPHu1T8mNX8Sf50Wg00tTU1KE83djYSFFRkc/Jj12lqKiIqKiogJEfU1NTGTduXFB+DAwqgP+nlDICvwPiAfsXb59S6qBSqg44ACQopQxY+8gts22zDGuPPDu3YG02/EPgm101SkR0wErgh0qpWqXUPqztX7xC0HnqAUSEH/7whzz44IPa2PHjx/nVr37FwoULtSfPsrIyNmzYwKToamKjrDeS8DAjjSZFYmKiz8RGmPoO1V73ryv1oiWu07t3b2JiYtqsPNXV1VFeXs6IEW6tKvsMeXl5fiM/9u3blwceeECTr9orMeGr8mNX8JXsR09RXV1NXV0dQ4cO7XxjPyA7Ozug5McuUKBsOrpSygKU0yKPFTlsZ6ZFvdoOLLW9XkKL87TUtu9J278Fbtg1EAgFCh3GLrpxPLfw/p34KqG6upply5YxevRonnvO2tPw3Llz/OQnP+HnP/85Bw4cwGKxUFVVxWc7N9JwzT1E9elFWIjik3wTtzhk41RUVKDX6zXZo6fRJ06FVKvEHdFQ5hUbuso111yjva6ursZsNtOvX7+Aqu5sNBrJysryK/nR7hRlZ2eze/du7r77bq0Suq/Lj66SkZHB8OHDA0p+DJTsx+bmZvLy8rwqP15BSutJhttfiEg4MBgotg1dSVL8EHhdRG4CmpRSZ0UkBhijlFotInpgNfAqVrmvK5Tbzh8P2GOsRgCZXTyeW/j/Y5wf4DjBPPLII7z00kvaRJOens7TTz/NwoULtVWlYcOGoQ9tKWZ4uszQ6li7d+9m9+7dPdKxvj1CZ7VMLFFhZiyNDV6xw1327t3L9u3bKSoqQq/X+7z86Cz+LD8OHjyYCRMmEBdnfdB1/O4EAnb5cdSoUd42xSNcvHiR6Ohorz3IeZpAkh/dIFZEHhGRCOBZIEMpldXRDjYZ7yBWmW+bfRh4U0QWYfU1woBLXTVKKWUCdgA/E5HeInIr7q1kuUXQeeoBLl68SK9evbQJZtWqVfzlL3/RvqA5OTl897vfZcGCBUyYMIF77rmHMbEtNXlqjC3tWkSEJUuWsHjxYq896WXWNdHQYI290ut11B/zSrye29xyyy0sXryYtLS0gIlv8Hf5MTIykkWLFhEWFobZbOaLL76gf//+9OrVy9umeQR7BlcgyI8mkykoPwYmmcB8rCs9C3E+wHs7MA14F0ApVQs8CqwHqoFvAQ+CVs/p1S7Y9m1aVsKeBDZ34Rgewf+/wT6OfYIZP358q/E77riDf/zjH9rSfVFREd/+9reZPHkyISEh3DTCISg7wsSl+pYA57i4OO3J/MyZM1y82HOyb1VVFfX19TTqW1oVGM4e7rHze5K+fftiMBgYMmQImZmZHDlyxGureZ4ikCSUI0eOsHnzZvr16+dtUzxCeXk5ZrNZ++76O5mZmYwYMSIoPwYeJqXUA0qpaKXUHKVUKoBS6nmllNZ/TCn1a6XUQw77JWPNsjvhsM1mpVSiUipKKXWNUuqobXytUmotHaCUWqCUev2ysVKl1BKbbbcopR5SSv3a/Ut2naDz1M1kZGRccYJZvHgxGzduJDIyErAGjN97772cOnWKIb101DXpUUphST3Mjs/PttnfaDRy7NgxTp8+3e3XAa3711kGtcgOUuQVydltmpubyc/PZ8yYMRQWFpKXl+fXzlNJSQmhoaEB42xER0fzta99jSFDhnjbFLfxp+xHZ2hoaAg4+dFRHQjiGiLSG3gY2OBtW3qKoPPUjTQ0NFBSUtLhBDNv3jw2b96sZUVVVVWxcuVKDh8+TC99GOYLJzBfOMHFz3e2cZJCQ0NZtWoVixcvBpwrNOgOhYWF2gQTNuVabTzCWNOt5+0uUlJSGDduHHq9nsWLF3PnnXei0+kwGAw0Nzd72zyXsKdXB0oBydraWmpqapg/fz5gXbU5evSo3zq3eXl5ASU/JicnB5z8eLk6EMQljmOV7P7ubUN6Cv//y/dhzp8/71R69ezZs9myZYvW8b6+vp7Vq1cjFw5hyU+zbaXYuXMnR48ebbVvdHQ0ISEhGI1Gtm7dSmpqandcCiaTiQsXLmjxDb3m3EjK6RI+35fJRzuSMbjYo8zbVFVV0dDQoK1qiIgWYP3hhx/y7rvv+lRbnM7IyspiyJAh2iqmv3N59uO5c+c4efJku2UMfB179uO4ceO8bYpHCDT5sSN14GpDKfWRUsplL1IplaSUulYp5V83AjcIOk/dRFlZGRaLxekJZsqUKWzbtk1Lx25qauJvP15HXugAJKYlC+zDDz/k0KFDbfa3WCxOtR/pKhkZGYwcOVKrcK6P7sUlNZDyknrMJgtVX37Zbef2NJ3FN0ybNo3p06ej1+u9YJ3rNDU1UVBQEDDNZduTH+fPn8/q1au12jv+tAKVlpZGYmKiX2Y/Xk4gyo+dqQNBgrRH0HnqBpRSJCcnu5xenZSUxLZt27RigUajkb2//SHnG6NpjG6J+zhw4AD79+9vdQMJDw9nxYoV2tJzXV2dB67Ein2CSUhIaDUeO2OG9tqfnKfCwkKtWGZ7JCQkMGXKFMAaC5Gdnd2T5rmMo/zo71xJfhQRTdo+duwYe/fu7daHBU9RW1tLZWUlw4cP73xjPyA3Nzeg5Edn1YEgQS4n+BfTDeTm5jJgwACio6Nd3nfUqFFs375dexJSFjOf/uUnvJMcQXPvlqejzz77jA8++KCVA2VfRamoqGDDhg186SGH5krp1X2nTdNeV5465ZFzdTeXy4+d8fnnn/PRRx/5rIRXVVVFY2NjQARVg1V+HDp0aIfyo9ls9nqbImcItP51RqOR7OzsgJEfXVUHgriOiLwuIs90ss1XRcRk+2cWEeXw/j/tbL9ARLonPsUFgs6ThzEYDGRnZzN27NguHyM+Pp733nvPIYBRYfjXD3nvbANjx7ZMXMePH2fHjh1tnsD79u3LtGnTSExM7LINdsrKylBKMXDgwDafxU6fDkBIqA7dxTSUxTcdDEcuXLjQSn7sjOXLl3PXXXeh1+u1psK+QqClV9vlx87+bufMmcNtt92GTqejqakJo0MdNF+ipKSE8PDwgMl+TE1NDSj5sSvqQBDPo5T6WCkVopQKAdYC2nul1EJv23clrgrnqSczp9LS0hgzZozbE0xcXBxbt25l2rTp2ljtvp9xJqOo1Rf+zJkz7N69u9W+Op2O+fPn06tXL5RSpKWldemmb7FYOoxviE5I4Pol41iyajKz58TR8OUxl8/RkzQ0NFBaWtpGfuyI8PBwrXnrp59+yvvvv+8zq1CFhYXExMRcUX70N1yRH0UEi8XCjh072LVrl085tWD97qSmpgZMAcna2lqqqqoCSn7sqjoQ6IjIbSKSKiI1IrJVRKJs47eIyFkRqRaRPSIy3Db+vK3o5R4RqRKRj0Skv4g8DTwA/ML22h2bnrUduxC4y+2L9ABXRQ36nsqiqKmpobq62mNPM7GxsbzzztvccPtDlKVbC1H+8+8vEfXYY8ycOZOTJ08SGhrKzJkzr3iMnJwc9uzZw9KlS12eyHNzcxk4cOAVJxgRQaJ6A9Yn/+aTh4ieMcelc/Qk586dcyu9OiIigqamJp+Ij7DLj9dff723TfEIlZWVLsuPOp2OqVOnEhoa6nMrb5mZmZ3Kj/5CoMmPdnXAl3s//vpU2fPAc05u/o9nZgx49LL9Xwa+2cE+LzwzY8Dzlw+KSAKwEVgMpAG7gCdFZBPwNnAP1rIEP8Ba3dv+Q7wfaxPgVcBe4DtKqRdEZCLWwpkvOnktbRCRG7FWKr8ea9Xz3R3v0TN4/y7QQ9TU1HgsBqg9HLNQPDnB9O7dm//57RvEjG/pJ/f3v/+dzz77jLlz57Jq1aoOnwZHjRrFXXfd5XINE4PBQE5OTqfxDab+w7TXSiur4HtcumRtqdSe/Ogss2fP5uabb0ZEtEBgb3HhwgUSEhKclh99Gcfiq65+dyZOnKhJ5IWFhT5RyqCpqYnCwkKPyOa+QKDJj55SBwKUVcAupdQJW7+6x4GjwH3AbqXUAaVUDfA8MENEBtv2e18pddC2z34cmgt7gPuB9Uqp80qpYqDLjpgnuWqcp1OnTvHJJ59QX1/fLccvLi4mPDxcq9XkSa4bHcGaX/2O0dfdpI3985//ZPfu3U71MBs9ejQiQn19Pfv27cNgMHS6T2pqKmPGjOm0QWZIUsuqV3hDWafH9QYWi4Xk5GSPpFfbb+579+5l27ZtXpHw6uvrKS0tZeTIkT1+7u6goKDAbfnRYDCwc+dO9u/f70HLukZKSgpJSUkBkf1oNptJTU0NmOKrNTU1VFVVMWzYsM43vjoZAWhNgJVSKUqpfwMjgQyHcSPWfnX2aHvHyb8Jz6paw4A8h/c914+sA64K2Q6slbynTJnSLRq32WwmLS2NOXO6R7IaE6unyRLJkmd+w/7f/4S0jz4AYNu2bTQ2NvLnP/9ZkybLy8vZs2cPd911V5ub0cWLF0lLS2PatGlaPan2qKmpoaamRkvX74jo6xaiTm1FRIiOAHNNFfqYvm5crefJzc0lLi7Oo7/7hQsXUldX55UbZKA1l83IyHBbfgwLC+OOO+7Q4tO8RWVlJU1NTQwePLjzjf2ArKws4uPjiYiI6HxjH8edFc6exiapPe/G/o9ilbpcpRzQbg4ich2QCFzC6sTYx/sAA2jt1HQXJVidOjs+0fXc/2dfJ9HpdNqyc3JyMufOnfPYsTMzM7t9gomQUHT6EBZ97xdMXLRSG//ggw9Yu3YtjY2NVFVV8cYbb5CTk8OGDRuoqKhodYyxY8fyyCOPaI5Te+nersY3hA0aQn2DNdtPdEL94QPuXKbHscuP7mQ/tke/fv20Vb/U1FSOHTvWI0HLnpAffQlPyo/x8fFER0ejlOLzzz+ntrbWAxY6T3dJ997CLj+OHj3a26Z4hO5UBwKI94AVIjJFRPoDvwcigG3AvSIyV0R6YXXsPlBKVXVyPBNWJ8sdtgKPisgEm0z4PTeP5xGuGufJjlKK1NRUUlNTPXKza2xs5OLFi90+wSTFWvV50emY/c3/5dFHWx4qDh48yNe//nVyc3O14phVVVVs2LCB0tLSVsexB7BmZGSwceNGqi9rq1JcXExkZKRLE0xTWMvTvjH5uGsX1s2kpqYyduzYTuVHd8jOziYrK6vbizba5cdASa+2y4+uZD86Q3V1NSdOnOi2VkVXoqCggD59+gRM9mNycjLjx48PGPkxLS0tYOTH7kIpdQr4EfA+kA4cBl5TSn0JfBf4J9ZVqAnAt5045D7g2yLypL2eUxds2oO1Z94nNnvedfUY3YH4WopvdzBr1ix1/HjLTd1kMqGUIjQ0FKWUW0+JJ0+eZOjQod2+TF/RaGF9SgU6AaVgdWIsW175A7/73e+0bWbMmMHPfvYzPvjgA21VKTIykjVr1jB06NBWxysuLubIkSMsXbpUC5w0m80cOnSIOXPmuLSKVvaXnzHg0kkAKpsjif3VW+5erkeoqanhzJkzXH/99d26EqCUwmAwEB4ejtFoxGw2d8sqZFZWFs3NzQGT/n706FFGjx7NgAHuPpi2pbq6mpiYmB5bATKZTBw6dIjrr78+IIL4KyoqtFCEQFhFS09PR0Q8vgLdESJyQik1q8dOGKRHuepWngBCQkIIDQ3FZDKxY8cOTpw40aXjVFRU0Nzc3GH8kKfoF6mjodm6eiICH+WbeOqpp/jJT36ibXPq1Cm+//3vs2zZMm0Cb2xsZOPGjeTm5rY63uDBg7njjju0n0NhYSGZmZkMGzbM5Rt/2PTrtNeRynNtYdyhJwtIiogWc7Zv3z7eeecdj1fANhgM5Obm9ujk351cunQJEekWxwmgT58+iAg1NTXs3bu32wtppqenM2rUqIBwnPwpNsgZekodCHJ1cVU6T3ZEBL1e36WUVW9MMH0d7MyqsWbMrVu3jl/+8pfaeEpKCv/1X//FkiVLNCfIYDCwadMmMjIyaI8jR47w5ptvcuHChS5NML1mzcNktEpWERF6mrLTXT6GpykqKiIyMrLHA4inTJnC1KlTPS4T9oT82FN4MvuxM0pLS8nMzOzWshL19fWUlZUFTPZjfn4+ffv21XoJ+jspKSkBIz8G8R2uaudJr9ezfPlypk6dCuBSGYP8/HxiY2N7dIKZMbDlqdasM2KxWCXXBx98kD/84Q9a9lVWVhbr1q1j8eLFWoaZyWRi8+bNJCcntznurFmzGDt2LNdcc02XMrh04eHUG/TU1zZTmF1F7dkzXbk8j2E2m0lPT/eKvDVixAhm2BomFxcXk57uviNpz360N4z2d3JycoiLiyMqKqrbzzVmzBjWrl3brf3L7NmPgbBKYzQayczMdLkunK/Sk+pAkKuLq9p5gpa6PVVVVbz++utOSXj2CSYpKam7zWvFDcNCMJqt9kaHWzh9qSVA+d5772X9+vXaKlp+fj6PPPIIN954I3369AGsT/xbt24lL691dml9fT1Dhgxh8ODBXLp0iU8++cTl4OfCvjM5sCONk5/mUZae48ZVuk9X5UdPc/ToUT7++GO3JLxA61/nDfnR/ndw/vx59u3b59GsyNLS0m6VH3sau/wYCAUkA01+DOJbXPXOk52YmBimTJniVFVgb00w4XrBYLDKNgaT8Hlh6wKNy5YtY8OGDdrNori4mAcffJBZs2ZpZRomTZrUqkCcY3o1WFetkpOTXa7U3HdGS7HMypMnXb84D+FL8Q1Lly7l3nvvJSQkBKVUl7LxioqKiIqK8nr9Ik+RkpLiNfmxqqqK6upqj8WjWSwWUlJSekR+7Anq6uooLy8Pyo9BgjhB0HmyYW+ma79J5ebmtvuE6u0JZlSvKLaf7sOGL/rx2vEwqi/reXzjjTeyadMmTa6rrKzkoYceIjExkeuuu44777yzlTSXl5fXSn689tprefDBB7V6Oc4G2sY69NcrP3qU2ivEV3U3ycnJTJgwwScKSIaEhGh/T4cPH2bbtm0uBS57U37sDqqrq6mrq/Oa/Dh37lzuvvtuLcvWXXJychg0aFCPyI89QSDVqAo0+dEXsDUAXu9tO5xBRCaKSKatSfFXReScbfx1EXnGE+fw/h3GB8nOzmbr1q2kpbXt1ebtCWbl+FDEEopFCeWN8NIXbbe57rrreOedd7RaTY2Njaxbt47y8vJWQZP2Ceby1TZ7Lajjx4/z5ptvOrUKFTl4MINusrWPUYr89X/s4hV2nYqKCoxGo0/GN/Tu3Zs+ffq4tOKSkZHBsGHDeqyxdXfiC81l7QkiJpOJ7du3c9KNFdLm5mZyc3MZM2aMBy30HqWlpej1evr37+9tUzxCIMmPQdoiIp1F/88GSpRSC5RSHyulPF4cL+g8tUNCQgJLlixp0xTXFyaY6DD4yfyW9/88A+dK2243Y8YMdu7cqTUNNpvN/M///A+///3vtafutLQ0SkpKeOedd2hqampzjEGDBjF06FCnY4fGfOtbDBjSi7mLRzMhKp+Gc10rAdEVHGODfJHJkyezePFircfgxYsdt2dqbGykqKjIJ+RHT1BUVER0dLRPyI86nY6QkBC3sq9SU1MZN25cMPvRB/G2OuDPiMgCEflSRH4vIrUickJEpjpsMkBEDohIk4h8LiIDReRWESlwOEYvEWkWkUQRuVFEkm3H+o+IdNhaRUQSRKRKRL5v+z9NRG60ffaQiOwRkV3ATtvYAyJyQUQqRWSTiMSKyFeB14A5tpWnBSLSpmKu7Vz/ttl2RESmufKz8v9vfjcgIlol2sbGRs6ePcs111xDcnIy1157rZetg6VjYN4IOJRndRp+f9zAy7eGode1fqJPTExk165dfP3rX9fa0fz2t7+luLiYH/3oR5w8eZLs7GwAXnvtNVavXq0Fl4M1c8zegqSxsZHKyso2xTYd6TdzJuOvG0NstDW2p/HdvxA1+TWPXvuVyMvLo3///vTq1atHzucOBw8eJC8vj0ceeeSKdYF8SX50F7v8eN1113W+cQ+g0+lYvny5tgJmNBpdWqGwy4/2LF1/Jzs7myFDhmgrzv6Ot9UBd4mPj++xytWFhYXt/ZCmYnU++gFPAG+LiP2p9DZgCXAMqwPzGPAi0EtEptkqkd8MpCqlMkXkP8A64CDwC+C3wIpOzOqDtb/eIOAO4F0RsT9FLgHuAXaJyBzbuW8FcrG2kvmzUmq1iKwFVimlbhWRBZefQER0wC7gr8CdwIPAdsDpp1X/n5m7mdTUVD777DNOnjzpMxOMCLywAIbGGLl7RjVjh9Ty+rnmdreNi4tj69atzJs3TxvbtGkTDz74YKsMoUuXLvHqq69SUlLS7nEOHjzI9u3b212hamXbV+/VXseqCpoyU1y4sq5hNBrJyspqs1LoqyxcuJDbbrvtio5TeXm5z8qPXSEjI4Phw4f7lPxov7GWlpby6quvtikieyUCLfuxubmZvLw8pxJl/AFfUAcCgArg/ymljMDvgHhanIp9SqmDSqk64ACQoJQyAB8Ay2zbLMPaI8/OLVgdoR8C33Ti/Ar4uVKqWSn1LlAM2FctjimldiilLMBDwMtKqdNKqUrgx8Dd4twXczagV0qtV0rVK6X+CuhcWX0KOk+dMH36dFatWkVdXZ1PxTckxsIdkwzE9bZmDuU31VNa3342V+/evXnjjTe4++67tbFjx46xfv16Fi9erK1u1NbWsmHDBnJyctocY8GCBSxbtqxTCa/PspVU28pl6fRC/Zt/6MLVuUZ6ejqjR4/2m/iGyMhIbUUvIyODffv2YTZbMycd06sDAbv8OGrUKG+b0i69e/dmyJAhTveju3jxIr169Wq1QuvPBOXHIO1QoGyxHTYnpRywF0orctjOTIt6tR1Yanu9hBbnaalt35O2fwucOH/tZQ2HLzmc33F8JJBx2XYRgDPplSOB8Tb5sUlEmrA6iU73WfP/b0w3IyKUlJSQlJREXl4e6enpLFq0yCfklHVTo/j9l81EhlmICFWsP9vAs3Pal63CwsL44x//SFxcHOvXWxMmTpw4wTPPPMOLL77IgQMHMBgMNDc3s2nTJu68885WN/CoqCitgWtWVhYlJSXt9r0SEcxfWQ7ndwPQ11BMc0EO4cMSPP8DwBrfUFFR4bcNP8vLyykrK8NisaDX6/1KfnQGewFJX/i+tEdkZCR33HGH9r6hoeGK2XN2+XHu3Lk9ZV63EpQffZMrSGk9yXD7CxEJx+pQFNuGriQpfgi8LiI3AU1KqbMiEgOMscloemA18Cq2eKUOiBGRPkope9f6kbbzXx4vdQkY5vB+LFCllKpxYvGpFDiulNLicERkBuB0N3HfnNF8iKqqKq2IZFlZGSUlJRgMBm+bBUDfCGFCTLT2PiSsiU8LrlzDRqfTcd999/H4449rTk9GRgbf+ta3mD9/vnbDNpvNbNu2jS++aCeVD+sklZGRccV6ObH3PEStbRVMH6Kj7o2XunR9zuDtDC53ufbaa1mxYgWhoaFafJ2/yI+dUV5ejtls7tbq3p7k2LFjbNy4kZqamnY/z8jIYMSIET4lP3aVQJUffUkd8GNiReQREYkAngUylFJZHe1gk/EOYpX5ttmHgTdFZBFWXyMMq8PjDE+JSLiIrAN6Ae3djLYA62xlCfoCzwEbnTz+YSBORG6znWc1sAfrappTBJ2nDri8Qu3s2bO5//77iYiIwGKxaFKLN1mRFEZdo1Wu0gl8mFeH2dL+w0FTUxMFBQX84Ac/4K9//asWc1NSUsLatWuZOnVqqzioffv28a9//atNTZybbrqJe++9l9DQUMxmc5vaRaLTYZh6s/a+T30exkvFeJqSkhJCQ0O1AqD+il0yeeuttzhz5ozHmwp7g8uLr/oDiYmJTJw4sd1Vv4aGBoqLi31WfnSVQJMfU1JSSEpKCvav8wyZwHysct1COg/wtrMdmAa8C6CUqgUeBdYD1cC3sAZmIyKvisirVzhOM1b5rRz4DnCXUqpNvRyl1G7gT8A+oACwAD9xxlClVBNwl237SuAp23mcXhkRT7Yq8FVmzZqljh8/7vJ+BQUFlJeXM21a6xgypRT79u2jqamJ5cuXe12S+LLUxJ78KvQ2M/rre/HNqW1jk06fPk1cXJyWMff555/zjW98g9raWgBCQ0N56aWXqK6uJj8/H4Do6Ggee+yxK1bp3bdvH+Xl5axYsaLVxKVMRhp+dA/RUdaxsugxDPj+bzx2zRaLhU8++YRrr73W75fpwRpvdujQIYYOHcr06dO9bY7b5ObmUldX51fOkyONjY0opTQJ7/jx44wYMcJvVtE6wmQycejQIebOnRsQq2hVVVWcP3+euXPn+tQqmoicUErN8rYdrmDLTFuvlHK5uqiIXAe8ppTqclVfEUnAmqnn3d5aThBceboCJpOJCxcutFvdWUSIi4tj4MCBXnecAKbFhRBiaXEgCpvqKa5rHTxeVVVFQ0MDQ4YM0cbmzp3Le++9x+DB1hg5o9HId7/7XYxGIxMmTCAsLIzVq1d32N4gISGBUaNGtXnik5BQGsfeoL2PqUzDVO25zvZZWVkMHTo0IBwnsMYGzZkzR3OcLl26xNGjRz3ah62n8Lfsx8tRSrFjxw62b9+OUsrv5MfOCDT5Mdi/zvuISG/gYWCDt23pKbx/5/dRMjIyGDly5BXTyWfMmKEFjtbU1LjUdqM7+NbUKBoM1l9nRKji5bMN2mcdxTdMnDiRXbt2tWrU+tOf/pSUlBQefvjhVs5We4wbN445c+YA1grfly61SNr9HvguDY1WaTMsTE/Zpr+5d5E27PJjoKRXFxcXt5Efk5OTOXnyZKelIXyRtLQ0EhMT/Sb78XJEhDlz5nD99dcDBFT2Y0NDAyUlJQEjPxYWFgaU/O1uwisAACAASURBVOjHHMcq2f3d24b0FEHnqR3sE4w9u6wjjEYj7777Lnv37u1+wzogJlyY3KcleDw0vIlP8q2xM4WFhcTExFwxHTs+Pp7t27cze/Zsbezll1/mZz/7WZvg+Ly8vHZrQdmlzN27d2sNcHXhEdQPn0VBdiUf7U7j7FvvY25uvx6VK6SkpDBu3LiAiG+wWCykpqa2yRacP38+a9asITIyEqWUzyQpdEZtbS2VlZVaZXt/ZdSoUSQkJGg9LgNhlQasjmCgFF/tSB0I0jWUUh91RbJTSiUppa51yJDr6vlz/EGyg6Dz1C6upFeHhoYyd+5cvvKVr/SAZR3ztXHW4HGzBU4XRPK7z3U0G6wTTGcNMmNjY9m8eTO33nqrNrZjxw6+/vWvazFRpaWlvPXWW+3WghIRli5dyrJly1r93Po98jTJqY3UVjXTVFxMwY4dbl1jVVUVjY2Nna6I+QtXkh9FRAtcPn78OJs2baKhoaG9Q/gMvtC/zpMYjUbOnz/PiRMn+Pzzz71tjtvYS2IEkvzYkToQJEh3EnSeLqOsrAylFAMHDnR6n4kTJ2rVoDMyMmj2wOpKV9DphPvG9mLn6b4czo7mVLGOP398yekJJjIykpdffpkHHnhAG/v000+55557KC4uZuvWrTQ3N2u1oOwtX+zExMRoP4dTp05x6tQp9BERjP7GN7RtMtavx9LFbLJAS692Vn4cOnQow4cP9/n4rpKSEsLCwvw++9FOamoqU6ZMYenSpa1WZf0RpRTJyckBJz86ow4ECdIdBJ0nBywWi1vp1dXV1ezevZsjR4542DLnmTRQz6qJLbVPX00bQMygBKf31+v1/PKXv+Tpp5/Wxs6fP8+dd97JrFmznKoFpZQiLy+PvLw8lFIkrF5NqC0moSEvj4vvv9+la+tMfvQ3nE2vjo+PZ9GiRYgIjY2NZGRkdLi9N7iS/Oiv1NbWUlVVxfDhwxk/fjyRkZFYLBan27j4Grm5uQwYMIDo6OjON/YDfL34ahArIvK6iDzTg+fbJiKNtqbEGSJy7ZUaA7tL8C/PgdzcXAYOHNjlCaZPnz7ce++9Xq9A/O3ZMMzmX9SbQ/ntYdd+zSLCE088we9+9zvtxp6fn8/DDz/M7NmzO60FJSIsX76cpUuXIiKYQkIYsWYNIaE6xkweSK+PX0WZXVt9ssc3dCY/+guVlZU0NjZqmY7OcvjwYd5//33q6uq6ybKukZmZGTDZj1da4Tx16hRbt26ltLTUi9a5jsFgIDs722+zHy/HLj+6og4E8X/ESmc3s7uB6UqpTKXUGKVUt61kBJ0nGwaDgZycHLcnmGHDhhESEoLJZOLDDz+kstJz6fnOEhEC35tujduLDjNTbq7hYJ7r2YArV65kw4YN2g2xqqqKhx9+mISEhFYBwYcPH2bbtm2tijvqdDpCQ0NRSrF7927O9e3DgtuTmDBjCH1j9FS98w+XbLlw4QIJCQkBEd/gTnr1vHnz+NrXvqatAPpCKYOmpiYKCwsDJvuxpKSE8PBwYmNjW41PmzaN2267ze9ihtLS0hgzZkzA9K8LpOxHX8RWdTtVRGpEZKuIRNnGbxGRsyJSLSJ7RGS4bfx5W9HLPSJSJSIfiUh/EXkaeAD4he11R+f8yHacZBEpF5H/szf4FRElIt/D2tcuzraq9B+bfSdE5AbbdvYl+fO2bXJEZM5l59GLyK9EpEhELorId7v6c/L/b5OHSE1N9egEU1NTQ05ODqNGjWozCXc3FouFATWnWZ40m0EDGgjVw/4CCzfE9yFU79rNeuHChWzZsoUHHniAiooKmpqaWLduHb/85S8ZP348qanW1dDz589TX1/PypUrWzUPFhGmT5+OyWSivvAkkeaLAISc/hdq1WOIE8vu9fX1lJaWMm/ePJds91UKCgq6LD+GhIQQHx8PWIPNv/zyS5YuXerVbLDk5OSAqe5sNptJTU3Vym84EhISQlJSEmCV6ENCQnxeBqupqaG6ujpgnA131QF/4YUXXngea7sRZ/jHc8899+hl+78MfLOjUzz33HPPXz5oK1K5EVgMpAG7gCdFZBPwNnAP1rIEPwA2A/ZifvdjbQK8CtgLfEcp9YKITMRa9PJFJ65jBXAz1kWdA8Bp2zkAZmBt3NsAnMXaI+924BbgPRGZpJQaIyIKGKaUKr7Cg+mTwExgMjAE2CsiXyqlPnbCvlYEV56wTjA1NTUMGzas842dpF+/fnzjG9/QZKaerAOVm5tLXFwc35oVjv3vp3ekiVfPdq1m0IwZM9i5cycjR44ErM7ZM888Q35+fqtA2pycHDZs2NAmrX7s2LFMmDCBXg8+hdlkLWPQO0qo3vlPp85vj0MLhPgGk8lERkaGR+THhoYGGhoavBo8X1lZSXNzs8vyo6+SlZVFfHx8qweAyzGbzbz77rv861//6kHLXCfQsh89pQ4E6ZBVwC6l1Albv7rHgaPAfcBupdSB/8/eece3VZ/7//2VbMvyih07TmI7wwnZi+yEDBIgoyXNIIEEQqEhlFJIS2lpuVwK97alv9vLvS23LWVlEcjeO2SSvcgm3tuO97Zly5Ysnd8fio7teEm2JEu23q8Xrxc5PuN7ZPk5z3k+z5AkqQz4T2C0EML8h39IkqRv7x9zgjrDha3gM0mSsiRJuoepX9STdX725/vnnoBp1t1HkiRVSJK0G7iNaYyMJawEfi9JUqEkSXeBNcDCVqzV7TzZ08CYowF5eXmsXbtWHnliT8wGZsCAAQwLUeIt1eag5OorySwzNnN00/Tr1499+/YxYsQIedtHH33E6dOnmTlzprxtwIABTUprql6RJJfXNk5UXN4HLchO+fn5CCHq5Vm5MvHx8TaTH4cPH86zzz6Ll5cXBoOBoqIiG6zQcurOr+sID2ez/NivX79m91MqlTz++OPMmDHDMQtrJTk5OXh7ezs88m0vbK0OuGmU3oA8BFiSpBhJko4DfYDEOtv1mObVmTXsgjrnqKJ1qlbdB2R+nXODSbLj/jqSpPr5Cg/u2xx9gHNCiCohRBXwDtC9FWt1y3Y5OTmo1Wq7GhhfX1969OhBYGCg3a5hJjY2lgEDBsgG5tWRPnx4oxoflRGVh8TqqAr+c3LT41aao1u3buzcuZNXXnmFM2dMUc5NmzaRl5fHqlWryMzM5PHHm34BEEIQsvIdjDv+hEIpCFBLlB7dRZe5Sxrd32g0Eh0d7fJl4mYqKirIz89n+vTpNjunORp34cIFbt++zYoVKxodbGsP7t27R5cuXTpM9WN0dDSDBw+2SH6s62AVFhYSHBxsz6VZjcFgIC4urlH50RUxqwN1X946Mvcltf9sw/GvYBrKay2F1HEm7s+r64/JQYmos70LEAKkt3aNjVA3WtUHaGyafL113GcAJknREvKAH0iSFAMghIiglUGkTh15MhsYe3eo9fX1ZeHChfj7+yNJEpmZmXa5jtnAmHNiAPy8BKO61j5Mvb2rOZnWegnRz8+PL7/8ksWLF8vbjh8/zh/+8AemTZvWIALxYEJz0MixFEu1oxR0x75uMuk5NTWV0NBQeTirq2PPKM3o0aOZPn26wxwns/xozgFydYqKiqiurpb7lFlKTEwMGzZs4N69e3ZaWetISkpqUX50FTqa/Ojk7AaeEUKMEEIEAx8B3sAu4GkhxCNCCD9Mjt1hSZJKmj4VADWYnCxLWCGECBVCDAJ+BjTWUfki4COE+JkQwkcIsQyTM3XMwmtsB/5NCOEnhBgInARGWnhsPTq185SUlERERIRDDUx0dDRbt261uYTXXAPJxQO9qNDWykSnMjXoDa2v0vLy8uLvf/87q1atkrfduHGDBQsW1Lsvo9HI1q1bG/SC8l7yOpLRdP1uAQo0Zxvmjuh0OtLS0urN3HNl7C0/+vv7M2rUKMDkCFy6dMmulXjx8fFERkZ2+urHhx56iKlTpzpVx3utVktWVlaHqX50hDrgxoQkSTeBfwcOAfHAZWCdJEm3gV8CX2OK/gwBXrPglMeA14QQbwohHhVCNNej5sr9/y4BX9zPZ3pwfVXAAuBlTFGyfweevr/dEv4T0GGKmJ0BvpQk6aCFx9ZDOEOps70ZN26cdO3atXrbtFotV65cYfr06Q5NRDYYDERFRTFixAibvkVlZWWRm5vL6NGjG/15TKGBXSnFeNxXJAKEL6893PaePOvXr+e9996TH9ShoaF8/fXXDBs2jAMHDnDz5k0AJk2axOzZs+V7LnznOYJVWgCKq73hd58ghJClzdu3bxMcHGzTJP72wmg0cu7cOcaPH++QKNrFixe5ffs2P/7xj+0SiaqoqOD69euNRhpdkfT0dEpLS9ssCel0OgwGQ7v3urp+/Trh4eEdIonfYDBw7tw5Jk2a5HJRNCHEdUmSxrX3OlwBIcRpTAnjlspv7U6njTyZ8xscXcGlVCoZOXIkQggqKyttEiEwGAzEx8c32915SLASNbUP7nx9JRmtTB6vy4oVK/jss8/kCEReXh6LFy/mzJkzFBTU5hA+2AvKa35tFW2QqorL6/7F3r17MRqNlJWVUV5eXk9+dGVSU1Pp3r27w+THyZMn13OcbF3peffuXYYOHdohHCe9Xk9SUlKbqx8lSWLnzp0cPHiwXXtvFRUVodfrrZYfnZX2UAfcuLGETuk8mQ1Me7+ZxcXFcfXq1TZXSSUmJhIREdFir5+fjVRTUW36las8JD6/ZZsZfPPmzWPz5s1y4rBGo+EnP/kJXbp0qfdQioqKYuPGjVRVVeE/eSZFWi+0lXrufpdFcGq2PIKkI82vM8uPDz30kMOuWXeo8M2bN/n666+pqKiwybnz8vJQKBQdpvoxISGByMhIPD09W965GYQQjBkzhnHjxrXb97ajVT+a5ceWqh/duGkPOp3zVNfAtDejR4/mxRdflCt1WvPGqtVqyc7OtsjA+HkJxgT7UaJVcOj7AL6+qea7LKsv2SiTJ09mz549skOq1+v55S9/SVFRUb1qubS0NNatW0dpaSmKhas4tSeWlNgC8o4eJ9BgIDs7m6ysLBITE52ie3ZbiYmJqVf96GhCQ0MJDw+3iZRkNBqJiYlxir8dW6DRaCgoKJD7l7WVwYMHExkZCZj+Lh1NRkYGQUFB+Pu3rprW2WgvdcCN45EkaYYrSXbQCZ2n9PR0pzIw5hyfxMREduzYQXW1ddGg6OhohgwZYrGBWTTAi6KiINKLTTLb745Dha6Fgyxk8ODB7N+/v14Tuw8++ICzZ8/W6wWVn5/PmjVrqOzTn64TTaXUksFA9H//N3FxcSiVStLS0lzeeSotLUWj0bSr/BgeHs6cOXNQKBRUVVWRnJzc8kFNkJKS4lD50d7YK0qTnp7O6tWrSU+3ZRV385jlx45U/egM6oAbN03RqZwnvV5PcnKyUxqYmpoajEajVYa8sLCwVfkN708X+NxXKZJL4N1vW+xVaTHh4eHs2bOHiRMnytu+/vprPvvsM+bOnSs7eRqNhvXr16N85hl5v5xvvqFLcjILFy5k/vz5KBQKqqurG3QsdwWcsbz68uXL7N+/n/LycquPra6uJj093aHyoz3Jy8tDqVTapT9T9+7dGTx4sEMH18bHx9OvX782y4/OgDOpA27cNEWncp6c2cAMHjyYpUuXyt2iNRpNs/vXLa+2logA+HNtIIhvkoxsuGs7ByUwMJDNmzezcGFt1/szZ87wzjvvMGfOHDn5s6amhhtZWYQ//TQAwaG+9Pt+F7r0ZHmo8OHDh9mxYwdGY9uT2x1JdnY2vr6+DmmMailTpkxh8eLFrYq6xsbGMnDgwA7R3dne8qNKpWL27Nmo1WokSbK786/RaCgsLKR37952vY6jcDZ1wI2bxug0zpMrGBhzhOLkyZNs3ry5WQkvPT2d4ODgVpeiPzUEnh4C4YE6nhlbTKaujKvZzbXgsA5vb28+/vhj3nzzTXlbXFwcP/3pT5k8eTJBQUH4+vqybNkyRrz7LgMm9OGROf0JDFRR+em7SPejcKNGjWLUqFEulffgqOar1uLp6UmvXqYmvunp6ezbt8+iB7tZfgwLC7P3Eh1CSkoKPXr0sHtLAbPzv3//frs6/x0pSdyZ1QE3buriOk+kNuJsEkpzjBo1inHjxjVZPWc2MG0dkPmHGRIzBlTg4yWhVMDBtDJKqmxn5IUQvPXWW/zjH/+QWxkUFBSwYsUKevfuzfPPP09gYCBlOh26KbPk44K8qyn+6u+AaQyGObqWmZlJamqqzdZnLxITE+ndu3eL1Y/tSWlpKaWlpS0+1JtrvuqKOFJ+FELQu3dv+vTpY7fPLjc3127yY3vgzOqAGzd16RTOk16vx9PT02UMTPfu3RkzZgxgymt6cJxLXFycTQyMr5dgUaQ/eoPJsPuojPzjlsbmidqLFy9m69atcofgqqoqXn/9dVmOi4qKot9PfkGBIYhoulCEF37x31KVmiCfQ5IkLly4wOnTp51awjNXP5qrrpyVESNGsHz5cry9vTEajZSUND5lISsrCz8/P7p06dLoz10Ns/xoyfw6WzBixAjGjx+PEAKDwWDTc3fE6kdnVwfcuDHTKZynqqqqZhtIOjMnT57kyJEjsuEtLy+nuLjYZgZmfE8PenrVSn9eKh1fRlna6d5yJk6cyIEDB+q1VPjLX/7Cq6++SkBAAH5+fpQs+SW76MsaBpLjFYD2s98j3XeUhBAsWLCARYsWoVAoMBqNTlmNFxUVxdChQ11CZjQ7EBcvXmTjxo0NEslramqIj49vcwNJZ6E95ceCggLWrVtn0xl4KSkp9OzZs907mtsKV1IHWsLWjWndOB/Ob+FtgKenp8samHnz5rFo0SL5QWeP/IaVI1RUVNZ28M2qruB6ju3yn8xERkayf/9+Jk+eLG87dOgQf/rTn8jPz2f/yVMYhUArPPiK/mR4+1L85UfyviqVSo6AnDt3jgMHDjhVFKqwsBCDwUBoaGh7L8UqRo4cyeTJkxsk6CYlJTm9/Ggp7S0/+vn5ERISYjM5qrq6moyMjA5T/Zibm+tS6kBLxMXFtfcS3NiZTuE8uXJrfx8fH9mgHDt2jISEBJsPyBQCfj3Gl1KtyUFTKmB/Whll1bZ3TIKCgti8eTNP36+wA1MJ/VNPPcXUqVPx9fUFwCAU7BF9uJmagDYxpsF5fHx88PPzc5oIjyuXVwcEBDB27FgASkpKuH37NpWVleTk5Di9/Ggp7S0/ent7s2jRIpuNTYmJiXGo/GhPzPKjq6oDD1JeXt7mqRFunB/nePK4aRGDwcDt27ft1qAwSC1Y0DcAXY3prVztZeSft22f/wTg5eXFH//4R5YuXSpvS05O5qWXXmLcuHF069pV3n5e2ZO9G75A/0BV2Pjx43nssccAkxyTl5dn83VaQ1paWpuqH52Fmzdvcv78eW7cuGFV81VnxpnkR0mSuHjxIg8OKreGkpISKioq6Nmzpw1X1n4kJyd3KPmxtS1k3LgWrm8ZOwkpKSnMnz+fhQsXIoRAq9XKQ3ZtxaQwJd08ah/+Sk8dG2Nsn/8Eps7ov//97/n0009lWai4uJgVK1bQrUcPegf4yvvGewSy7q//3eR8tpMnT7Jnzx6bfx6WotfrSUlJaXP1ozPw6KOPMmfOHDw9PQkNDXXKvDJrSUxMpE+fPk4jP+bn51NUVNSqz7a95Udb01Hlx651XgDddEw6jfPkyg+Bqqoq7t27x0MPPYRSqUSSJPbv38/evXttfl+vPqyiVFMrc57JqCGjzKaXICcnRzYw8+fPZ8eOHbI0qdPp+NWvfkWhAYZJtY1Cc3RGVn/+Ofn5+Q3ON3v2bObNm9duDRzj4uLo379/hyivFkKQlZXF8OHDuXPnDnv37nXp5NfKykpyc3Pp27dvey8FMH2+8+bNY/bs2a1yfjIzM/H39+8w1Y8xMTEMGjTILT+6cTk6hfNUUVHBkSNHnCq52BoezG8wN48cOXKkzd8+hYDfjvWlUOPBqTg/jsf684sjoLdRlbXRaCQ2NraegRk7diwHDx6sF7n56KOPOJVVyviSVHl2TGl5OevXr2/QPNTPz0+eHxcVFcWxY8dsXhbeFObqR3PzSVcnLS2NkJAQOfcMcGnpLioqyunkR/PfcVlZGVeuXLH4BaimpoaEhASna77aWkpKSqisrHTLj25cEuexKHbGVSNPxcXFaLXaBgZm8ODBsrORlpZGQUGBza7Z1UewrH8XkvJNEaibOfA/l2xz7qSkJMLCwhoYmN69e7Nv3z6mT58ub9t36BCf3Emn56lTKO5HP0YHBjYrv5SVlVFaWmqbxbaAM86vaws6nY6UlBQGDBgAmKrwFi5ciFKpRKfTUVZm4xCknSkoKECSJKetfoyLi+PKlSsWf18TEhLo06eP3HDWlelo8mNddcBN56BTOE++vr788Ic/lCfLt1dujLXUnV/XlIExGo2cPHmSU6dO2fTaY8MEv32k9t+fX4dvU9t2zqqqKjIzM+nfv3+jPw8ICOCrr77i+eefl7fdjI3li/x8go8cIezGDWr+9S+0WVlNXmPy5Mk89dRT8gO/uLi4bYtuhtzcXFQqVYfJb2hMfjR/744ePcq2bdtcRsKTJIno6Ginrn4cO3YsL774okXzDysrK8nLy3Ma+bGtZGZmEhAQQEBAQHsvxSZ0pOpHN5bRKZwnMD0EjEYju3fv5sCBAy4Ribp3716LBkahULB48WJ++MMfAraNsL0yBmb0Mf2/h0JiS3w53xe0/uFpSX6Dp6cnf/nLX3j//fflB3d2WRkf5eVRcvs23ko9Jf/zptw8s7y8vMF8NvP5v/32W7Zs2UJVle2T3g0GQwP50ZUpLy+npKSkSflx0qRJTJ061WXyuhqTH50NhUIh5y6lpqY2+1LnSs1XW8IsPzpD9aMtaEodcNOxcf2/RCtQKBSMHDmSESNGOH2ouKamhsTERIsMTJcuXfDz80OSJI4fP87NmzdtsgaFgL/Nhv5da1g8uoR+3arZnVxOuc763DGzgenRo0eL+woh+NnPfsaaNWtkea/SaOSfej3ZA4Lo2UVH0Rf/hU6nY9OmTaxfv75RSWnixInMmDHDLn2+kpOTCQsLc+keYmYskR+7desm59pkZmaSmJjoyCVahVl+dJXqx8LCQnbt2sWNGzca/Xl+fj6SJNGtWzcHr8w+dDT5sSV1wE3HpFM5TwDDhw+XdenMzEy0Wm07r6hxEhIS6Nu3r1UGxmg0UllZ2WRJf2sI9oH3pgt8vUwOk8rTyCd3rOv/1FoDM3fuXHbv3i03FjQAvz2fyv+7moFf+hW2b1hPbm4uOTk5rFmzhuzs7HrHBwYGypGhrKwszp07Z5PIXEvyo6thrfx45coVzp8/77CkfGuJi4vjoYcearfqS2sJDg5mwYIFcqPSuhiNRqeXH62ho8mPlqgDbjomnc55MlNVVcWePXs4ffp0ey+lARUVFa0yMEqlkvnz5zNlyhTAJMXY4gE3s48Sf2r7P0lKHdvjLXc67927R5cuXVplYEaOHMnBgwfryWP/up3NqrPJhCddlWWM8vuVeLGxsY2eJzk5mbi4OJtIeNHR0R2mvLo18uO8efNYvHix3DbDmSRwc8FAREREey/FKsxtSGpqaqisrJS3p6WlERoa6tTyozXcvXu3Q8mPlqoDbjoerv8NbiXe3t7Mnz+fRx99tL2X0oC2zK9TKBQIIdDr9Wzbto2jR4/aZE1vjFORX1IrUSVpKrlrQf6T2cAMGjSo1dcOCwtjz549PP744/K2wynF/PWbczzqoZWlM/M9X7x4scEDfcqUKSxfvhy1Wo0kSfUeUNZQXFxMdXW1RfKjK5CcnEx4eLhV8qOXl5c8B+/s2bOcPHnSKRyouiNyXFFCkSSJnTt3yjmZOp2O1NRUufrR1TH3aOtI8qO16oCbjkOndZ7AVB7v4+ODJElcuHDBKUqx8/PzEUIQEhLSpvN4enoyadIkRo8ebZN1KRXwm/Gm/k8ACgXsTimnQt98/lN8fLxNDIyfnx/r169n5cqV8rab+RW8tX4LU/r3qTfv7/jx4xw8eLBe1E0IIedPXb58ma+++spqebOjtSYwy4/9+vVr1fGSJCGEcJrPIicnB29vb5vPfnQUQghGjx7N+PHjEUIQGxvrUvJjc3Q0+dGsDvTp06e9l+KmnejUzpOZkpISbty4QXx8fLuuw9YGZvjw4XIFSFRUVJvL9sP8BLPD/anSmx6WXh5GPrlT3mTUoaKigvz8fJvlNyiVSv74xz/yx/d+j+L+8zqzQsdLb/2OwYMG0rt3b3nfGzdusGnTpkZz2gYMGMDw4cOtnhN47949AgMDO0x+Q3R0NIMHD261/CiEYPr06Tz22GMIISgvL7dLZaMlGAwG4uLiXL6B5KBBg+jXrx9lZWWUlJS4nPzYFB1NfjRHODuC/Oimdbh/80BQUBAvvviinLDZXhJEamoqoaGhNh/+W11dzdmzZ7ly5UqbzzWrnxK1sTb/yaDQsyuh8fwne0VpVr76cz5565f4epq+vhV6Ay+vfBkhBCNHjpT3S0lJYd26dZSXl9c7PiQkhKlTp8oP/Bs3brT4O9fr9W2WH52JoqIiqqur5WT8tmBuA7Jnzx727dvXLn8/SUlJVsuPzookSRw5coQ7d+60mzNqSzqi/GgLdcCNa+N2nu4TEBCAEIKysjK2bNli047dlqDT6UhLS7OLgVGpVCxbtkzOGWrrw+03E1RkF9d2CI/XVBJTWD//KS8vD4VCYTcD86Nfvc2Xi2YQ5muSAw2SxLvvvsutW7fq5bH5+fk164x+//33nD9/vkXJNiEhgcjIyA6R32CP8mqFQsH06dN55JFHHC7jabVasrKyOkz1Y3Z2NiEhIYSFhbnsSKm6xMbGMmDAALf86KZD4XaeHqCmpgadTufwLuQxMTF2NTBBQUF4enpSU1PD7t27iYqKavW5PBTw23E+5Jeb1ppd6snfLinNI+jkAZn2NjATP/iUdaeptAAAIABJREFUbXMGMyqkVgpYs2YNX375JfPmzaN79+4888wzzcpSkydP5vnnn5ebFTb2e6+oqKCgoKDD5DdkZGQQGBgoJ33bir59+8pNNuPi4hq0jrAXZvmxI0goBoOB+Ph4pkyZwlNPPeXyMldZWRllZWXy7ElXx17qgBvXw/WtjY3p2rUrL7zwglxNZcueSU1RWlpKeXm5QwyM0WiUE33bQkSA4PEwf66k+HDgTgDHkhSsvd+bMzU1le7du9vdwCh9/ei25A2+nBbJBHVt5+vjx4/zzjvvMG/evAYz9B4srRdCyP2N4uLi2LBhQ4NZY+byamdJjG4Ler2epKQku5ZXGwwGLl68yOXLl+12DTNFRUXo9XqbyI/OQFJSEhEREbL8qNVqOXr0KEVFRe28MuvpaPPr7KkOuHE93M5TI5jfYBMTE1mzZg2ZmZl2u5ajDYyXlxeLFy+W+/rk5+e3uhfUDx9SMibEBwnTuv9yAa6mmwyMowZk+j86l6K+s/iJUcHsOhGmqKgoFixYwN27d+vtf/bsWXbv3t1ohMnf37/BSA97y4+OJj4+nsjISLuOWVEqlTz99NPyyCB74eqtCR7ELD/WrX40Go2kpKSQ1cw8R2clOzsbtVpt0ew+V8De6oAb16JdnSchxItCiFQhhEYIcUYIMUiY+JsQolQIkSWEeLXO/pFCiPNCCK0Q4ooQwq6lNeHh4YwcOdKub7XZ2dn4+vo61MCYHzRarZZt27a1aajw21Ng1P2PR2+EVUckuvUZ5FADM+AXv6DL4ME85enJcg8PlPfvLycnh4ULF3Ls2DHAlN90+vRp7t6922irgrCwMBYsWICHhwd6vZ6oqKgOld+g0WgoLCx0iPzo5+eHSqXCYDBw8OBBu4xzycjIICgoyObyY3sRHR3NkCFD6smPvr6+vPTSSwwfPrwdV2Y9HaX60Ywj1QE3rkG7OU9CiAHAv4CfA92Ab4CtwEvAE8BQYA7wgRDC3KzoK+AM0B3YCWyx5xrVajUzZ87Ew8ODmpoa4uLibHp+c35DexkYtVrNE088wcSJE1t9Di8lfPwDCPACgUSvUAP7irzRGRxXcaVUqXj4ww8RSiXTPDx4q4sPfl6myIpWq+Wll17i008/rRdBzMjIYPXq1eTm5jZ6zjt37rB582a8vLw6TH5De0Rp9Hq9/OCx9XmTkpI6VPVjU/KjuUghOzu7gaTsrCQmJtKrV68OU/3YkeRHN7ahPSNPjwHfSpJ0RJIkLfC/wCjgeeBDSZIyJUn6HpODtFQI0RsYDfxRkqQy4K9AXyGEQ3rj37x5k0OHDsldcm2B2cCoVCqbndNaBg8eTEBAAJIkcf78+VYl+fbuAv/1uMTjg8sZ37cSpdLI7iT754rVJXD4cAa8/jrB3X15df4gDs4fTHigKSIhSRIffPABR48erdelvLS0lHXr1jXa32vYsGGMGjWKCRMmyOdwZfLy8lAqlQQHBzv0ut7e3ixbtkxu1lpdXW2T88bHx9OvXz+7yo+Ooq782BTV1dXs2rWL8+fPO3BlrUOr1ZKdnd3q5qvORnuoA26cn/Z0nvYCv6jz77FAJTASuFlnexTwECbHKkaSpGoASZKMQOz9n9mdsWPHsmTJEpuNFjAbmMjISJucr61UVVURGxvbanlljF82Sl1t7lSKpop7mpbHt9iSgatWMWjaULxUHgwIUnPoh/0Z07+v/PPt27fz17/+lSeffFJ+m9fpdGzZsoVLly7Vc5Di4uKYMmUKHh4e5ObmsnHjxjY3GW0v2ru82lztWFxczPr16xvkoVmLWX6s2xTVlUlPT29RflSpVCxYsKCe8++sNCY/uiodTX50Yzva7dstSVKuJEmpAEKIxcAB4N+AAKBubLr8/rYHt9f9md1RKBSysc7JyeHYsWNtGrobFRXlVAMy1Wo1y5cvZ+rUqYDJmbI02mI2MG9NVJNZbIoECAHbEjQYHBixEUolXd75hPIK0zW7+Xqx9ZGuPPVYbd+nq1evsmrVKmbOnCm3JwA4duwYBw4cwGAwUFpaikajISwsDKhtX+CqPZ5SUlLo2bNng8pDR+Pv709kZKT8ubaWjpQkrtfrSU5Otkh+NMtgkiTZLIJnazpa9WNiYiK9e/duV3XAjXMi2lOOEEKEAGsxRZ1ekyRpvxAiH3hEkqSE+/v8FFMO1FfAv0uSNKXO8ReADyRJOtLIuV8BfgsEBgYGhuzZs8dm605KSiI+Pp6ZM2e2StM3GAxUVVU5bQ8XnU7Ht99+S0REhEXRCrMhV6lU7CuMpGsfXzzvF751r8gjsibPnsttgEdeJmOvb0WtNiWtV1Tq+UN5GJsPHJYdQm9vb1577TUUCkW9BpldunShb9+++Pv71+sPZW7vIEkS+fn5hIaGOvSeWoskSWg0GqdMqi4oKCA4ONgqJ8jch62j5KFVVVWhUCgsdswlSeLixYsIIZg8ebLTOZAajQYfHx+neSlsC5IkUVFRgZ+fX8s7N8LMmTOvS5I0zsbLcuMktJvzJITwBa4DF4FfSpKkub/9HPA3SZL23P/3R0A+sB24CoRIkmQUQiiAPGCUJEnN9hIYN26cdO3aNZuuX6fT4eXlhSRJ6PV6q4zfuXPnGDNmTKv/KO2NeY39+/dvsbpEq9Vy5coVpk+fjkKhoKoGXvmmkqHhlffPBa8OCyJI1br5aa2l/NIp1Af/jsf9ES5llXBhygu88du35So7IQS///3v6d69O3fu3AFMA5WnTZvGtGnTGj1vbGwshw4d4qmnnnIaybU5bt26Rbdu3ZyuSiglJYXdu3czb948i5O+jUYjZ8+eZeLEie0eRbMFGo2GGzduMG3aNKucoFu3bqFUKp0ugTktLQ2NRtNhqlOvXbtGr169Wh1FE0K4nacOTHu+HiwHsiVJesnsON1nE/CWEKKrEGLi/f12SJKUCMQDvxZCqIH3gLstOU72wuwsXb58mc2bNzc6gLYx0tPTCQ4OdlrHCWoHvpofuAkJCWg0mkb3fVB+9PaA54eoyS9X3j8X7EzUODzh2n/yY5SPXohkNF03wAcmXt7Bnt275PuSJIk//elPXLx4kRkzZiCEYODAgXKSeGMMHDiQuXPn2mzYsT0pLS2loqKizTKZPejbty+zZ8+2quGgs8iPtqK1FVwPP/wwI0aMcCrHySw/Dhw4sL2XYhMKCwsxGAwdRn50Y3va03kaCTwqhKip+x9wBLgFpGFqR/Ars4QHvAA8DRQBjwMvtsO66xEWFmZxSa4rGhhzh+MLFy40+JnZwDwoYc3qJ9Bo/Lnvt1Co03On0PE5GkFLVlDYbZT8764qLd33fc6hQ4fkIdAAmzdv5p///CfTpk1j9OjRzeY3KBQKOd+msrKS3bt3U1JSYtf7aA3OXl4thGDEiBGmaGVVFVevXm3Wwa6uriYjI8NhzVftTW5uLp6ennJ3+9aQmJjIt99+a8NVtZ64uLhOVf3oxk17JoyvkiRJIUmSxwP/pUmS9LokSf6SJPWSJGlznWPiJUmaKEmSWpKk6ZIkpbXX+s306dOHxx9/XH6YNtfKIC4ujv79+7uUgVGr1SxdupQZM2YAtSX7zRkYIeCdRzyIyqqNEBzNqKBC7/ghpyG/+AMF1HYHD9FnIh3cyvbt21m8eLG8/dKlS/z6179Gr69fIZiTk8M333zTaHFASUkJ+fn5VFVV2e8GWklWVhZ+fn71kuKdlfj4eC5cuNBkzy0wdXceOHBgs3MKXQXz7Edzl//Wkp+fT0ZGRrsnj5eXl1NcXNyhqh+dXR1w0/64flafE3Hs2DF27drV6OgPs4ExD051Jbp16yZ3i963bx93794lLS2tWQPTNxBGB/tQVmX6ipVoFeRUtM+E+OB3/0Wx1uSwJt7N4+pfPkZz4wZ///vfeeedd+T98vLyWLhwofw2r9Fo2LJlC1euXGHTpk0NpNmwsDBWrlwpz0EsLCx00B01T01NDfHx8XadX2dLRo4cyYsvvih/jg9SUlJCZWUlPXv2dPDK7ENycrJN5McJEyawfPnydq8E64jVj66kDrhpH9zOkw2ZOXMmP/jBDxqMJjFLKK5uYAwGg1ztlJKS0qKBeX2cIPqeH9fS1Gy9Fsi6G+0zE0p4euH/9idEJ9UQczMHqaaGa6+/jiY5mVWrVvG3v/1NfgCVl5fzwgsvsHbtWq5fvy5X4qWkpLB27doGDpL5d52ZmcmGDRuIiYlx7M01QmJiIn369Gn3h6o1mOWr9PR09u/fL7+AmCOczio/WktVVRX37t2zifyoVCpRKpXU1NSQkJDQ8gF2wBbyozPhiuqAm/bB7TzZkC5dushzwxITE+UHaW5uLl5eXi5vYMxDhT09Penfvz+lpaWNRtnMqD3h52O8+C7NF4Mk2PQ9fN+0MmNXPLqGEPnhF6ju52fpy8q4+vLLVBUU0LNnT7Zt2yZHNoxGI++//z6HDx9m+vTp8jkKCwtZs2YNycnJDc7fo0cPpk6d2u45OZWVleTm5rpEQntjlJaWUlxcLMunmZmZ+Pn5ERDgkHZudsce8uPNmzfZv38/BQUFNjunJdhKfnQWXFkdcON4OoXz5OhKL0mSuH37Nrdu3UKv1xMbG9thDIxGo6GkpIRu3bqxfft2Tpw40ez+s/vBjPtzaCXgvdPIieSORt2zJxO++ALlfbmkMiOd9LdXEuzrw/jx4zl8+LA8RgRg48aNfPrpp/WiiVVVVWzcuJEHW18olUomTJiAp6cnBoOB48eP1+sf5SiioqJcurvziBEjeP7551Gr1ej1emJiYjpMd+eSkhK0Wq3N5ceHH36Yp59+mpCQkJZ3tiHJycmEhYV1iOrHjqIOuHEcrmlhraS4uNjiVgK2QAjBggULWLhwIampqXTv3r1DDcgcNmwYarWaxx9/vMWhwkLAH2aYBggD3MyRWPd9FTmVTUes7EngiBGM+egjlJ5KJszoy8BeCnrs/AjJaCQ0NJQdO3awcOFCef8LFy7w5ptv8sQTT8iNJiVJ4tChQxw5cgSjsWEeV2FhIXFxca2aE9gWCgoKMN6/D1fGHJXZvHkzUVFRHeJhZs/qR09PTzlZW6fT2fTcTWGWH/v37++Q69mbjqIOuHEcncJ58vLyYtu2bTaf7N4cHh4eCCG4d+8eiYmJHD9+3OWHy+bm5qJSqWQDM2jQIIKCggC4ePEi6enpjR7XNxBeGQN+KgNPDi+jwKDhQGq5Q0e31KXHrFmMe2053SNMUlCwqpLC//41YKou/Pjjj/nd734n75+amspPfvITRowYUS9qcPXqVTZv3tyg2i40NJSVK1fKzR8rKyvtfUtIkkR0dDTDhw+3+7UcQWVlJZ6enowdO7ZBDqErkpmZSUBAgF3lx9TUVL744guHyHcdrfqxI6kDbhxDp3Ce/Pz8GDp0KNu2bXPocNfo6GgGDRpE165dCQwMdOk3aIPB0KSB0el0xMfHk5SU1OTxq8ZDTz/o2cWUy1JYbeC7PMdFAx9E9cIbpGtqo4Eh1WkUrv1fwBQ5fOONN/j888/liGFZWRkrV67Ew8Oj3meQlJQkdyevi1nKKCoqYt26dY3uY0tSU1MJCQlx2pE/1hIVFcWsWbPk7tsajaZBGwlXwZzQbe/qx+7duzuk15I5kt9Rqh+TkpI6jPzoxnF0CucJTGW9EyZMYNu2beTl2X/WWnFxMdXV1fTs2ZMpU6bIXasLCwsdEomwNcnJyYSHhzcqP3p5efHss8/y6KOmAbyVlZUN5Cy1J7w1Scm1tNqZZGezKimubv1w5dYiSRLRMTH4/fYjSrS1UY2g1POUfrNT/ve8efPYs2ePXEJvMBh47733uHXrljxAefjw4YwfP77JawUEBDBs2DC7JnDrdDpSU1M7THl1fn4+kiTRrVs3wOR8bNu2jW+++aadV9Y6EhIS6Nu3r90HS6vVan74wx/atbdXR6x+zMzM7DDyoxvH0WmcJzD1k5k5cya7du0iM9N+U13qNpCsa2CMRiN79+7lwIEDdru2PTAbmH79+jW5j0qlQqFQoNfr2b59O8eOHWuwz5z+0EWpJl9jCvUbgW/SHT+6JScnxyQ/du+B72//habS5OgplAL12a+ouH1V3nfkyJEcOnSIUaNqO5V/+eWXrF69mvnz5zN//vxmHyIeHh7MnDmTgIAAJEni2rVr8mw9WxEXF8dDDz3UIeQto9FIdHR0vearHh4eTJo0iXHjXG9MWEVFBXl5eXIVrqOuefr0abvkP927d8/u8qMjiYmJYdCgQR1CfnTjWDqV8wSmPJ25c+eyb9++RkvObcG9e/fo0qVLAwOjUCiYO3cujz32mF2uay/M8qMlBsbT05NRo0Y1WiElBPzhUcHFJD+54i5No+dukeM6JBsMBuLi4mTpzTMkFOVLH1B9PwLm5aVEbPoAXXaGfEyPHj3YtWsXP/rRj+Rt586d44033iAjI6PB+b/77rtGE8lLSkq4cOECUVFRNrufsrIySktLiYiIsNk525O0tDRCQ0MbyI/Dhg2TZaKMjIxGP19n5MHZj46guLiYmzdvkpWVZdPz1tTUkJiY6DLNV1vCLD821ZzVjZvm6HTOE0BkZCQLFizg6NGjNm9qaDYwTU2KDw8Pl+WI69evk5KSYtPr2xqz/GiNgRk9erT8ph0bG1uv6iwyCBYN9OT7zFr570Sm40a3NCY/qgePQDfnVQw1pjX4+CjRfvQGhsraCJFarebTTz/lN7/5jbwtKSmJH/3oR5w/fx4wRRwPHz7M4cOH2bZtW4OxGUFBQfz4xz+WZb7memRZQlMRTlfFLD82Nyy4sLCQHTt2NGgV4YyYRzWZ/94dRUREBK+88orNpWJHyY+OoKPJj24cT6d0nsDkxCxZsoSzZ89y69Ytm503Pj7eIgNTU1NDVFQUsbGxNru2rWlrebXBYODSpUtcvny53vZV4yGjwFce3VJtkDiVaVspqzGakx/9Z/yQ0iFz5H938ZEo/OC1epKiEIJf//rXfPLJJ7LzVVJSwvLly/n666+Ji4vjxo0bgOl7sG7dugZDg7t27YoQAq1Wy9dff92mRPKcnBy8vb3likdXJzY2lgEDBjQrPwYHB/Pkk0/W68fljJjlx/aqfjRH7mxVINMe8qM9aUodcOPGUjqt8wSmN8KlS5dy7do1Ll++3Obcm4qKCgoKCix64/Pw8GDp0qXMmjULMM1UcrZWBhkZGQQGBsr9jaxFqVSydOlS5s6dC5gcRkmS8PGEd6cJzibUzsWLKq4mucy+PWqio6MZPHhwk/Jj1+WvUeBvSrrWlFbx/aGrJHz8cYP9FixYwK5du+jevTtguq9/+7d/Y/PmzUyePFneLy8vj9WrVzfawkGpVNKtWzeCg4NbdS9m+bGjNJAsKyujrKyM8PDwFvcdNGgQnp6e1NTUcPv2baf7uwFT9WNoaCg+Pj4t72wnUlJSWL9+vU2i2+YIp6s2X61LS+qAGzeW4Pp/CW0kMDCQZcuWERsby5kzZ9pkiO/evcvQoUMtjtKoVCo8PDzQ6/Xs2LGDs2fPtvratkav15OUlNTm/AYfHx/UarXcWPLYsWNIksTc/hDp70V8Xu38tbPZlXZ7EBYVFVFdXS07PE0R/OZ/kVkTyvlvkqjU6Ij7v/8jc//+Bvs9/PDDHDp0iBEjRsjb1q5dy4YNG5g9e7bsoFVWVvLVV19x+/btesd7eXkxb9482VlITExs0C+qOZKSkpqsfnQ1WhvhjI2N5cSJEzbP7WkrOp2OtLS0ZuVHR9CrVy8mT57c5pYC+fn5CCEc3sHcXliqDrhx0xyd3nkCUx+opUuXkp2dzdGjR1uVjJqXl4dCoWiVgfHw8CAiIoKwsDCrj7UXCQkJREZG2rRnTPfu3QkNDUUIIXcev5rii1YviMtVodIH2CX/wJr8BqFQ0PM/PqbLuNrO6bfefpuiRnJsevbsye7du3nyySflbadPn+add95h9uzZctTBYDCwd+9eTpw40ahzWFFRwaFDh7hw4YJF96PVasnKyuow5dXZ2dmo1WoCAwOtOm7YsGE8++yzFkWrHIkl8qMj8PDwYPLkyW1ysBurfnRlrFEH3LhpDrfzdB+1Ws3ixYvRaDQcOHDAqmRe84DM1hoYIQTTp0+X31QzMzMdOk7mQTQaDQUFBTbNbxBCMGnSJDlXJTs7my7GQl4YoWDbtSBOxfnzPxcVlFoefLGYjIwMgoKCLJYfFZ6ejPvXv/C7P+TXqNOR8O9vUHn3RoN9fXx8+Oyzz/jVr34lb0tISGDFihWMHTu23qiUCxcusH379gYl5L6+vixZskTuHdVS9M0sP3YECcVgMBAfH98q+VEIIb9w5OXlcfXq1RaOsD/WyI+OoqioiD179qDRaKw+1hnkR1tirTrgxk1TuL71tSFeXl4sWrQIhULB7t27G1RLNUVKSgrdu3e3iYGprq5m7969nDp1qs3nai32ruCSJInjx49z+PBhXh8vEaQyfQ0LtfDXyy0cbCVm+dHa/AbPgAAmrlmDV9euhPcNZNzEEKSv/oAut6FEpFAo+O1vf8vHH3+MSmWSIUtKSlixYgX+/v715JvY2FgOHTrU4Bzh4eGoVCo5SnX37t1G11VUVIRer29RfnQVkpKSiIiIaLP8GBUVxa1bt9r1pcOe8+vaghCC3NxcCgsLrTrOWeRHW9EWdcCNmwfpFM6TNU0JlUolTz75JEFBQezYsaNFY1xdXU16ejoP3Y9StBWVSsWCBQuYMWOGTc5nLXl5eSiVylYnMluCEIKFCxfy5JNP4usl+P202kjL13cgKh80NmpdEB8f32r50adXLyb83/8wanIESqUCXx8FlX/9BUZt4x3iFy1axM6dO+t1xn777bdJSEhg0qRJgCnK1Fyfr5qamiajnh2tNYFZfmyu+aqlPProoyxfvrxdR2xkZ2fj4+Njtfxob4KCgvjpT39qdSQ5JibGKeRHW9BWdcCNmwfpFM6TtREhhULBE088QZ8+fdi6dStlZWVN7hsbG8vAgQNtamAiIiLw9fVFkiS+/fbbBo0Y7YXZwDhiQGZAQIA8YDgg9wLD/IoA8FAY2RCj4YvoYrQ1bXOgNBoNhYWFbZIfg6bMoHRArbMT6GOk5A8rMeoaj0qOGTOGQ4cO1fsMv/jiC7Zs2cKcOXNYtmxZs+MzVCoVS5YskUvc8/LyZJnPWvnR2YmOjmbIkCE2kR8VCoX8N3Pp0iVu3rxpgxVajrNXPyqVSiRJIjU11aKcztLSUsrLy51KfmwLqampNlMH3LiBTuI8CSEwGAycP3/e4oomIQTTpk1j+PDhbNu2rdGQd2lpKRqNxm6J3lVVVaSlpXHv3j27nP9BbCk/WookSRgMNSwNjUYhJGYPLaerfxU6o8T1/LYlQNkqStP1xTco9K1Nzu7qXUXxf65E0jfeWiE8PJy9e/fKLRoAjh8/zvvvv9+oo5Cbm9ugnxSYopo7d+7k+PHjrZYfnRV7yY+SJJGXl9fgM7U3iYmJ9OrVS5ZtnZGMjAx27drVYm85Z5UfW0t1dTVpaWk2UwfcuIFO4jyBqaHgd999R2pqqlXHjR8/nsmTJ7Njx4568/AcYWDUajXPPvusLPlUVtqvlN/W8qOlCCGYMWMGL8ybwrwBgric2ofPtXwtOkPr7tfW8mPX33xIoaL2QR/sVUHRH19BMjQusfn6+rJ69Wpee+01eVtUVBQ/+tGP6jXGTElJ4YsvvuDAgQMYDPWHJKtUKubOncuUKVOIj4+nX79+Nq1+bC/qVj/aGoVCwbx585gzZw5CCIc4UFqtluzsbJvIj/akV69ezJs3r0UHPDs7G19fX6eTH1uLPdQBN246jfMUHh7OihUr5L5F1hjV4cOHM2fOHPbt20diYiIAWVlZ+Pr62nWCOZgeoOaO1Js2beLcuXN2uU57GxghBKvGQ2K+ilKt6WtZZZC4VWh99Mke5dVCoaDr7z+hUOoqbwtWllL4p1eRmpBBFAoF7777Lv/7v/8rf645OTk89dRTfPPNNxQVFbFt2zaMRiM3b95k06ZNDSKj/fr1w8PDg8LCQjIyMpx+nI8lpKen07VrV/z8/FreuRUolUqEEFRWVrJ9+/Z644HsgS3lR3sihGhxRqWzy4/WYm91wE3nxbn/2m2M+U2qsLCQDRs2yLOnLCEyMpJFixZx4sQJbt682ery6tbi7e3NsGHDGDhwoM3P7SwGZlAIzOhVw617tUm/V/O01Bitix6kpKTQs2dPmycPC4WCru9/TlFN7UiHEAop/PPrTTpQAM8++yybNm2SHW2tVsvLL7/Mtm3b6kUBUlJSWLt2bYORGlFRUQwcOJDk5GSrI6fOhl6vJzk52S7f4weRJAmtVmvXCrzCwkKXq37MzMxk06ZNjX4uiYmJ9O7d26nlR0sxqwMdpcDCjXPRqZynunh4eFgdZenZsydLly7lm2++oaCgwKEdaoUQPPLII/KA3ujoaMrLy9t8XmfLb/jVI57E5XhTUW1ai0Zv5PtCyx9+9pYfhdKDoP9YTZGuNi8sxJBD+kcfNHvc1KlT2b9/v9ycT5IkPvjgAy5evMj06dPl/QoKCli7dq2c55abm4tSqaRHjx4sW7aMRx99FLCvhGtPHCk/+vr68sILL8hymq0/L3vKj/ZEpVJRU1PToO+TWX6MjIxsp5XZFrM60FHkRzfORad0noKDg1m+fLk8ULWoqMjiY1UqFaNHj6aqqorjx4+3qht5W9FqtZw6darBwN3WkJWVhZ+fn93lR0t5uAdMjhDcyayNGl3Jq8Jo4YMvJiamRWmirQhPLwLfX0NxlQpJkrh96R53PtlA4uefN3vcQw89xIEDB5gwYYK8bfPmzXz22WfMmTNHXnNFRQUbNmzg7t279cqrvby8UCgU6PV6tm/fzrFjx+x2j/ZAo9FQVFRE7969HXZNs5SWmJjItm3bLO7dZgnp6ekEBwfbTX60FyEhIbzwwgtySw0zUVFRLiE/WkJbmq8h5CPSAAAgAElEQVS6cWMJrv9X0krMUZbo6Gg2bNhQLxm8OaKjoxkzZgxLly6lvLycffv2odfr7bnUBpgTyc29oFr7Rl1TU0N8fHyb59fZmlUTICrbm2q96XdUojMSVVjZ4oT4kpISKioq2jzLyxIU3moC3ltDbJ4f6Ykm5zvmww9J/vLLZo/r2rUrW7du5amnnpK3nT9/nrfffptZs2bJUmNNTQ27du2ipKSkQQNJDw8PRowY4ZCWErakPSUUs0Ngq5cdR8qP9sBcgWyWgQsLCzEYDC4lPzZHR5If3TgnndZ5MtO/f3+Lh2cWFBRgMBgIDQ3Fy8uLhQsX4u3tbVEzTVsTHBwsT5bfuXMn33//vdXncFYDMykcRoYq+D6r1mk4lVLIps2bm/ycrZlfZyuUvn4M/GgNwferIQGi/vQn0rZsafY4lUrFP/7xD9566y15W2JiIi+//DITJkyoVyF4+/ZtDh8+XO94IQRjx46lV69eAMTFxVndPdrR5Obm4unpKff2cjT9+vVj6dKl8pDqtkp4cXFxLl/9eO3aNXbv3k1RUZHc1qMjUFlZ2aHkRzfOSad3nlQqFZMmTUKhUFBVVcWNGzcaNaySJDWo4FIqlcydO5eIiAi2bNlCaWmpI5cOmN6kW5O/VVlZSW5urlMaGCHg9fHwfaYa/f3qfa2HD8OmzWoyCby95EeltzcTvviCoDFjAPDwVOB3dh1F6/7a7HFCCN58800++eQT2XktKipixYoVhIaG1nOgmnPsa2pqOHPmDOfPn7fB3dgHRzZfbQ4hBEajkSNHjnD27NlWn6e8vJzi4mKHyo/2YOTIkSxcuJCysjKXlB+bIjo6mqFDh3YI+dGN8+L+dtXh7t27nDlzptG3+LS0tEYNjHmo7+jRo9myZQt5eXmOWi6AHAEza/u5ubkWDTV29vyGx/pCvyAF0dneFGiUCJ0/j44wzdjKzc3l1KlTcl8ks/zYXvkNHr6+TFy3juCHRzDpiX4Ed/cjKPUcRRs/bvHYBQsWsH37dtlZ0ul0vPnmmyQmJjJixAimTJnCmPuOWaPX9vBg2bJlzJkzB7CdLGVLkpOT7VL92BqEEHh7e+Pt7d3q6FNHGZGjVqvp1asXKSkpLis/PohZfqw7kNuNG3vgnE/OdmLs2LE8//zz8uBIs3HV6XQtGpjRo0fz2GOPsWvXLtLS0hyyXjNmI67VatmxY0eLQ4ULCgowGo1ObWCEgFXj4UqqLztuBPL1DRXlOtN9pqenk5iYKCf/JiYm0qdPH4dWPz6Ip78/4/71MZ4+JqlRKASB8Sco2bGmxWPHjRvHwYMH632/duzYwcmTJ5kyZUqD/R986AcEBODt7Y3RaGT//v1cuHChjXdjO6qqqrh3757TdHcWQjBz5kwmTpwo5/1YQ05OTrvKj7YmLi6OmpoaDhw44JLVm3WpO/vRjRt743ae6iCEkCtQ0tLS2LNnD9XV1cTFxdG/f/8W8xsGDhzIvHnzOHz4MDExMY5Ycj3UajVz5syRO5I3hll+dIXy6rn9oW8XAQjKdfD1bdP28ePH88ILL+Dj40NFRQUpKSlyC4D2xKtHBJ6rPqKi0hT9USgE/ncOULJvY4vH9u7dm3379tWrxNu7dy/Lli2rFwmtrq5m7dq1REdHN3oeHx8ffH1923gntiMmJoaBAwfatfrRWswvG0VFRaxfv97ilx2j0UhsbGy7y4+2wiw/9ujRA6PRaNNKxPagKXXAjRt74HaemqCyspKKigrKysooKSmRk3NbolevXixZsoSzZ89y7do1O6+yIQMGDCAgIABJkjh79myDKsK0tDRCQkKc6gHbFEoFvDau9t9rb0Hl/cJGcwXa7t27uX37drvkmzWGqnc/FD/7byq1poiGUqnA/9oOSo/saPFYHx8fXnvtNZ577jl527Vr15g3bx7x8fEYjUZ27txJZmYmO3bs4Pz58/WiBQqFglmzZvHwww8DJnmzsrLSxndoOSUlJWi1WodUP7YGHx8fgoKCLJYTk5OTCQsLcwr5sa3UbSA5fPhwlixZ0qCq05XQ6/UdSn504/y4nacmGDJkCM899xyJiYkMGTLEqodQt27dePbZZ7l79y6nT59ul3B4VVUViYmJJCcny9vM8uOAAQMcvp7WsmAQRPib/r9IK/HlXR0Xcky/i4KCAsLDw5k2bZpTNcJT9x+MeOnPaOs4UL4XNlJ28kCzxyUkJNC/f38+/PBD/uM//kOOkKSnp7NgwQJOnjxZr13DyZMnG8zEMx9jMBjYt29fg0o9R+FszVcbw9vbm8WLF8vydXMtR8zyY//+/Zvcx5XIzc3Fy8uLrl271pP9LW3Z4mzExsZapA64cWMr3M5TM+Tn56NSqYiKiuLrr7+2qh1BQEAAS5cuJScnh0OHDlmUxG1L1Go1zz33HFOnTgVMhjE2NpaHHnrIpQyMpxJ+NhY8FBJPjymhjDLOZ1dSVFVDVFQUEydOZMqUKQghqKio4OzZsw7/rBtDPWgE0vL3qaoyOTYeHgrUp9agOdd4Y8vKykry8vLo27cvQgheeeUV1q1bh4+PqZN5WVkZK1euxM/Pjz59+sjHNTUTT6lUMm/ePB577DE73WHzZGZmEhAQQEBAQMs7OwE3b95k48aNDT5HM45ovuoompIfjxw5wsGDB52y6KA5ysvLrVIH3LixBW7nqQkMBoNsYIYOHcqYMWOsDter1WqWLFmCJEns3LnT4b2gvL29EUJQXV3NunXrOHv2LBEREQ5dgy14ZhgEeQsqdaavqwScSMqjW7du9eTHlJQUbt68SUlJSTuttD4+I8dhWPI21dUmB8rTU4HXkY+puNqwTD4qKqpBefXs2bPZs2ePPJLHYDDw7rvvkpaWxogRI+T9mpqJFxYWJic2X7p0iYSEBJvfY2PU1NSQkJDgdM1Xm6Nbt2706NGjUeeouLgYrVYr/x5cnaSkpEblx6lTp7J48WKnrcBtDPf8Ojfthev8lbQBnU5n9THJycmEh4fj7e1Nt27d5ETe4uJi4uPjLT6Ph4cH8+bNIywsjC1btrTYJdseeHp6olKpeOyxx1zSwHh7wE/HwI2M2nlyKdUe9IysX8E1fPhwVq5cKVdLOtpZbQzfcVPQ/ehX6HQmB0qr0XH1N2+jSUmR98nPz0eSpAbjMsB0TwcPHqyX4P/FF19w4MCBepV4D87Eq0tNTQ0pKSkOGyqckJBA375927X60VoiIiL4wQ9+gKenZ73IS3s0X7UnVVVVZGZmNio/hoaGyn87rkJubi4qlarDVD+6cR06hfNUWlpKXFycxfubDYx5oGhdrly5wokTJ5oM7zeGuRfUuHHj2LZtm8PzCnJzcxkzZowcpo+OjrZqnp8zsHwEVFZ5kF1qagYqCQU3Chs6xeZKm+TkZFavXu0UORz+jzxG1eyfk59byaXjyWiy8rj0/PNUZmRgNBobNF99kJ49e7Jnzx65lxPAsWPH+PDDD5k5c2aDmXgPVuJ5eHjw9NNP8/jjjwOm77e98vAqKirIy8urJy26ElVVVWzfvp07d+4AcO/ePZeSH1siOjq6WfnRYDBw6tQp+f6dmbrqgBs3jqZTOE9BQUF8++233Lp1y6L9o6OjGTx4cKMG5oknnuCZZ56RK1OseQiNHDmSOXPmsH//fqucubZgMBiIi4uTDYxOp+Ps2bNcunTJIde3FX5e8NLDgpt1ok+3CqvQ1jSenxEaGsrgwYOdppdVwIwfoFj5Z4xKUzSmKieHSz/+MYnXrhEaGtpi9aOPjw+rV6/m1VdflbdFRUXx5ptvMnXq1Hoz8a5fv97ge+np6VlvqPCJEydsfIe1a3Ll7s5eXl6o1WpUKhU1NTUkJia6lPzYHMXFxVRXVzcrPyoUCgoKCigrK3PgylpHXXXAjRtH45oWzkrMXZivX7/OxYsXm3V4ioqKqK6ubnJApoeHhxzavnPnjtXJ4JGRkSxevJjTp0/z3Xff2b0SLykpqZ6B8fLy4tlnn+WJJ54AsLpJYHvy4iiJ3BIFhRqTU6s3wvX8xiOAfn5+zJ49W57/d/HixVbJt7YkeNw4xn/+OYr7clZlRgalf36bPp6WyUFKpZL33nuPDz/8UB7Hk5OTw8svv8ygQYMIDg4mJCSEp59+ukmJycPDg0GDBtmlaWV+fj5Ao/Kjq6BQKFiwYAGDBg0iPj6+3Zuv2oq6DSSbkx+FECxZskQuNHFWmlMH3LhxBJ3CeQIIDAxk2bJlJCUlceLEiUYrSqzNb6iurkan01mdCxEaGspzzz1HdHR0k2uxBVqtlqysrAYGpkuXLqhUKgwGA3v27GnTnC9HUlWSww/CCrmZUZvoei1fi87QvAOakZHB5cuXycjIsPcSW6TblCmM++QThKcnEf2CGDeqC9Lqt6m+l2rxOZYvX87GjRtlKUmr1fL666+jUCh47rnnmn0TF0IwceJEeaZhcnIyGo2mTfcEyPKjKzRftYSKigru3LnDpUuXXL55JJjkxy5dulgkP5qjhkVFRU5RudoYLcmPbtzYm07jPAH4+vryzDPPUFxczMGDBxsYhoyMDAIDA/H397fofOPHj2fRokUolUqqq6utegj5+/uzbNkyysrK5E7mtiYmJqZJ+RFMD9KgoCCXSLY0y49vzexCRpGKUq3pq1tlkLhV2Hz+WWRkJCtWrJCTZNv7Ydh95kyG/+e7jJwUjlAIfNUK9P/4FfrcLIvPMW3aNPbv3y/nFkmSxJ///Gf+67/+q0G/omvXrjVaqFBdXc0333zD6dOn23Q/AKmpqYSGhsqtFVydqKgoBg8ejBCi2f5ProBZfhw0aJDFxxQWFvLll19y9+5dO66sdZjVgY5S/ejGNelUzhOASqXiqaeeAkzdqc0PUr1eT1JSktX5Deao09GjR9m6datVb2oqlYqFCxfi7+/Ptm3bKC8vt+razdGS/AimN8zHH39cjhZkZmY6ba5DUlISERERRHT15plhglv3aqNPV/O01Bibjz4FBQUBpofC2rVrSUxMtOt6m0OSJLJ69ye31xSk++v28xFU/e11agotHyw9YMAADhw4wPjx4+VtmzZt4sc//rHccd0sLTdWiadSqViyZImcSN5adDodaWlpLtV8tTny8/MRQvDwww/z7LPPuvy4j/j4eKurH7t27cqMGTOc7nfa0aof3bgunc55gtr2AUFBQWzfvp2Kigri4+OJjIxsdQNJc7NGcy6KpSiVSmbNmsWQIUPYsmULeXmWPzybojUGxmAwcPjwYY4ePdrm69uaB+XHV8dCUp43FdUCgxH8lZ7oW3CezPj4+NCnT59mnUp7k52djVqtJuyV31HY6xE5781fDZUf/pyaUsvbWQQHB7N161YWLVokbzt37hwLFiwgPj6eQ4cOAU1X4oWGhqJWq5EkiSNHjljVhsNMbGwsAwYMsPq774w8WP1oHh587Ngxp4zCtERFRQUFBQVWz34UQjBmzBinG+NkrTrgxo296JTOE5iiLk888QT9+/dnw4YNpKWltam8unv37gwZMgQwRXCsaUgohGD8+PFMnz6dXbt2kVKnB1BryMjIICgoyCoDo1QqWbRoEbNmzWrTte2BufrRnIsRHgDzBwlOxvmz+bsgTsb5o/aw7KusVqt58skn8ff3R5Ikbty4YVXbibZiMBiIj4+XvyshP/0dhd1Gyz8PUBup+H8/xaCxPALo7e3NP//5T9566y15W0JCAosXL2bs2LH1KvEam4kHpuhRSUmJ1TMCy8rKKC8vJzw83KrjnJXU1FS6d+9eT34UQsj36WrcvXuXoUOHtjpKU1hYyLFjx5wi96m16oAbN/ag0zpPYDKKjzzyCAEBAcTGxsrVQm3l6tWrXLhwwepKtsGDBzN//nyOHj3a6j4rZgNjTX6DmZCQEHlG3JkzZ7hy5Uq7zOWrS1FREXq9vkF+w2vjILvUC021kjNpcM3ydCGZwsJCzpw549CIgll+rJvUHbLqP8gPGCL/u4vagOaDlVY5UEII3nzzTT7++GNZnikqKuLnP/+5XIln5uTJkxw9erTe71alUvHMM8/8f/bePKzJO93/f3+yswrIKggCgqiIFRdcqbt1A3cEnXZmnM5MZzrfdqZzOl2m82vnzNbT0znzPdfpnPm21VoVBVxQcAVXFARRUJElEFllCbIGkkC25/dHzCORLcnzBFDzui6vXkmTJ5+Q5PPcz/2+7/dNS4BdXV3DfvYvmruzQX58thORw+Fgy5YtWLBgwSitzDKam5vB4XAYGV92dXWhvLwcLS0tLK7MMioqKhipAzZssMlLHTwB+g0mLCwM69evx/Hjx1npyNqwYQO2bt0KLpcLiqLMCkB8fX2xY8cO5OfnIysry+zghan8COili+7ubsjl8lE9KfZtr36WIFdgQ59yjD/fAMyN89zd3bF7927Mnj0bAKx+dT1Y9yMAePzmL3hs/9T1eZxIY3YABQCbN29GSkoK3QTQ09ODX/3qV3B1dTWSbvLy8nDy5EmjAN/QWKBUKpGYmIhr164N+VqNjY2wt7cfU0OZmVBaWjqo/GjIera2tuLUqVOj3nQwHDqdDqWlpUOar5pCQEAA3nzzzVEvzu7u7kZLS8tza75q48XjpQ6e+tY3TJkyBevWrUN6ejpjA0s+n09LZllZWTh79qxZdgRubm5ISEhAQ0MD0tPTTe726e7uRmtrK+MNhsPhYN26dVi6dCkAvUO7QqFgdExLqK2tHVJ+/LeFAP/JN7igEUir0KFFaV4A5OHhAUIIlEolDh48iKKiIqbLHpSSkhJMnTp1UANJ93/7As3CSfRtZ6EaFb/7JbRmnqjnzp2LU6dOwd/fH4BeKnz//ffR0dFBy4WAvpg8OTm53/dLJBJh9uzZQ9oOPCs/Pu90dnaiu7t7WPmxu7sbTU1NY7axwkBVVVU/+dESCCEQCoUARnfckSkeVTZsjCQvdfBUVVUFHx8fuiYkICAA27Ztw9WrV3Hnzh3Gx6coCnZ2drCzszP7R28YKszj8ZCcnGySDQKbGwwhBBwOBxRFIT09HcePHx9RCU+tVqOysnJI+dF/HPDGTEDI02F+oBxFsnak13RbtE4ulwsPDw+r2TYY5MehCtUJIfD44O9oFk4CpaNQmF2HivM5uP2LX5gdQAUFBeHUqVNGmYf/+I//QFFRESIjI+n7KioqkJSUZPQ3M9TgGeSekpKSfidOiUSCiRMn0ifW5xlz5MeAgADs2bNnTBuB9vb2ora2llUj1MzMzH7fk5GiubkZXC7XSHq2YWO0eWmDJ8MG8+yATE9PT+zcuRNFRUW4evUqo82CEIJ58+Zh2bJldNFpR0eHyc/n8XhYu3YtQkJCkJiYCKlUOuhjrbXBEEKwYsUKLF26dESv+srLyxEUFDSs/PireYCLEJgxQQkel4JUqUFVl/m+PAKBABs2bKAzD+Xl5axl2/p2Pw6HIYCqcp6D+ir9d6X56lWLAihPT08cP34cCxcupO/bt28fzpw5Q99nMM0c7LOVyWTIyMjA7du36fuUSiUaGxtpo83nncbGRjg4OJgsP/J4PLrZYCzMTnyWsrIyhIaGstr9GBQUhBkzZljN0HcwDPKjbX6djbHGSxs8lZaWDrrBjBs3DnFxcZBKpQOaaZoLIYRuBT9x4oRZG5Dh5LZs2TIcP358wC4+U4bLMsHHxwcTJ04EoO/esfZYGYP8aJCdhsJFBLwZyUFJ09MC7KwGBaP1KZVKnD9/HtnZ2RYfoy8G+dFUvyBCCALf+xghv/wlfV/z1aso/OXPoe0yrxvOyckJhw4dwoYNG+j70tPT8c0332Dp0qWIiYlBaGjooM93dnbGzp07jQKw4eTH5wmD+aq58qNarUZhYSGKi4uttDLLMMiPEyZMYPW4wcHBmDNnzog7erMlP9qwwTbP/+5nAZ2dnZDL5UNuMHZ2dti6dSsA4Pjx44z1fkIIVq1ahdWrV1t00gkNDcWWLVtw+fJl3Lp1yyg4eFZ+tCZ1dXWorq62avBkbgfX6xFAc4cdtE9i0ialBnXdlge8dnZ22LlzJ6KjowGA0dW2KfLjQBBCMOXXv6YDKL6Ai8l2DZD/+SdmB1BCoRD//Oc/8cMf/pC+Lzs7G3/5y18G/A08+9l6e3uDy+VCpVLh0KFDaG5uHlWfLDaRSCTw9/c3W34UCASIi4sbU9YeBvnRWgaSFEVBIpEMmQFnE2vIjzZssMVLFzyZs8EYzDS9vLyQnJxstgfOs7i5ucHPzw+Avj7p9u3bZgUh3t7eSEhIgFgsxoULF6DRaEZ8g3nttdcQGxsLDoeD3t5e1gvJpVIp+Hy+WfKjkAe8M48LsfTpCfByPbN1eXp60vP/Tpw4gVu3bll0HFPlx4EwBFChv3gL81cGwmW8PZxFGsj/Yn4AxeVy8ac//Qnvv/8+fd+DBw+wadMmI1+xrq4ufP3116iuru53jO7ubjx48OCF8XRiKj86OjrSzQbPmo+OBg0NDXBwcMC4ceOscnyNRoOMjAwUFhZa5fjPYg350YYNtnjpgqeGhgY4OjqavMEQQrB06VLMmDEDSUlJrDmAV1dXo7Ky0uwMjpOTE+Li4tDb24tjx46hsLBwRAdkEkJoH6HMzEwcPnyYtRZ/JvUNG0KA3h57GIzGm5Rq1MuZzySjKAr29vYWyQbd3d1oa2szSX4cDEIIQn/zHlQTnkqyzkLLAihCCN555x188cUXdPazpqYGmzZtQlFREXp6epCYmIimpiYcOnQIZWVlRs/v6upCQkICIiIiAIxu9xUbFBcXY9q0aYzlx/z8fGRkZIyqiaZGo7F69yOfz8eOHTtGJNtmLfnRhg22eCmCJ8PJ3bDBWOJQO3v2bLruaKCrcnMghGDdunXYtGkTOBwOVCoVVCqVyc8XCASIiYmBi4sL0tPTzZpZxSZz587F/PnzWbsyZCI/EgL8NoqLiuan2acLNcyzYoai/b7z/0w9SbJlIKkvIv/yGRsDywIoAEhISMDevXtpo86WlhZs3boVV65cgVwuB6CvBUpJSUFBQQGAp/Kj4eTc2NiIb7/9dlRnBDKhtbUVWq0Wnp6ejI+1YMECxMfHj+rIkIcPH1okP5qLu7s77V9nLawtP9qwwQYvRfBkOLlbWt9gIDQ0FBs3bsS5c+cYu1IbMjiGQvJjx46ZXVvj5OSEmJgYHD16lPFIF0vw8vIyCiqercUyh97eXtTV1TGSH+f6Avawo80ym3vVaJIzz4oZNnDD/L/z588P+xyD/MiW9YGhC08qfCox0QGUzPQOTgOrV69GUlISnYGVy+V46623MHHiRHrNBpuKGzduoKysDMHBwbT86ObmhtDQ0OcyMzCU+aol8Pl8ugasvr5+xEeZKBSKEe1+bGhowIEDB6zmdWWuOmDDxmjwUgRPgP7KLD09na45shQ/Pz/s2LEDN2/eRG5uLuMrMEIIpk+fjunTp5slH9TX18PR0RFRUVH0SJeRqkUYCLFYjKKiIrMyaH0xdD8ylR/fi+KhquVpJu5UFXs1Wc/O/xvss7dWezUhBJ4ffNkvgFL89U2LAqi5c+ciNTWVdo9Wq9X49a9/DaFQaOQofenSJeTn5xv9doRCIdasWQN7e3u6kHi0R/mYSk1NDcaPH29y96OpdHR0ICUlBXl5eawedzhKSkpYkR9NxdHREVwu1yrGuUzUARs2RpKXJnjKz88Hn89n5apw/PjxiI+PR0VFBTIzMxl7n0yePBkzZ84EoL/qqq2tHfLxGo0GFRUVtITi6+uLnTt34t69e7h48eKIe7EAwLJlyxAfHw+hUAiKosyqh+no6IBCoYCPjw/jdQS5Aj6ip/VJ9XINFGr2Tup95/9duXJlwBE6lZWVVut+pAMowST6PicGAdSUKVOQlpaGkJCns24+++yzfk71dXV1SEtLG3Beo1gsxqlTp0Yl+2kuarUaVVVVQ9ozWIqLiwvWrl1LzwccCVpaWliTH03F2dkZu3fvtsrIFqbqgA0bI8VLETxpNBoEBgbi7bffhoODg9kn94FwdHTEjh070NXVhVOnTlmccXmWGzdu4NKlS0MGQBKJBAEBAUa1Ti4uLti5cyc6Oztx4sQJ9PT0sLIeUyGE0EXVOTk5OHDggElXptaob/g/c3goaxThstgRB/NccaLUOm3bOp0OOp3OaN09PT149OiRVbsfCSHw/PDvaBI8zUA5CTVo+eNbZhtpAvrg+8SJE/SMPwD46quvcO/ePaNi93v37iElJaXfOJcpU6YgNjb2uTDNFIvFRvIj24SFhUEgEECn01l9hAtFUVb1dxsOjUbDSgONAYVCAalU+lx8j2zYeCmCp56eHoSHh9O1Tzdv3sShQ4cYp52FQiE2bdoEe3t7pKSkmDRCZThiYmLoQvKBhgobNpi+Q14NiEQibN68GW5ubjhy5Aja2toYr8cSQkJCMH36dJMyL/X19XB2doazszNrrz/eHpjn4QixVAQdRfBfuUA3O7EtDSEEK1euxKuvvgpAnz3r6OhgTX405fW9PvySDqDaH8tRcPo27rz9tkUBlJubG5KTk7Fy5Ur6vpSUFBw7dsxIfiwvL+9nHkoIweTJk0EIQVdXFy5dujTidT+m0NXVhfb2dtrw1ZqcO3cOx48ft+rfwVryo6lkZGSw+h6Li4tfGPPVsfj9t8Euz/+31AR4PB4cHBzo20FBQZgyZQorsgqXy8Xq1asxefJkHD58GI8fP2Z0PJFIBFdXVwB6I8MzZ84YSSXDtVdzOBwsX74cs2fPRnJy8qhIKZ6enli8eDE9kiYvL2/AehiD/GiN+oYfzwImPDmntCiBfzEfVTggBvf48+fPIzExEXK5nBX50dTX9vrwS9QLQ5F7qQoatQ7Sy5dx51e/siiAsrOzw969exEXF0ffd/fuXRw4cIDOSk2aNAmLFy8e9Bg1NTUoLS1l7InGNubMr2ODGTNmICoqymoeRSqVymryo6lERkZi3bp1rFwotLS0QKfTjaj8aNGhKCQAACAASURBVE2e1y5UG6bzUgRPz+rn3t7eiI6OBiEE3d3dePjwIaPjE0Iwf/58LFmyBMeOHWMtYBEIBBAIBHSg1NLSAoqiTBpKGhERQReSm2vGySalpaW4devWgBJGRUVFP/mRLUQ84N+eThTBNwVAvcw6fwNCCFavXg1fX19ERESMaHs1IQQTPvgbJv3kZ/R90kuX9AGUBdI0j8fDl19+ibfeeou+r6CgAF988QVeffVV7Ny5c8iAIDw8HD/+8Y9pk9OxUkQulUohFAqtNvj5Wfz9/emMnTVqEK0tP5qCt7c3AgICGH/fDfKjKbMfnwcM6oCNF5uXInga6sedk5ODc+fOsWL4N3XqVDpguXfvHuPjzZs3D6tWrQIhBJ2dncjJyTGrvsHX1xcJCQkoKSnB+fPnRyWVPG/ePLz++ut027GhFkuhUKC5uXlA+ZEtNoUB0z0o+LqosHpaJ74uklvttRQKBYKCguDs7IyioiJcuXJlxAr3CSEIe+89TO4T8AhqCqD499eh7Wy36Hjr16/Hb3/7W/q3I5FI8N577/W7MKAoql9gbKh9KykpwdGjR9FrQRaMTbRaLcrKyqxqIDkYNTU12L9/P6sGml1dXejo6BgR+XE4VCoV8vLyGA1Irqmpgbu7u5E68DxjUAdsvNiMavBECHmbENJICJERQk4TQnyInr8TQjoJIQ2EkJ/3eXwgIeQGIURJCMkjhDDeDZctW4atW7ey1hnl6+uLuLg43LlzB9euXWPFygAAkpKScPfuXbO7UAyDXTUaDWt1WeZACKEDJ7FYjH379qGlpQUPHjywens1hwBvz9cgJkIGXxc1+MIeFDSxH9A8Kz+2tbWhtbWV9dcZir4B1MRgV8yc7wcngQbKz38KnZldeO3t7VAqlXj33Xfx1Vdf0dmNpqYmbNmyxWhUzY0bN/DPf/4TNTU1A66Jw+GMeg1LZWUlJkyYMCKzH5/FyckJDg4OA3YpWsJIy4/DQQjBnTt3Bvz8TWEsyI9sYo46YOP5hoxWWp0QEg7gGoB1AIoBfANAA+AqgHcArAXgBuAKgFUURRUSQq4DyALwOYCfAdhFUdQrw73WnDlzqNu3bw+7JrFYjKqqKqxatYqxjq9UKpGWlgaRSIR169YxSq+rVCpkZmZi2rRpFneiUBSFvLw83Lt3DzExMSNWl9OXtrY25OfnY+bMmaitrcW8efOs/poUReGzvE6IhPqs2+NOO3y5lN0r3NLSUohEIqPPRqvVgsvlore3F0qlkrY3sDYURUH61/fgrXqaIepW8WD/wTfgOA+/BoqikJ2djYiICLqIPysrCz/5yU9o93GRSIT//d//hYeHB9LS0gDo5b5t27b1G4BMURQIIdBoNFCr1SMewPT09CA3NxdLliwZsRFG1qSpqQkNDQ2IjIwc7aXQKJVKiz/XoqIiuLi4jIksGlN0Oh2uX7+OOXPmwMHBAYSQOxRFzRntddmwDqN5SRgN4DRFUXkURXUDSAQwHcBuAP9BUVQ9RVFFAI4AiCOE+AOYBeCPFEXJAHwJYBIhhLVq487OTnR0dLAit9jZ2WHr1q3g8/lISUmhTzyWUFZWhoiICPrkLBaLzZaFDHVZK1asQGpq6qgMMnVzc8OqVasgFosRGhqKvLw81q7IB4MQgqW+Tzf2cQ49uFjFXvZJLpejubnZyBMJAH2ivnz5MpKSkkZMunq2Cw8AHAUaKP72JnSy4SW8R48eYdy4cUbdj9HR0Th+/Djc3d0B6AOSPXv24O7du3Snl0ajQXJyMu7evdtvPQBw9uxZHDt2zOqf97OUlJSM6OzHwVCr1bh8+TIaGxstPoZBfhxrkpAhcDJ335TJZOjs7GRsXDxWqKmpgYeHxwsjP9oYmtEMnr4B8CYAEELGA9gFIAdABIC+VtnFACYDmAmglKKoXgCgKEoHoOzJ/2OFefPmYfv27bSZJpOAB3g6Fy0wMBCHDx+2SMqRyWSQyWRGG4xUKoVUKrUoyJs8eTJ27NiBnJwcZGVljbihZk1NDTw9PdHU1ITs7GxGJxNTedVPAK1Gf/IU8CgcLVdCy9LbHq77cf78+Xj11VdH1PTPEEA1CoLo+/QB1E+HDKA0Gg0kEkm/7BGg7x47deoUHSTqdDr8/ve/ByGE7g6lKAqnTp1CTk5Ov+dHRERg5syZIxrEtLe3o7e31ypmjuai0Wjw8OFDPHr0yOJjVFZWwtfXl55JOJYoLi7GN998Y/JFQt8ROWNBfmSKSqVCdXX1CyM/2hieUQueKIpSUxSlIoS8CaAFwCYASQCcAfTtc+56ct+z9/f9f6xh2NwvXryII0eO9DMENBdCCBYuXIiFCxciJSXFrNqAweoboqOjsW3bNvB4PKjVarOLUd3d3bFr1y40NTXh5MmTI2aoadhgQkJCEBYWhjfeeIMOCpn+nYeCEIIVfk9dxye49iClhHn0ZLClGKq+wdXVlS5Urq+vH7FCckIIvD/8T7MCqPLyckyaNGnQ7sdJkybh1KlTRl1RX375JaRSKT3bDQAyMzORmZlpVO83adIkREREAACam5tZLaAeiLFWG2RnZ4fXX3/dYvfxnp4e1NfXIygoaPgHjwJubm7w9/c3+Xfc1NQEoVBIB97PO2VlZZg8ebLVrClsjD1GrebJaBGE2EGfhfoDAC6AeRRFVTz5f28CWAngAICPKIpa1Od52QD+RFHUuQGO+VMA/wbAxcXFxT01NdWsNbW3t6O9vZ3Vzerx48fIy8vD9OnTTapdUqvV0Gg0Q9YT3L59G1KpFKtXrza7rkqn0+H+/ftobm7GggULrD4VXqlUgsfj9Vtne3s7srOzMW/ePKv5vFAAsu2mgSfQXy/cqxHgh053IeRaHsh0d3fD3t7e5ILo0tJS1NbWYtmyZVaxZxgIiqLgeeYApjs+LRrvUABFy/dAK3pqrqjT6aBQKEwyXFQoFPjHP/6B4uJi+r4FCxZg0aJFRp133t7eCA0NNQpedDodMjIyYGdnRxuMWgNTfjujhUwmg0KhMCsjplAowOfzR9WagE26u7sNdUGjvRTGDPbbWbZsma3m6QVmNAvGPwZQQ1HUoSe3+QB6oZfp/kBRVOqT+/8LwGMAKQBuAXCnKEpHCOEAaAYwk6KoIftkg4KCqOLiYos30ubmZshkMlZGbrS1tSE1NRUhISFYsmTJoJuHVqvF9evXMX/+/CHT9C0tLWhoaKCv6i3h/v37yM7Oxtq1a61mHSCTyXD//n0sWrSo33vu7u7G5cuXsXLlSrrN3Rrcbu7BxXp9t6FSReAKN7wbZdnmXVlZid7eXrPb33t7e+n5f11dXaw6qw8GRVFo+utv4aOqpO/rVvFg/4eD4Dz5bt26dQtBQUF0XdNw9Pb24t1336ULxgFg4cKFiI2NNfJNCwsLw9atW42uyBsbG43MYNlGrVbjxo0bWLRo0YgFqeZw9OhRyGQy/OhHPzIp8G5ra4NYLMb8+fPHfLDR2dkJtVo95PeovLwchBCjeYrPKxRF4ebNm5g6dWq/77OtYPzFxmTZjhAynRAS3effKkJIE4PXfgzgA0JIECFEBOC3AIoAfAXgt4QQN0JIFPS1UEcpipIAKAfwmyeZqk8APBgucAL0ZpNJSUkWux7n5OTg6tWrrPgkubm5ISEhAQ0NDUhPTx80zf3w4UP4+fkNW9/g7u5OB05SqRT5+flm2yNERERg48aNOH/+vEXPH47h5tc5OjoiJiYG9vb2oCgKBQUFrM0K7MssDyG4T77ydgIKd9vkuNto/ntVqVSoqamxaPM31D7l5ubi4MGDVpevgP4Snk5HQZz7EIW/+x0onQ6PHz8GIcTkwAnQv4+vvvoKe/bsoe/LyclBcnKyUUFzRUUFmpqMtwkfHx/6RFNQUID2dvO9qIaioqICgYGBYzJwAoA1a9YgLi7OpMDJUBvE5uxHa0FRFFJSUnD16tVBH6NUKtHQ0DBm5UdzaWpqgp2d3QsjP9owHZOCJ0LIn6C3CEgCkPzkv2cB7GPw2vsAZEJfJC6FvvtuK/SF5HcB1AA4BuBdg4QH4HUA2wG0AVgB4A1TXsjR0REzZ87EkSNHLHJ+XbduHX31PNC8OXOxs7Oja5aSk5P7nUAt3WBKS0tRWFhoUQ2Tn58fEhISIBaLcebMGVaDl8bGRtjZ2ZnUrt/U1ISrV6+irKyMtdc3wCUEK/vUPvm6qPDLCxQazYxfysrKEBISwqi+YerUqZgzZ86IzSXrG0AV3qhFQ00nGk6fRtFnn9GFu+bC4XDw2Wef4cMPP6TvKygooMe58Hg8xMfHD9pNJZfLkZubi/v371v8vgY6ZktLS7/ux7GEs7MzHB0dQVHUsPMn6+rq4OLiYnVJnQ0IIXjttdewevXqQR9TUlKCsLCwUe9+ZAOtVguxWDwq5qs2Rh+TZDtCSAOAxQBcAPyaoqgfEELeByCiKOqPVl4jYww+TxUVFcjMzMRrr71m8ZVPbm4umpubsX79esYbAEVRuHXrFu7evYvY2Fi6BuLOnTvw8/MzKsI19XhyuZzemHt6esyWKtVqNS5dugSpVIrY2FjG/kQG+XHBggUmd5w1NzfDw8OD9gdiswiToigcqehGoVSL0w+coVBxEO4JHN0G2JtQTiKTyVBUVISFCxeylgmQyWQoLi4eEVmGoig8+OwzVB88SN/ntmsXFv2R2c94//79+Pjjj+nbkZGR+Oqrr+Dv7z/k8zo6OuDs7MyakWZeXh6Cg4PNyqKNFrdu3cLNmzfxxhtvDPg7M8iPixcvfiFqndra2lBeXo6oqKgxn0UzhfLycnA4nEHLOWyy3YuNqTuWE4A66GWzmU/u+xeAX1pjUdYiJCQEsbGxyMjIQFFRkUXH4PF4RvPmmEAIQVRUFJYvX44TJ06gvLwcbW1tUKvVZgdOhuMZMhn5+fk4ePCg2Y7ifD4fa9asQUREBI4cOcJ4Tp9EIoGfn59Zrfqenp4ghEAul2P//v1GhclMIYRgx2RHxASMg0qj/wwfNAO/yQB0w1xHWKuDSywW4/bt2yMyTJcQgvA//AETNmyg7+s6moTW//kUYJBR/eEPf4g///nP9O2CggL84he/6De6RS6XG3Uburi4gMPhQKlUIj09nZGM2dzcDA6H81wETgAwbdo0LFq0aNC6t/LycgQGBj53gVNraysyMjKMShJeNGuCF01+tGE+pkYAN6APlngA+ISQVQCWQ9/E9Fzh6+uLHTt2IC8vD9nZ2WZLcHPmzMGaNWtACIFCoRg27W4KISEh2Lp1K65evYpjx46xYoIXEBCA0NBQiwzbCCGYNWsWNm7ciAsXLuDWrVsWSZVKpRKNjY0WbzA8Hg/e3t6snwx5HIIl/gR/Wvb0vnMS4Mvcod9jY2Mj7O3tWXcLnzNnjlH2wRr1Xn0hHA5mffEFRK+8AqEdDwtXB2F8yz10fvMnRsd9NoAqLCxEQkICHUB1dnZi7969OHnyZD+7BplMhkePHlk81kan06G0tNQi+XG0cHR0xJw5c8DhcPr9vrq7u9Ha2jqm5cfBUCqVEIvFaG5upu+rq6uDq6vrcyE/moJBfhzt0UM2Rg9TP/kfAGjFE9kOwHcA/hfA21Zal1Vxc3NDfHw8qqqqkJGRYbbrseHK6cKFCzh27BgrheReXl5YvHgxOjo6cP36dcbH9PLywtKlS+kMTkFBgdkBkJ+fH3bt2oWKigqkp6ebfVIvKSnB1KlTLd5ghEIhNmzYQGfhiouLGRuX9iU+HNjzZLiPv5sKVQoZTpQO/DfSarUoLy+3Sn0DIYTOPpSWlmLfvn2sF1E/S5dSCbtf/hLTX50GZ1e9tDuuoQCyI//D6LiDBVBSqRT79+9He3s7ioqKkJaWZhRAeXl5Yc+ePXS3p7nf1aqqKnh5eVm1W9NaNDc34/vvvze6EHueszS+vr742c9+Bl9fXwB6+fHhw4cDmq8+jxjUgbFgvmpj9DDprEZRVAtFUe9TFFVNUdR5iqL8KIryoSjqmLUXaC0cHBywY8cOKBQKpKamWjQ+Y+nSpVi1ahUrNTlqtRpSqRRvv/023bXCVqBQVFSE69evWyQLOTk5IS4uDkKhEIcPHzb5pM5EfhwIhUKBS5cuGQ2lZYOPlgCbw5VYO10Gfzc1ztV1405D/xO3RCLBxIkTre4U7u7uDn9/f6taGNDdj5GR8Pj0/6Gt4+nFg1PZRXSfSWR0/IECqD179hjVP927dw/p6elGQZKhO66qqgqJiYlQKBQmvV5vby9qa2tZsRIZDRwcHMDn8+mLk+bmZnC5XIwfP36UV2YZhBD6szRcdAQFBT138uNA9JUfbbzcDBk8EUKmEUJyCSEyQsgVQsjzb8zRB4FAgNjYWIwbNw7Jyclm1we5urrSZpcSiYRRm79hg7G3t8f69esxadIkJCYmGqW+LSUqKgq7d++mZSFzs1o8Hg+rV6/GrFmzkJSUhMrKyiEf39eagC3s7e2RkJCAxYsXAzD/PQwGjwNsnQpwnlzgB3v04h8FStT3KdUxyI+WDmU2Bw8PD6xbtw5cLhcqlQoPHjxg3TqisbERDg4OcHFxgcDDE3a//r/olOnrUwghsM89CsX1s4xeY6AAau/evUbfibt37+L06dP93h+XywWPxzM561JWVobQ0NDn1t3ZwcEBCQkJ8Pb2hk6nQ0lJyZibX2cuFEUhNTUV6enpaG1tHbZx4Hmhtrb2hZIfbVjOcJmnr6G3KFgAvUHlt1Zf0QjD4XCwcuVKTJkyBUeOHEFLS4tFx3n48CHKy8stGnxqqG8wbDCGkS7R0dE4duwYJBKJRWsyQAihr2IlEgm+++47s2UhQghmzpyJmJgYZGRkIDc3d9CTem1tLcaPH896G767uzs9d/DYsWPIzs5m5biLfUSY7PTUT2u6rwIfXuuF/IlKyVR+tJQHDx4gIyODlQDawEDt1XYBgeD+5C/o7tYHUBwugeDC/0PP3f4z6sxhoADqu+++M7pqLygowNmzZ42+S/7+/oiLi4OdnR0oihoyK9zZ2Ynu7m5MmDCB0VpHG0IIdDod0tLSIBQKn0v5sS+EEPj4+EAmkz238uOzqNVqVFZWvjDyow1mDHc2mA3gY4qiigF8BmCW9Zc08hi63hYtWoSjR4+irq7O7GOsXr2a9oLSaDRm1QcNZiAZFhaGzZs349KlS6yZVzo5OcHT09PiwMbX1xe7d+9GZWUl0tLS+p3YDBuMNQdkGgwd2SokJ4Rgc7ADXPvICqE+XfjgihqPW1qh0WhYkx/NYdasWYiLi6Nfm43PXyKRwN/fv5/86DgtAtpt70Op1Gf0eDwOOEc/h6riAaPXGyiA2r9/v1Hwdvv2bVy4cMHo/Rl+CxcvXsTRo0cHNJMdznz1eaOzsxPXr1+3esPASBEUFIRp06Y9t/Ljs7xI8qMN5gwXPHEpilIDAEVRCui77V5Ypk2bhnXr1uH06dNmmzQSQmg38IsXLyI5OdkkaUkqlYLP58PNzW3A/+/j44OEhASUlZXhwoULrBSSx8bG0hmc4uJis0/Kjo6O2LFjB+zs7HDkyBGjQlexWGz1DYbL5dLZQgCorq5m3PXIJQRvhDmBB713F48LuLrI8KerilGrbyCE0EW3TU1NOHDgAKNCcqVSiaampkHlx3Hzo6Fc/hOoevXZUwGfA913v4emwfRh1gMxUAB14MABoyv4vLy8fsOEAf0JODg4eEBJrqGhAY6Ojhg3bhyj9Y0Vampq8NOf/hTR0dGjvRTG9JUfa2trWZeeR5pn1QEbNmx9ls8QEBCAbdu2ISsrC3l5eRb96ENDQxEWFjZsDYahvXq4+gZD0bZKpcLRo0dZKyQvKSnB+fPn0dDQYPZzDXVQkZGRSE5ORkVFBbq6utDe3j6iG4xOp8OlS5dw6dIlxscS8TjYM80ZFKXPYtgLKPC9HHG+1ny7B7bR6XTg8/mMRo4UFxcPKz+6rYlFZ+RWaDT6TjiRgKD9y/egtaChoi8DBVCHDh0yGnFz8+ZN3Lt3z+h5wcHBWLBgAQgh6O7upi8eNBoNysvLERYWxmhdY4XOzk7I5XJ6kLJcLh8R3y9rUVVVBR8fHzx69AhHjx5FbW3taC+JEdbwd7PxfDOkwzghRAvgguEmgFUAMvo+hqKodVZbHUsYHMbNobu7G6mpqfD09MTKlSstdhNvaWlBR0fHgJ1AEokEWq3WZA2doijk5uaiqKgIMTExjFtlKYpCfX09PT5Dp9NZVNfT2NiI06dPQ6fTYevWrSNuUiiTyUAIgZOTE7RaLTgcDqNNrlqmxhFJJwyHqGkV4MdTnTDXd3Q3ToqiQAgBRVFoaGigs1Km0NraColEgqioKJMeL/32v+BRcxWtzXLkX62B58rVmP3f/w3C0FX/+++/x0cffUTfjoyMRFxcHB4+fIjQ0FBs3759wIsOlUqF77//Hn5+fli7di3KysrA5/MRHBzMaD1jAYqikJOTg/DwcIwbNw46nQ779u2Di4sLtm3bNtrLM5ve3l7cvHkTS5YsAUVRkEgkmDx58nNb0C+VSvHo0SPMnj3brOfZHMZfbIY7U/4Y+ll2hnl2e578t++MuxcSR0dHxMXFQalU4sSJExbNiwP0w1IvXbrUr2ajp6cHjx49Mqu9mhCCBQsWYNmyZThx4gRKS0stWlPf4xkCp5aWFuzfv9+i2X8+Pj5YtWoVurq6cOnSJVa9mEzB2dkZTk5OoCgKFy9exJkzZxjJBJOc+ZhKnrbbBYxX4Q/X1XgkG+JJI4AhICwuLkZSUhLq64ediQ3AsvZqr5/8GvXj5+HW5WpoNTo0nj+P+598wlh+eeONN/CXv/yFvl1QUICkpCQsXrwYO3bsGPQEKxAIEBUVhcjISCgUCkil0hHpfhwJnpUfORwOli1bhuXLl4/yyiyjtLQUoaGhdNekKVn4sYqp6oCNl48hgyeKor6nKOp76Mey7AbwMYBPn/n3wiIQCBATE4Px48cjKSkJHR0dZh9j7dq12L59O/h8PiiKoo0BDe3VlmS0QkJCsH37dmRnZyMrK6ufW7Ol2NvbW9Tlo9PpUF1djbfeegu+vr44dOiQySd2tnFzc4ObmxujzFNXVxfGd9dj6jgRtDogs9QJJVIB9qQB3WOgljcsLAxr1qwxucOspqbGou5Hv3c+QsAbP6Rv1yYno+zLL806xkA8G0AVFhbi73//+7BWIREREfDy8kJJSQnGjx//Qkgog8mPwcHBdB3k81Qv1NHRAblcDh8fH/o+iqJQVFRklWHf1qayshI+Pj5mzwi18eJjqkZzCEAegB8CiH/yb+eT/77QcDgcLF++HBEREUhKSjK7PqhvMfjt27eRmpqKx48fQ6FQGG0w5uLh4YFdu3ahsbERp06dsjgzZsDd3R1xcXF0BkcsFpsclFVWVmLChAlwcHDA4sWLsWLFCqSlpaGwsHBEN35CCObOnYuFCxcC0GfTLOmcNGRpYgIdsNDdBbVt+s60slbg/5wHtOzEqhbD4/HoDrPu7m6cO3du0M9frVajqqrKou5HQgimf/QRfDdtou9TXUxG55e/YTQHD9AHUH/961/p23fv3kVCQoJRnU9WVhZycoztElpaWtDY2IgLFy6gpKSE0RrGAoN1Pxq4evUqzp07N8KrsgxDhvPZ7kdCCIqKiiAWi0dxdeZjiTpg4+XB1ODJC8C/UxSVQ1FUXt9/1lwcm2g0GkYn8sjISKxatQonT55EeXm5RccQiUQQCoUoKytjpb3azs4O27Ztg5OTU7+uN0swrKeurg6nT582SRY0bDB9a08mT56MnTt3oqioCOfOnRu11uvr16/j3LlzZnUoNjU10QEvIQTLAnj4ax/15FIV8Dkz+yNWaWpqQmVlZb8BvAbEYjGCg4Mt7n4kHA5e+dvf4LlsKSaHe2DmfD+M66pC578+ZbBqPa+//vqAAVRHRweuXLmCK1euIDMzE7m5uQD0J+eSkhJa0nre/XZMkR+FQiHs7OxYyy5bk/r6+kG7Hzdv3oyYmJhRWJXlMFEHbLz4DFkwTj+IkK8AVFIUxTxnPwrMnj2bev/99+Ht7Y1XX32V0bGkUilOnjyJyMhIzJkzx+wAqK6uDu3t7QgJCUFHRwej7FNf7t+/j+zsbKxdu5aeD8aEyspKBAYG0gXKg73PwsJCeHl5DSghqdVqXLx4EVKpFDExMYPaMViL3t5eyGQyeHh40JLpUBuhTqdDVlYWoqKi+qXp/3YD+N87gICnw4JAOdb6OyA+fGw0q/b09NA2GQqFgpZeu7q6cPfuXSxevJhxoK5RyNH+8Q/gMe7pfiGb/Cqcd7/L6LgAcODAAXz44Yf07VmzZmHTpk1G0u+6devg4eEBhUJhVH+iVqvR0NDwXA7Qzc/PR0BAADw9PUd7KYzRaDS4fv06Fi1aNGRH6FB7yViio6MDJSUldKenJdgKxl9shhvPUkoIKQGwEsAXhJB2w32GfyOzTGYQQuDl5cXKJuXl5YX4+HiUlJQgMzPTLEdxjUYDiUSCsLAwXL16FcePH2cstxmIiIjAxo0bcf78edy+fZuxXBYUFARCCJRKJRITE1FT09/rp729HUqlctAAkM/n47XXXsOsWbOQnJzM2CndXIRCITw8PADog7yUlJQh/94G+XGg+ob3FwHrQ7XY+konpvn04ubjLuQ+Ghu1KIbASSKR4Ntvv0VDQwNtIMlWezXP3gHjPvkGbR1PMyBOFVfRfep7xsd+NgNVWFiIkydPGn2vzp49i5ycHCNrA0Bvb3DixIlBM29jlZaWFuh0OpP3pJaWFty/f9/Kq7IciUSCgICAIQMnw3QDtvY8a/Gima/asA7DXTr/HMBbAH4KYBmATX3uM/x7Lli6dCntatzQ0AClUmnxsZydnbFz507azsDUzaCiogKTJk2CQCDAq6++ig0bNtAnPjbw8/OjAzs2DDUB0F2Cz3bLDFbf8CyGsS6xsbG4fPkyqwXu5mCQEwarLRlIfuwLhwA/m6uBf3akSAAAIABJREFUi70+WJ7gosbe4m7UdIyNAArQdz1OmzYNHh4ekEqlEAqFrGb7BG7jYf9v/4NOmf57RQiB/Z1UyK+kMT7266+/jr/97W/07cLCQqSnpxvZcVRUVODBA2PH86ioKMTGxlp1kDLbGORHc2Y/FhYWIjs726IB5tbGID8Ol/F2cnKCi4vLmA+e6uvr4ezs/Fx9p2yMPCbJds87fX2eVCoV9u7dCz8/P2zcuJHRcXU6HS5fvoxHjx5h8+bNQzodKxQK5OfnY8mSJf28lKqqqlBeXo4VK1aw0tKrUqlw4cIFdHV1ISYmhvGMub6p9urqavj5+aGxsRHt7e2IiIgw+TgKhQJnzpwBAKxfv37U5ncplUrU1tYa1cwUFhbC29t7WBn1fI0Cd9sU9O3yRgf83xV2cBo4JhsVtFotrl69Cnt7e0RFRbE+k09eVgzq29/B0VFfR6XV6KDa9B7s5jB3xj548CA++OAD+nZkZCRee+01o5mTmzZtwsyZM/s9t6mpic4yj2WqqqqgVCrNan/v6emBTqcbkzPv8vPzMWnSJDrL+zxjqvxoCjbZ7sVmbBRtjCACgQAbN25kXPsE6DvxVqxYgfDwcBw5cmTITrwHDx5g2rRpA57IWlpaIJVKWcvICAQCbNiwAUFBQUhMTGRsG2AInDo6OpCamoobN27Q8qM52NvbY+vWrfD29sbBgwdHzc4gPz8fZ8+epaUeg/xoiunoGn87+IieRkoh3nJ8cLV31Dvw+lJZWQm1Wo2cnByL3OOHwyFsOnQ7P4ZSoc9Kcnkc8FK/RK/43jDPHJ4f/OAHRhmogoICHD9+3CiDdurUKRQVFRk9j6IonD9/HpcuXRrTrf0qlQrV1dVmdz+KRCLY29uDoqgxJVE+fvwYFEWZFTj19PSMuBecqVRUVAwrP9qwAQDcTz/9dLTXYHW+/vrrT3/605/St52dnSEUCkFRFLKysqDRaCyWNwghmDBhAlxdXXHmzBk4Ojr220geP36Mjo6OQTdMX19fTJ8+HXw+H1qtFh0dHYx9RQwGmK6urjh79iwEAgG8vLwYafgikQheXl7gcDjw8vKyyGuHEIKAgAC4ubnh7NmzAIAJEyaMaG2Bn58fAgIC6ELy/Px8REREmCShEkIQPl6AgsdqaCgdCAHsRSpcrxLgVf/Rvxbp6elBaWkpli1bhuDgYLqQn+1CXeGEiejmuIBTfgs8HgdcLoGm4AqosChwnV0ZHXvmzJnw8vLCxYsXAejd0ZVKJSIiIqBQ6LN+ZWVlcHd3p2uGCCHw9/fHtGnTBpVmxwIlJSX0fmEJ2dnZyMjIwIwZM0Z9QK1Op8OdO3fwyiuvmBxsaDQafP3111CpVGPO5FQul6OiogIzZ85k5bfy2WefNX766adfs7A0G2OQ0d/tRxG1Wo26ujpWrs6Dg4Oxfft23LhxA9nZ2fTVr2FA5nDuzga5Ljs7G4mJicMaBpqzrri4OBQWFiIzM5OVwcIdHR3w9/fHyZMnLfbaCQoKwq5du1BRUYFTp04xqkEzFy6XS482ycvLw/Xr183K+vE4BG9OcwZ0+p8Pnwv0cGU4VDT66aeSkhJMmTIFXC6XzqQ1NzfjyJEjrGcsXFesQ9e8ndConwwS5lKo+P9+By0LNS27d+82KiK/c+cOLly4gPHjxwPQf4bPXmC4ubnB0dGRLvhlo+aPTWQyGTo7O2lXf0sICwvD/Pnzx0RmpKamBp6ennBwMH32I4/Hw9KlS82q9xopiouLB1UHbNh4lpf6WyIQCBAXF4clS5YA0F95MEn5e3h4ICEhAbW1tTh9+jTUarXZG8ysWbOwZMkSxnVKfXFzc0N8fDyUSiVSUlLQ1dVl8bEMBpJqtRpqtZqR1Ojs7Iy4uDg4OzsjMTERTU1NFh/LEjQaDR49eoSgoCCzM30OfA5+PNUZOp3+CtVJpMO9jk6cFI9eANXe3o7e3t5+8qNKpYJarbZKds99yy60hqyGqleDW1eq8fDKHdx55x3oWAhcFixYgPfee4++fefOHVy8eBHe3t7YtWsXgoKCBnxeU1MTLly40K+4fDRhq/vR3d0ds2fPHnXvIYP8+Gz3oymEh4ePOXuGx48fA8ALUbdlY2R4qYMnQN9OTwiBSqVCUlISMjMzGR3PwcEB27dvB5fLxeHDh1FaWmrWBuPk5IRXXnkFANDW1oYrV66YZYcwGEKhEDExMQgODra4Durx48cghMDd3R0ikQjbt2+nryAt7WDkcrlYvnw5oqOjkZqairt3745YzUp5eTlmz56N+Ph4CIVCaLVaFBcXm/z6nvY8bAp0os22xztqkVyuQHKxFRc9CEOdnP38/PD666/T7vGtra2svrbXD99GXcBreNyoz5ZKL15E0R/+wOhz7OnpQX19Pd555x18/vnn9P35+fm4efPmkNkbHx8f7Ny5c8Ci8tGiqakJIpHIYrnuWWpqauhh3KNBWVkZo2G/bW1tYya4NVUdsGGjLy998GSAz+dj5syZrPyAeDwe1q5dCx6Ph+LiYotPVtXV1SgrK2OtuJIQgqioKKxevRppaWlmBSoDbTCGk7RarUZaWhoyMjIsXltoaCji4uJw7949nDlzxuot2XK5HI8fPzZqry4rK8P58+fNGukyzU2AaG9HUBRQ28ZHbpUD3r8I7Cu0wqKH4NGjR3BxcRm0vdrwWd2/fx8HDhywaAD0UAT//JcIfvNN+nZtcjIq/us/AQtP7iUlJQgLCwOXy8Xu3bvx7//+7/T/y8zMxO9//3uj765cLjeSun19fUEIgVwuZzxAmylarRZisZjV4bJyuRzNzc2syfvmIJPJIJPJGMmPDx48wMWLF8eEbYEl8qMNGy9lwfhAGAq/DSef0tLSAesqTKWrqwu9vb2IjIzEmTNn4OrqanZRuo+PD8LDw2kJTy6Xs1Lr4OrqismTJ+PatWtobm7GpEmThtX5q6qqIBKJBnQS53K58PHxQUhICEQikcXFyXZ2dpg+fTrq6uqQk5ODiRMnWq01u7CwEFOmTDHaMD08PODn5wd/f38A+oDRlPfh78TDOB4PJ4rsIZXrH3+tRu8NFeULWLsWXq1W4+7du4iMjBxWznFxcYFIJEJISAjrMp77woVQ1NVBVlYGkT0fU8d3oKfsPkTzV5n1R2hra4NUKkVYWBi9xlmzZkGpVMJgOXL//n3w+XxERUWhtbUV33//PcrKyjBjxgyjv8GNGzeQm5uL8PDwUasTkkgkcHZ2ZtVCwd3dHTNnzhzxgbUURaGgoADTpk1j9NoeHh6IjIwcdesFlUqFoqIizJo1i/VaJ1vB+IuNLfM0ACqVCteuXaNnaplLX4faKVOmYMuWLbh06RJu3bpltpRh2KCKiorw3XffsSa5uLq6IiEhAb29vUhOTh6yDkqlUqGmpmZI+dHX1xcuLi4AgCtXriAnJ8ci2YbP52P16tWYO3cuUlJSUFzMvgbW3NwMDocDd3d3o/sNHVuAvn5o//79JsubER4CJG4hmNPHJurvuRT+nK1lOkN3WCoqKhAYGGhS95VIJMLcuXNBCIFCocDNmzdZk34Ih4OZf/sb/JYvweLXguHkIsK4Tgk6930+/JOfMJT56kcffYRNfYYUf/7550hKSsJ3332H9vZ2NDQ04MSJE0bvJzo6GvHx8aOWVVAqlWhoaBi0PstSCCHgcrnQarUDuv9bi6amJtjZ2TGWH+3t7cdEpqesrAwhISGs+OvZeLmwBU8DIBAIsHPnTqxcuRIAzD65NDY2wt7eng4mvL29kZCQALFYbLHzt7+/P8LDw1mrmQCeel6FhIQgMTERjx49GvBxhrotUzYYnU6H3t5eqFQqRpmN8PBwbN++Hbdu3UJGRgbtdM4UnU6H0tLSYeVZiqLM3uDHCYGDm4FoffyFOf5KwK4dn1y3ng9Ud3c3WlpaLJrtJhaLcevWLbS3t7O2Hg6fjxn/9T+Qa55mecbV5aHr+LcmPb+urg4uLi5wcnLqf2wOB3//+9+xcOFC+r7f/e53RtlQsVhsJB/zeDy6CFgikbBe7zUcfeVHa3D79m0cP36c8VBwUzDIj4ZJDUyRyWRIT09HY2MjK8ez5PVlMhndeWvDhjnYgqdBcHFxgUAggFarRWpqKvLz8016nlarRXl5eb8NxsnJCXFxcejt7cWxY8fMrmMaN24cli5dCg6Hg56eHuTk5LBSSE4Iwbx587BmzRqkp6ejoKDAKGPU2dmJrq4ukzcYDoeD1157DdHRerfp9vZ2i0/Ohu5FlUqFw4cPs3KCqK6uhpeX17BygZubG+Li4ugAWCwWm/T3tucD324Ets/owdxJCvA4gL1jFz641gM184+rH4buR0sC1VmzZuGNN96g2//ZykDxHB3h9OG/0N759A073DsN+aWTQz5PrVbj4cOHQ5qvCoVC7N27l/59aTQafPrpp0ZZ0by8POTl5fU79sWLF5GdnW3JW7KItrY2qNVqqzqez5w5EzExMaxeVA3Gw4cP4efnx9pIKYFAgMbGxlEx/bTNr7PBFFvwNAwURUEkEpms70skEvj5+Q1o1CcQCBATE4OJEyfi8OHDaG5utmhNEokEeXl5dHstGwQGBiI+Ph4PHjzA2bNnoVKpLN5gCCF0/UBGRgZSU1MtPjELhUKsX78er7zyCpKSkiz2lQKA3t5e1NTUYPLkySY93vCem5qacPr0aZMHswp5wAcL+NBq9X8DDgHGu3Tj/atK9LBoPdTc3Awul0sHP5ZgCA4lEgkOHz5Mm1AyRejpBeGvvkRXlz5jyOEQiK58h5471wd9Tnl5uUnyo7OzMw4cOECP0pHL5fjiiy+Miv/Pnz+PsrIy+jafz8eOHTuwdu1aBu/KdAzyI1uDmQdDJBJh8uTJIIRYtUvVGvKjSCTCm2++aTQmaaRobGyEnZ0d/f23YcNcXorgSSaTWXzy5vF4WLduHd2S39TUNGhLvlKpRGNj45AbDCEEixYtQnR0NI4dO4by8nKz1xQeHo4f/ehHtJ8PW51pLi4uiI+PN7JZcHBwYLTBrFmzBmvWrKGDKUs2eMNw4W3btiE3NxcXLlywSMYrKytDaGio2fUN3t7e2Lp1K936bsp7cLfj4lcRLtBpn8o1Xm5yvH9Nga5e5ic5g/zIVgcXl8uFQCBgtfbDPigE2P0HKORPx7hwTvwnVBX9W9S7u7vR2tpqsvw4YcIEHDp0iG7waG5uxv79+408rk6cOGFkgOvm5gY+nw+NRoOioiKrBht1dXVwdXUdUH60BmKxGIcOHWJN3n6WkpISTJ06lfWiakNgOZKGpgZ1gM3uRxsvHy9F8KTVanHy5EmLg4y+LfknT54ctCXfnA1mypQp2Lp1K65evWrkSG4qhoDm0aNH+Pbbb1mbE8fn87FmzRpERERg3759jDuUXFxcaMmvqKgIZ8+etXij9PT0xK5du6DVapGYmGhW/UpnZye6u7sH7BY0BUNHYm9vLw4fPgyJRDLsc5wFHLw7cxygexqQTHBT4OMbcrQpmJ24q6qqTJIfTSUwMBDbt2+npWq2amicZs1D79pfovdJyk3A50D33SfQNBnbQViSpQkLC8PevXvp72h5eTkuXrxID+hWq9U4fPgwOjo6jJ5XVlaGjIwMq8z9M7zuw4cPRzSjYm9vD5FIZBWLD2vLjydOnKBHNY0EQ6kDNmyYyksRPLm6usLR0XHYrrLh4PP52LBhw4BDhVtbW83eYLy8vLBr1y7U1tYiLS0NKpXK7DWNGzcOgYGBjKSbZyGEwMHBAVu3bsWNGzfMHl8yGD09Pejp6WF09SoUCrF27VrMmTMHycnJJhntseXuDOivkA1ZGlOw53Pwm1fGgUc9laImuPXg09xuNHZbFkD19vaitrbWZPnRVAx/m6ysLFYlPNfla9E1Zzs9xkUkAFT//S50nfoAjYn8uHDhQvzjH/+gb+fk5EAsFtN1OXK5HIcPHzbyE5o+fTp27txptULh8vJyBAUFjejsuYkTJ2L79u2sTiYAjOVHaxEQEEB3uVobU9QBGzZM4aUIngBg1apVmDp1Ko4cOWJxrRGgd2s2ZH2ysrJQUlJi1F5tLgZHcpFIhCNHjqCzs9Os5zs5OWHdunUQiUTQ6XTIz89nnLo3bDBRUVHYvXs3Ghsbcfz4ccbz5+bOnYstW7bQRe+WZssIIQgPD8eOHTtw+/ZtnDt3bsjAs7GxkbH8aMDBwQFxcXH0Zl9TUzNs0CvgErz7ijPs8DTgmuDai8/zu1DdaX4AZan8aCqzZ89GdHQ0qx487tteR8ukpdDp9O+3+3En7n/2R2g1GsbuzrGxsfjkk0/o22lpaZDL5XSQ7u7ubtTtRgihA6eWlhZG+8GzGOTHkQoGnkWpVBrVejGltrbW6vLj7Nmz6akK1sZa8qONl4+X5htECMHcuXMRHR2N48ePo6qqitHxNBoNGhsbIZVKUVtbi/Hjx1t81cfj8bB69WqEh4fjyJEjg1oGDMejR49w/fp1VFZWWvR8A30HZNrb22Pbtm3w8vLCwYMHGbcVG7IbOTk5OHr0KCP3dHd3dyQkJIAQgsTExAEL6Nlurwaevge5XI7U1FTcuHFj2OfwOAS/esUJ47hPpQI3RzXePK2FxAyFjKn8aArOzs6IiIgAoM+oMinS74v3m79Bk+M01D1sx60rVag7lY7cjz+Gj48PY7PHn/3sZ9izZw99e+/evXB0dMSCBQuwffv2AbNAFEXh9OnTyMzMZK3+aSSKxIfCcDHBJMNuQK1Wo7KyckTkR61Wy1rpwWBYog7YsDEYZKTmiI0mc+bMoQzOxABQX1+P9PR0LFy4kD5JWIJWq4VGo0FOTg5eeeUVODk5Mc4GVFVV4fz581i0aJFFa3v8+DHta6PRaMxeT2trKyQSCaKiovr9v4qKCly8eNHitfWlt7cX9fX1dPrcUldyA8XFxbh27RoWL16MGTNm0McSi8Xg8XgIDg5mtN7BMIx2sLOzM+k9UBSF70rkaOjpwZkiZzR0CjDeDjiwCQgfZlYqRVHIyclBeHg4Xddjbc6dO4eamhr86Ec/YqVGhKIoFH3yCWqOHKHvC/vd7xAyzAQAU9Bqtfj5z39uVD/zz3/+E7GxsYM+p6WlBUKhkJXMilQqRV1dHebMmcP4WJbS09MDmUzGyuDd4uJiODo6WuQhZi75+fnIysrCnj17rNIBR1EUrl+/jsjISNalzcEghNyhKGr0vgw2rMpLGTwBev+h1NRUTJ48GUuWLLH4xP3gwQPY29sjKysLLi4uRg7IltLW1oZTp07B398fS5cutchgr6OjA8nJyVixYoXJtTGmbDBtbW1IS0uDj48Pli9fzkpdR319Pa5fv44NGzYw2tgMa/Pw8MDKlSuh0+mQl5eH6Ohoq6fpdTod0tPTERQUhBkzZgz5WIqikFmtxTvneFA8UVidBcB3scCcIRJK9fX1aGlpGdGBtxqNBl1dXbSPENMgFwAorRa3334bTX0aL+b98QN47XpziGeZRk9PD+Lj43Hr1i0AenuQxMREI2PNlpaWfu7yFEWhsrISQUFBFr0/nU6HrKwsREVFjfjIlMHQarUWm3N2d3ejoKCA0d5o7us1NTUhMDDQKoaiNTU16O7uHtHhv7bg6cXmpZHtnsXV1RXx8fFoaGhAenq6RXVCXV1daG9vR2BgIBYsWIDIyEhW1ubm5ob4+Hh0dHRYXGskFArh7e1t1jw9U+RHNzc3JCQkQK1WIykpqV8nkyX09PRArVYzztq5ublh165d4PP5OHToELKysmj50dpoNBpoNBqTjDQJIVgdyMOhzYDzk2SOTAX88rwGGZUDP1+j0aC8vHxIA0lrwOPx6MDp9u3buHTpEmOJi3C5CP70U/CeyEFTZnrBq+Isug7/N+P1ikQi7Nu3jzbNVKlU2LNnD8rKyqDT6XDu3Dn861//QnV1tdHzqqurcfLkSYjFYotet6qqihX5kS3u3LmDgwcPWmykO9IGko6Ojpg8ebJVAieD/BgaGsr6sW28vLy0wROgnxu3bds28Hg8pKSkmF1/07e+YcqUKXSRaElJCeOaKpFIhM2bN8Pb2xuHDh0yu6jVzs4OsbGxdPD04MGDISeYm7PBCAQCrF+/HtOnT8eRI0dMatsfiuDgYOzevZsuepdIJBafoA2z8cLDw3H16lXU1tZa1c/HgEAgwJYtW+jC1+bm5mGD3tk+QPJWwN0OcBJqsWqaDDcedyBV3N/KQSKRwN/ff1TbqxUKBZRKJeO/J0VRKJVIMO/rrzFlyVSERuhrUBzFl9F95jDjdbq6uuLQoUN0bYtMJsPu3buRmpqKW7duQavVIjk5GS0tLfRzJk2ahI0bN1pU32Ot7kcmuLi4YMKECRZdFEqlUvD5fLMHmTOlt7cXBQUFRp8LG4jF4hHvfrTx4vNSB0+A/sp67dq1CAoKGrToeCAG22B0Oh0KCwv7jTmxBA6Hg+joaCxZsgTHjh2zuIumvb0dmZmZKCwsHPQxYrEYwcHBJm8whBBERkYiNjYWly9fRlZWFiM7A8MVbklJCU6dOmVx0TygPzn39vbi7bffRnl5OVJTU1lrux+KvoZ/qampOH/+/LDPmeYBJG+jsDFCBkehDvYCCvc6O5FY/PSkp1AoIJVKERgYaLW1m0J0dDTWr19P+11Z6tfV0NAAR0dHjPfzg/+fvkJnp/69EkJgn5sCZc7APmrm4OfnhwMHDtCzCRsbG7Fv3z66g7CnpweHDx+mL5gIIQgNDQUhBEql0qyu19LSUkyZMsVq8+ssITg4GKtXrzZ7lArb5qvmvva1a9cYN7z0xaAOjFb3o40Xl5e25mkgysrKcOXKFaxevXrIAuPh6htUKhV0Oh1EIhFUKhW4XC7jjbW5uRlpaWkIDQ3F4sWLzZaipFIp3bL9bC1EV1cX7t69i8WLF1uUplcqlbT55fr16xnVLel0Ojx8+NBo5IS5a6quroZcLsf06dOh1WqRnZ2NkpISrF27dkSKXwF956O9vb3JV+8FUjXO1snA4+p/j2ot4M1zxpuvCJCfn4+AgABWioDZQKfTISUlBSKRCLGxsWZ9PhqNBtevX8fChQvpLFp3yX2QfR/BwZH/5DE6aOM/gXA683KRrKws/OAHP6ADveXLl2Pq1Kl0RsbPzw+vv/46fdFAURQOHz4MiqKwa9euYd9bR0cHiouLsXDhwjE5I62trQ0tLS0mS1YSiQRarXZURqYA+iyhwTWeDXJzcxEaGjriWTTAVvP0ovNSZJ46OztNkuTCwsIQExODzMzMITNHlZWVmDBhwqD1DQKBACKRCBRF4ezZszh58iTjLJTBXVsqleLEiRNm10F5eXmBy+VCpVIhMTERd+7cAcCOgaSdnR22bNkCf39/HDp0CLW1tRYdB9Bn20JCQvD/s3fecVHl5/7/fKcytKEXpSodxcraABU7dl0FRUx2N9kkm3uTm9zceJPcJJubm5vy29zdTTbJZkt2bWABFQVUsNJWAUWU3pHe2wwzMOX8/hjPERCGqUB03q8Xfwwzc+Y75ZzznOfzPJ+HEIKBgQGt7RFkMhnq6uqYkwWbzUZERAS2bduGq1evIisryyADlafCzc2NOWBnZ2erzfoBwFJnLvZ7CzEiV30HXDbQoRjA77NVo4VmS+AEqL6j4OBgBAUFaf2bqa6uhqen5xj50TIoBPK9P4JUogpwOBwWcPp/IGus0XutEREReO+995jbt27dGiMLNTU14eLFi8z+SQhBeHg4IiMjNeqcnO3DZXNycnDr1i2NsoRSqRRNTU0zKj8aMnCaKfnRxKvBrAieCCHBhJBhQojLs9s/JoR0Pfv71ajH2RFC0gghYkJICSEkTJPtj4yMaJyGnzt3Lg4dOoTHjx/j5s2bL0hR9AFGk9Z3Qgh8fHyYLIq+CAQC7N+/Hw4ODoiPj9epNoAQAgcHB8bNub29HTweT+8DDCEEq1atwrZt25CWlob79+/rHTDKZDIQQrSq85lMfvT09MTRo0fR0dGBs2fPam1GqitKpRJdXV3o7Oyc8vMItOfgqJ8QwzLVbslmAQrzEWTIgzHbEsQLFy5kAtTm5maNgnlafhw9wJdGuGYdxOFHIRtRBbZ8HoH8b/8BRY/+BpYHDhzAsWPHmNtnz54dk7ktKyvDjRs3mNseHh6Mj5Y6v6Tm5mZYWVlNm22ELqxbtw5xcXEaNWOUlZXBz89vxuVHeqqBPsyk/Gji1WDGgydCCBvAp4DKfpkQsgHA9wCEA1gGIJYQQvf/fwigA8BcAP8DIJEQMuWcDAcHB+ZgqEkGSigUIiYmBv39/bh48eKYQmtt6xsWLFjAtJY3NzfrlZUBVFf969atw6pVq3Du3DlUVVVp9Xwul4uoqCh4eXlBqVTi1q1bcHd312tNo/H09ERsbCxqa2tx6dIlvVzJ7e3tceTIESawq6qqUps1ousbJns/5ubm2LdvH3x9fREfH69zZ5U2sFgs7Nq1Cxs2bAAhBIODg2o/Ey8hB28HCSEdUf2+WATgmEvxq1wJlLMsgAJURb6XLl3CzZs3p3xsSUmJWndn+x0H0BuwDQqF6oJFwKMg/eO/gJKI9F7nv/7rvyIuLo65/fHHH4/xdsrNzcV4ab+2thafffbZhPusXC5HVVWVQc1XjYGVlRVT96Vu3+nr64NEIoGrq+t0LW1ShoaG9K5RnEodMGFCX2Y8eALwQwD3Rt0+AuBjiqLKKIpqAPA3ADGEED6A1wH8F0VRfRRFJQDoBBA51QvQWZ+GhoZJD4bjobvdbGxsmJb83t5eSCSSMZPbtYFOoRtiTlxQUBD27duH27dvIzs7W6csT3l5OWpqanD//n291zMaKysrHDx4kOl6amtr03lb9HfX2dmJy5cv4+HDhxM+TlP5kXaa37t3L7Kzs5Genm60SfQ0LBYLbDYbFEXhypUrSExMVPt9uVqy8Z1AAcSjzh9m5mL8IkcMmWJ2RVB8Ph+7du3C+vXr1T4VYHnbAAAgAElEQVSuq6tLI/nR6eh30OGwDNSzSNGCK8Pgb9/R3x6BEPzmN7/B5s2bmf+9//77YwwZ09LSxlyMuLm5YenSpRM6UldVVcHT01PvwdnTRVpa2qTDd2eb/Lhp0yZs2bJF5+drow6YMKErMxo8EUL8ALwB4Oej/h0CYHSBSAkAHwB+AEQURTVNcJ9GODk5YcGCBRoHPywWCxs2bEBISAjOnDmDu3fv6nWA2bVrF/bs2QMWiwWKovQOolxcXHDkyBE0Nzfj0qVLaq0IxiOVStHR0YFvf/vb2LhxIwDVlamhGgjYbDbWrVuHtWvX4uLFi3j06JFe23Z0dMT+/fsZL63x22pvbwefz9dYfqQ/O7lcrlWXpT4QQrB27VqNjAc7GqpwZI4EwyPP5ZZmkRLvpAFS3ZrcjIa7uzssLCxAURRycnLQ29s75n6lUqnV7EeX7/0crVwvAIBsRIHiG49Q9dFHeq+TzWbjr3/9K5YsWcKs669//SsjuwkEgjHdaTweD2vXrgWfz4dSqWTqhoaGhtDR0TGh/DhbcXR0hJOT04T7YHNzM6ytrQ1ab6QP9L6h60XNbJEfTbzczFjwRAhhAfgcwPcpihqtpVkDGF2QMvjsf+P/P/o+jRAIBNiwYQN4PB4UCgXy8/M1Kh5eunQpFi5ciIcPH6KxsVHTl3sBMzMz5ko3OzsbFy9e1Lndm4aePScUCnH69GmN66Bo+dHR0ZEZLZKSkoIbN24Y1BfJz88P0dHRePz4MVJSUjA8PKzztry8vMBmsyGTyZCQkICSkhIAqqCvvLxc6/oGPp+Pbdu2Yfny5Th//rxB7CWmYu7cucxJt7y8fMJC8oGBAfT19cHPyw3HlgkhG+GirouHO5WWSK8lOHwB6NFvRrNRoLs2Kysrx/y/oaEBjo6OjHw0FYQQuP7s/9AktUNueg2628Wo+OADtGhg/TAVAoEAx48fZ74DkUiEM2fOwM3NDW+99daEkq9SqURSUhJTFzV69uM/C6GhoVixYsULQTstP063+epUFBYW4u9///uUQ7fHQ6sDs0F+NPFyM2NWBYSQfwWwjKKorz+7TQFwBXAFwE8pisp49v9NUNU3vQkgnaKouaO2cRpANkVRf5tg+28D+A8ANjY2Ng4XL14cc39LSwu++uorrFmzRqNM1ODgIJRKJb766iu4ubnpffCsq6tDf38/Fi1aZLBU+dOnT1FUVITFixerrWNSKBSQSqVjTmYURaG0tBQ8Ho9xZzYkCoUCRUVF6OjowIoVKxjXal0YGRlBXl4efHx84OLiwgRk+hhIikQi5OXlgc/nY9myZVr74+jC/fv3IZVKER4ePua3JBaLYWZmxlw5KyggsW0e0jqee9U484bwo3mP4cTXPNs4HQwNDUEgEDA2E4Dqs9Vldhwll0Py4YdQ0IOJ+XyY/+xnYM+dq/6JGtDe3o5f/vKXTEG4v78/fvrTn05aWF1eXg4zMzO4ublhZGSE8Yv6Z6OjowNcLpfZ/6RSKVgs1qyTH7u7u9HS0gI/Pz+t9muRSARzc/NZEdiuX7/eZFXwEjOTwVMSgB0A6AXwAQwDkAD4b4qi3n/2uO8DCAbwA6hqnNwoiup5dl8hgO9SFJWr7rUm83kaPURXqVROusOVlZXBzMwM3t7ekEgkuHz5MpO1MITj88DAAHp7ew3iQUT7Qfn6+r5wUgbASCshISFq0/Stra2gKIoptDcU5eXluHXrFjNcWNfAkfZ/kkqluHTpEqKiovSWHRQKBXJzc1FSUoLNmzczQ4uNhVKphEwmA5/Ph0wmg0KhQG9vL1pbWycc9fN5IfDrTHqHobDOdwjfCjHDarfZJ08MDg4iLS0Nrq6u8Pb21rkpYaS/H1l79mDoWZ2ija83Vn76d3Dd9a9nycrKwuHDhxn5/Jvf/Cbeffdd5v7xfmhKpRJZWVlYtmzZtA2XNSRyuRyfffYZ5syZg127dkEsFqOgoGDC48Q/I42Njejt7dV7aLmhMPk8vdzM2B5DUdR+iqL4FEWZURRFX+Z7ATgE4B1CyFxCiD+AfwMQ/0zauwzgXUKIgBDyFgArADpXO9OBU3d3N7744gu0tLS88BixWDymvoEe6WJubo4zZ84YpOU9Oztbb0mLhvaD6urqQmJi4gtdK01NTRAKhWoDDYqicPfuXaSnpxukuH00AQEBiImJQVFREVJTU3V+z3TQVVRUhPr6emRmZuq9NjabjfDwcGzfvh03btzA7du39ZZV1cFisZjgOz09HQkJCSgtLZ1UfnxrCfDXKIDPBlZ4DSHQVYLrLX24VGncgnddGBkZQXd3N3p7e+Hm5qbzdnhCIUI//hhsc3MI7QRYvpAP2cfHQEm1G6U0EeHh4fjxj3/M3P70009x+fJlAKoT8UcffTRGpm9oaIBcLseVK1cMsq9ONxwOB/v27cO2bdsAzH75kaIotLe3a/RZy+VyVFdXzzr50cTLy6zbayiKugbgBIAnALIA/JmiqDvP7v4+gAAAPQD+BcDrFEXp7XjIYrFgaWk5YU3G6Pl1NGw2G5s2bcKCBQuQkJCA5uZmvV5/w4YN2Lt3L3Mi1TcbKBAIsHfvXsyZMwenTp1iTCbpA8xU7sGEEOzevRu7du1iitsNaSxJDz6mJ95rO7ePpre3F0qlEm+88QYiI1VNl4YI9tzd3REXF4fBwUGt6sj0YeHChbC3t4eHh4dayTDKF/hstwIhbqqiJwGXwpOBfnxSNLtO5nZ2dggJCUFERAQIIXoFG9b+/ljym19j5UZvCCx4MOcqMPjev8EQ5lff/e53x3Tg/fu//zsyMjLw5Zdfoq+vj7kAGRkZQX19PebPnw+FQvFPGTwBqosrLpeLjo4OUBTFXEDORtrb23Hq1CmNZmdWVVXBy8tr1smPJl5eTONZJqC2thbe3t7o6upCfX09QkNDJ31sXV0drl27hvDwcI27idRRXl6OoqIi7Nq1yyAeJdXV1cjIyEBYWBjYbDbMzc217hLKzc1FY2Mj9u3bZ/DhmmVlZbh9+7bWMh5FUcjOzsbixYuZehq66N3S0nLK1nlNX6O4uJgZJ2LI+rTxSCQS3L9/H+Hh4ejs7ER3dzeCg4Mnffz9VhmuNw2Ax1Htv0oKYMks8OPXBGDNfLc5Wltb0dbWhiVLlqCyshI3b95EdHS0XmasLf/7I8wZee46PuCxAtZv/qfea+3v70dUVBTq6+sBqLzZIiMjme5VHx8fLFiwAHZ2dnB3d9dpZNBsor+/H++99x4OHDgwaySuiaAoCiUlJZg/f77aYyEtP9KB+mzBJNu93My6zNNMU19fj4sXL6K0tBSlpaVqT2AA4O3tjYMHD+L+/fu4e/eu3pkPiqLAZrMNdgXl4+OD6Oho5Obm4vr165irQ7Gtra0t7O3tNXIp1pbAwEDExMTg0aNHSEtL0/iKvqmpCTY2NmMKkSmKgrW1tcHqUQghWLhwIWJiYvDkyRMkJycbbcBwWVkZAgICwGaz8fDhQ+Tm5qpt1V7hysVhHxtIRlS7MIsA4Inx8xwxpLKZvSBSKBSoqKhgDCQdHR3h6empU9H4aFyP/R5tA8/3C6uGexjKTNFrm4DKFPfTTz9lMn7FxcVjssnV1dV48uQJIz8SQqBQKHD79m2tRgfNFjo7O+Hs7Kxx9+NMQQjBggULpryInEgdMGHC2JgyT+OgKApVVVVgsViQy+Uaa+gSiQRXrlxhHLz1KSSnr2xlMhlaWloMUkiek5PDXFnv3LlT5+LqwcFBNDU1GdxZWSaT4fbt22hqasKOHTvUminKZDJkZ2cjLCxMbSaMNufU1dR0NKMHDG/dutWgHj89PT2oqKjAypUrmROzSCSCUChkJNPJAtcuiRIfFQ3AjP+8NqtHxMd/LreEnWBmTiaVlZUghEzYtalQKNDf369zBmqkuwviX38dtjaq710mp4C33wPXQ/95bElJSfje977H3P7BD37ATCQghODo0aPM9y6RSHDq1CmEhIRgxYoVer/2dDEyMoKcnByEh4cb5WLI0CiVSlRVVUEgEMDDw+OF+zs7O6dUB2YKU+bp5caUeRoHIQReXl5obGyEm5sbkpKSNKp5oefOWVlZISEhAT09PXqtAQDy8vJw4cIF9PX16bwtQNWBx+PxcPjwYfj5+SE+Ph4NDQ06baugoAAZGRkajbnRBi6Xi82bN2PlypVITExUa6pZVVUFb29vtYETRVG4c+cOrl69apA6qNEDhq9fv26wYnJamhhtvspmsxnjxnv37uHs2bOTZuQcBCz8eJkQ8pHnGRk7y2H8rmAAdb2GLfbXBIlEgpaWlkndnW/duoWEhASdx/bw7B3A+fovIRlSZeW4HALZ3//TIAXk+/fvx9e+9jXm9p///Gcmi0lRFBITEyESqUbFCAQCxMXF/VMFToCqLMDX1xccDgdKpRIVFRUGbwoxJIQQZGVloaio6IX7lEqlRuqACRPGwBQ8TUBZWRl8fX0hkUjQ09OjsZTEZrOxceNGLFmyBGfPnmUyPbry2muvYdeuXYyxpi5ZQnpAJp3WDg0NRVRUFK5evYp79+5pvc2IiAjExMQwKX9DjzYJCgpiZLIrV6684JouEonQ1dU1ZTZuoqJ3Q5wk6AHDIpEIp06d0rnYnaaxsRG2traTSlq0M7Q6GdecS/CTUCtwlc8Lze0sZfiwaBAlxjdOH0NpaSkCAgIm7eAKDQ3FunXr9Krns1ocisGl+5gZeOZcBUR//IFBCsh/+ctfMg7kcrkc8fHxjJwnFouRlJTE/I7o//f09Gg9Y3ImGBgYwMDAACPd19bWIiUlBbW1tTO8sskhhODAgQPYvn37C/fV19fDycnpn9Zzy8Q/N6bgaRz9/f0QiUSYO3cuHB0d8eabbzIHm66uLo2CjUWLFmHnzp24du0aCgoKdO6e43K5zBV8W1sbzpw5g4GBAa22UV9fD2dn5zEHGA8PD8TGxqKurk7r4b1sNpuR1CorK/HFF1+8MI5DX+huPEtLS5w8eXKMhYQ29Q0CgQD29vYAVIaUiYmJBgn2BAIBduzYgdDQUCQmJiI/P1+n71gmk6GmpkZt96OPjw82bdoEQgjEYvGkDvdsFsEPl1rAkaP6noflBHeqLHDgPHBXtySj1vT09EAmk6mVSW1sbJhMQWdnp84ZWqeYN9HGf57dspJ1YvDLP+i0rdHw+Xz8/e9/Z2TFtrY2FBcXM/fX19fjzp07Y56TlZWFW7duGdXWQl8mml83f/587Nu3b9bPgBMKhS8E4yMjI2hoaDCKoa8JE5pgCp5GMdFwWdokr729HSdPnsTjx4812pabmxsOHTqE0tJSXL9+Xe8Dq1QqxcjIiFZ1CsPDw2hoaICPz4v1IPTwXhsbG5w+fRrt7e1ar8na2hqurq56FwJPBIfDQWRkJNavX4/k5GTk5+ejvb0dbDabCYi0wcrKCjY2Ngar8yCEIDg4GIcPH0ZNTQ3Onz+vdWBbWVk5pfw4mszMTCQnJ086w5AQgrcWmiPA0hJZlVboHeJALAPeSAbOlWi1NK2h5UdNJRSKopCamoqrV6/qfHHh+p9/QFv/889uICsDgxq0tU/F3Llz8cEHHzDHgFu3bo1ZY1ZW1phC8Y0bN+Lw4cOzuoaotbUV5ubmYwYhE0Lg7e39T1Fo/fjxY9y+fZu5TasDs/kzN/FyYwqeRtHa2goLC4sxBxgaJycnhIWFaWXCJhQKERMTA5lMhnPnzjH1Errg5eWFo0ePwtzcHBRFaSQJlpeXw8/Pb9IDDJvNxvr16xEWFoakpKQxV9ia4OLigp07d4LD4UAmk2k8K1AbfHx8EBsbi6qqKnz22Wc6F2oHBwdj8+bNIIRAJBK9MH9NV2xsbHDw4EF4eHjg9OnTqKio0Oh5IpEI3d3dWjUDrF+/Hnv37p1ydMweXzP8ZQsPc541HSoo4D9uAP93T2EIZWtCnj59qlZ+HA8hBDt37kRUVJTOJ28WhwO7n/0FPb0ylDxowcO7dSh45x3Ino1c0QcbGxt897vfZW5/9tlnsLKyApvNRlRU1JjsmoWFBfO+JzLanWnGdz+Op6KiAufPn5/VtU99fX3o6OiAUqkcow6YMDFTmIKnZygUClRWVk56gKHrhegJ6zdu3NBIruLxeNixYwfmzZuH06dP63VwpU8y5eXlSEpKUhtA0QcYTcarBAQEIDo6Gvn5+UhPT9cpS1ZdXY2srCyjnDysra0RGhoKb29vJCYm4umzUR26kpeXh2vXrhms6J3FYmHlypXYs2cPcnJycPXq1UmzQzS6tFebmZkxJ4yqqiqkpqZO+l35OwAXo4EgB9VtB0s5ROw+/DJXjBG5YSMomUyG2traKc1Xx2Nvb8/MWCssLNTJrZ/n4Aj2N3+H+lrVhYmopgaFP/oRKD0CAVp+PHbsGGOgSVEU/vGPf2Djxo0IDQ2d8HsrKytDQkKC3r9PQ1NdXQ13d/dJO4AJIVAqlToX8U8H4eHhiI6OBiHkBXXAhImZwBQ8PWOqA8xo+vr6UFlZqbGzOCEEK1euxIYNG3Dp0iWUlOinofj7+yMqKorJWoyXPSaqb5gKe3t7HD58GMPDw0hISNC6wy8wMBBxcXHMDDNDOjAPDw+jqakJBw8exLZt25CWlobs7Gydr5TXrl2LgwcPMkXvhqpVcXV1xZEjR8DhcHDy5Ek0NTVN+LiOjg6d5UcauvhXneTlYgmcex2I9FYgKrgfPA4Fc3MJfp4rQp/UcAFUZWUl5s2bp7OBqlgsRk5OzoQdVZogXLAAi377W+Z2+40bqPrLX3QqIB8tP7JYLHzwwQdMtrO3txfHjh2bNIPs5+eHjRs36jWOxtBIJBK0traqndPo6+s7Zn+YjdDHsZaWlknVARMmphNT8ITnBxhvb2+NHm9nZ4c33niDcRTXNIPh4+ODgwcP4t69e7hz547OJ38Wi4XAwECmiPjs2bNjur5aWlpgaWnJtLtrCp/Px44dOxAUFISEhATU1NRM/aRR0KMeOjs78dlnnxmsi2e0/Ojp6Ym4uDi0tbXh3LlzWtcZASq5kpZdqqurcfz4cb3tIGh4PB42bdqE9evXIyUlBVlZWWOkTEO1Vy9btgwHDx4El8uFXC6fdP1WfOCjrQRc8ly6tbUaxv/mD6CxX3+ZhpYfJ/Lg0RQLCwvExsYiLCxM52247dqFeW+9BQBgsQj49xIxeOI9rbczXn4cb6BZXV2NH/7wh0zQKpfLmfonNpuNRYsWgcViYWRkZFbIYKWlpQgMDFQ7v44QwozQ6e7unsbVaUdpaSk++OADjY/TJkwYE1PwBN0GZNKt1gMDAzh+/Djy8/M1ep6DgwMOHz6Mrq4uJCUl6Z0qHxoaGuN6TcuPug7IJIRg2bJl2LVrF27cuIGsrCytTwIWFhbw8vKCs7OzTmsYzUTyo4WFBfbv389IofrUL5mbm8PBwcFgruQ0Pj4+iIuLQ3d395j5eHV1dXB1dTXI6B26meHu3buIj4+f9LdkwWPhp6HWYCmeZ1XtLGX4a3E/itr1q1EzlIRia2sLFosFiUSC69ev67RfBP74x3BZsxKrNs+Dp68dLGtzIMm+pvHzJ5Mfg4KC8Ic/PO/kS01NxSeffIKenh58/vnnOH78+BgJf2hoCCdPnsSDBw+0fg+GpLu7GzKZTOP98OLFi0hJSdF7tqax6O3tRVBQkEmuMzErYL/77rszvQaj88knn7z79ttvT3hfd3c3uru7ta7XoOFyuZBKpQgMDNT4hMjlchEQEICenh7cvn0b7u7uOqfMLSwssGjRIubkn5mZCTc3N7UO3ZpgbW2NoKAgPHr0CMXFxVoN3eRyufDz8wOPxwNFUcjLy4Otra3WI2coisKDBw+wYMGCF4qkCSGYO3cu3NzccOPGDXR1dcHDw4MJKDTFysqK8SWSy+V48uQJnJ2dDXKA5vF48Pf3B5vNxtWrVyGTydDd3Y2lS5cadJK9ra0thEKhWrmIRQhWu/JQ0kkgpVR2DWZcCkVdI1DKuZhno/162tvbIRKJJuzm1JX29nbk5OTA3d1d68wpYbHgsCYM8txUCMzYqu+wsgCU32tgWdtO+fzy8nI4OztP6H4eFBSE7u5uRlrMzs6GmZkZenp6oFAo8PTpUybrxOFw0NfXB09PT63fg6GgKAoFBQVYtGiRxvudUCiEr6+vUbpn9UUikeDp06fYvXv3lA0Ts4Vf/epXre++++4nM70OE8bhlcg8icXiCa+mtG2vnggWi4WIiAim8DU/P18j7xr6eeHh4UhMTERZWZleawBUkkNSUpLBfJfMzc2xb98+uLu749SpUzoVwvb29uLevXsoLy/X+rmayI+urq6Ii4uDXC7H6dOn0dmpuytkeXk5bty4YdCid3o+3uHDh5GdnY2qqiq9ui4nwtbWFosWLQKgkkwfPXo06Vq+HWKO+RaWUD7bHSz4SuR09uN0yYhWrznafNWQuLm54Rvf+IbOMiDfyRmsIz8b40A+8rdjoKTqZxJqIj+ONtBUKBS4cuUKs++1trbi+vXrAFSfc2RkJFP/NxM8ffoU9vb2WmVU3d3d4erqasRV6c5odWBwcNDg5rwmTGjLKxE8SaVSZGRkvNBGr8sBRh1DQ0PIz8/XquU/ICAAr7/+OnJycvQeLNzR0YHDhw8zB3hDpN9ZLBZWr16NrVu3Ii0tTWtXcjs7O3zta1/D0qVLAUBjOUYul2ssP/L5fERFRSE0NBTnz59HYWGhTu89ODgYhw4dYjraDG16uGrVKixZsgTx8fEoLi42ijxSVFSEe/fuqe32O+BnhpUO1pApVNk1PodCvXQAHxYMa1xjXVtbazD5cTz0Nuvq6pCSkqK1/YVw+SoMhOwa5UAuh+j9H6otINekwWK8gWZlZeWYC4qCgoIX9v2CggJkZWVptX59oeVHPz8/rZ87PDyMzMzMWdUx2N3dDYVCAScnJ7S3t+OTTz6Z1a7oJl4NXongyc7ODkNDQ0hMTGRO3vocYCbD3NwcR44cYQpfNe04c3JyQmxsLDo7O3Wug6IPMKtWrWKKiJOSklBtANNAQOUzFRsbi/r6ely8eFGrNdrY2IAQAolEgpMnTyInJ2fK51RXV8PDw0OrAcvBwcGIiYlBSUkJkpOTtf4cCSFMbVVnZyc+//zzSTvmtIHOcIaEhOC1117D/v378eDBA1y+fHlMvZohiIyMxKFDh2BmZsYMFZ7wcR487PYUYlimOgSwWcDdRgX+6zYgnyJ+Hx4eRmNjo0Hluono7+9HT08PRka0y4oBgPORt9HG8WJuWw23Y/Dk/0342Pb2dnC5XI2GFc+dOxd//etfmYxTWlramCD48uXLY2Zh9vX1oaenZ1qLxysqKnTufmSz2SgvLx9jAjqTjFcHHB0dERERYZBh3yZM6MMrETwRQrBr1y64uLgwQ3srKiowf/58ndurJ8Pa2prptklISEBmZqZGzxMIBNi3bx+cnJxw+vRprWamTSQ/joyMGLzjx8rKCgcOHIC9vf0LY1M0gc/nIygoaMpxEENDQ2hvb9epq4Ye7WJnZ4cTJ07ofAXN4/Hg6OhokJbo5ubmMfIjHSzb2trixIkTWnc1qoPFYjGvU1BQgPPnz08axIc4cvBWoBCSETZKW/l48FSAU0+Ab6UAQ2pUkbKyMqaWy5gsXrwYsbGxEAgEoChK60yd60/fQ1vf8zVa1mRBknt9zGNo+TEoKEjj7YaHh+PYsWPM7S+//JIJ8mUyGc6fP8/ISuvXr2fmK04Hg4OD6O3t1Vn25HA4+PrXvz5rBh6PVwdYLBZCQ0NnrJbMhAmaVyJ4AlQ73dq1axEaGooTJ06goqLCqDUJHA4H8+bN0+ogRq9xzZo1SExM1LhOqKGh4QX50dzcHDExMUxmrbm52SDeS2w2G2vXrkVkZCSSk5Px4MEDjU9qLBYLYWFhzFVjYWHhhF5ZJSUlU7ZXT7XGiIgIbNmyBWlpaS/YBWiCUCjEvn37YGlpCYqiUFRUpFOdhVwuR1VV1QvmqxwOBxEREdixYwdu3bqF9PR0g3pjAapA3sbGRm3BsJsVGz9cLISQWAJQSVY36oCYJKBngsRdX18fhoaGpq02hs1mQ6lUIj09XWv5i8XlwvYnH6G3T/W9EULASf0b5M11zGN0lR+/+93vYsuWLQBUAVNSUhITTHZ0dCAtLY1ZP20p8vDhQ61eQxd0MV8dD/17mWnTzMnUAYqi0NTUpFd9owkT+vLKBE80CxYsgJeXF+rr6/HkyROjvQ5dEE4b7FVWVo5J56sjMDAQr7/+OrKzs6esg5LJZKirq5tQfqSDj+HhYVy6dAk3b97U/o1Mgo+PDw4dOoSysjJcvnx5Skft8cjlcjx69OiFWYFdXV1QKpV6dwsCz0fadHZ24syZMzoX0re1teHmzZs6mZtWV1fD09Nz0gDGzc0NR48eBUVROHny5KSDf3XB398fW7duZSTTyU42Qj4LH24l+M6y5/970kHhWOYQmgafB8a6mK8aAkII2Gw22Gy21tknvrMLWId/AildQM4lGPrbz0BRFKRSqc7yIyFkjIFmS0vLGMuMR48ejbn4KS4uRmZmpsGHaI9GG/lxKmpra/Hxxx9rlQE3NJOpA0qlEpcuXZqWYNSEickgs9XTw5AsX76cKigoAKA6Eba0tMDLywvJycnw9PTEunXrjJpWl8vl+OKLL+Dk5ITdu3dr/DyJRIK0tDQoFArs2LED5ubmLzzmyZMnEAqFU2a4GhsbYWNjAysrK1AUZbCTn1wuR2ZmJmpra7F9+3atMhJSqRQsFgs8Hg8SiQRsNhv37t3DsmXLDOp2TFEUCgsLce/ePUREROh0Zd7a2goXFxcQQqBQKDSSrOgGgvDwcI1+XzU1NcjIyEBAQADWrFljUEn58uXLaGlpwZtvvqk2E3WiCPjlHQqRASL4Og2jR8zBt4KtMd+WhaamJnR3dzOdfdPJ6N+spp//aNpPfAzHqqvo75HgQeZT+PoK22sAACAASURBVPzkF+gNDISzs7NGI4wmo7S0FNu3b2fqst555x2m0WDevHmIi4tj1tzf32+QwGYilEolMjMzsWLFCoMU8UulUuTm5iI0NHRGrAsGBwfx6NEjhIWFTbivtra2ws7OTquayOmGEPKAoqjlM70OE8bhlQqexh9gpFIpUlJSQAjB9u3bjeofMjg4CA6HA4FAALlczqTzp0KpVCI3NxelpaXYuXPnmOBkqgPMRFAUhYyMDFhZWWHVqlU6v5/xVFVV4caNG3jttdewdOlSrYITiqJw7tw5dHV1Yd26dQZvf6fp6upCSkoK7O3tsXHjRp1OMhKJBAkJCVi5cuWUdTL5+fnw8vJinNc13f7NmzfR2dmJrVu3GkweGxgYQHd3t0Z1ZKdKRtA08ty5vW+Ijdj5Fugqy8GaNWu09usyJH19fbhw4QLWr1+vdU1c9U+/j/LzqaCUFAiXC+uf/xzhhw/rfSHx8ccf49e//jUAVe3i22+/jWXLlmHt2rUTDuXu7u7WazTPRFRXV0OpVBq0AWamoCgK9+7dg7+/v9GCzenAFDy93LxSsl1NTQ3mzJnDnDTNzMywb98+2NraMoXkxsLKyoopfE1NTcXVq1c1kiDoOqHIyEhcunSJkbloCUXbLArdgWXo7h9fX98xMp429RKEEISEhMDa2tqoB38HBwccOXIEFhYWeklkowfaTkZXVxcoitIqcAJUJ98dO3Zg9erVSE5ORnZ2ttb1WhNhbW3NBBu1tbUoLCyc9LFHgnlw4z/P/NmYKxBfI4LYxndGAydA9fkIhUKdAl/vX/weVv4q6wtKJsPw3/8OmQ7jfcbzzW9+kymwlkgkuHr1KlavXj1h4FRfX4/jx48brAsWUGWJmpqapmzE0IXu7m6UlpYafLvqaG9vB4/HmzJwKi0tfUH2N2FiunhlgiepVIrm5uYXDjAsFguRkZFYtmwZzp49Oy3+JnPmzGEkIE3x8fFBdHQ0Hj58iPT0dDQ3N4PP52t9ZcZisbB161asXr0agOokb6ig0cbGBjExMbC2tsapU6e0aneWy+VYt24duFwuampqkJ+fbxQfJA6Hg8jISGzcuBGpqalaF5MLBALs3r2byQhVVla+ECgqlUq9zVf9/f0RFxeHzs5Ovc0/x1NWVobi4mK17/tIkAA+FpaMNZLQXImvpObIadI/kNMHPp+P/fv3M00H2hTxs83MsPyjj8B6JglLm5tR9bMfgBrWzy6CzWbj/fffZ2T1yspK/O53v5vwse7u7ggLC9NrFuB4jNn9WFBQgDt37hjc82wylEolysvLNep+rKqqmvbAzoQJmldmPMvKlSvh5eUFa2vrCR/j7OwMZ2dnpKWlgcPhGM1HhB4rQp98W1paMDAwMOm6RiMQCBAYGIiysjKkpKRg06ZNOtUG0YNAKYrC5cuXUVJSgkWLFhmkDorFYsHb2xvW1tZIS0sDIQSurq5qtz0wMICGhgamELmgoAANDQ3MZHtjYGtri6CgIDx58gQPHz6Em5ub1tmMoaEhnD17FsPDw2Om1tfX14PH4zFmm7rC4/EQEBAADoeDa9eugaKoKT9LTfDx8YG/vz94PB6USuWkNXBB9hz0D7HRJh0BISozzbLeESjlPJ3GuRiaJ0+eIDU1lRkFpAksS0s0SiSQFxRgXqADAudxIH74FfhrogA9PlcbGxvY29sjIyMDAPDw4UOsXLkSHh4eoCgK9fX1zPy+uXPnMsXv+n6Xvb29aG1tZQaFGxoXFxcsXbp02mqLqqurYWVlpdHx19vb22DHLWNgGs/ycvNKBE9/+9vf3t26dSsCAgLU7mhCoRDz58/HnTt30NPTA09PT6MWklMUhatXr6KiogIhISEaHQQ4HA5YLBbYbDby8/P18iIihMDT0xNeXl5MITn9f32xt7eHr68vvvrqK9TW1sLT03PCAmi6mDsoKIgJXubNmwc/Pz/w+XzI5XIMDAwYpR6NnjGoVCpx7do1mJmZwcnJSeP3z+Vy4eXlhYCAAKalXiaT4cmTJwabX0cIgZOTE/z9/VFQUIDHjx9j7ty5ehUFE0LA5XKZ+rfi4mL4+/tP+L5tFb1oae3DAEcAFgF4HArVA8MQSXnwt5v5AEokEsHPz0/jrEtFRQWcQkLgRMTwc5WBEAK+XARRdx94QfqVpyxcuBCPHj1CXZ3KCiE3Nxfbt2/HlStXcPv2bTg5OTEyrlgsRmJiIvh8vs71T/Tsx4ULFxqtXpPH4zH7rSEbTSZCKpWirKwMixcv1mjf4XA4szZwAkzB08vOzB/9pgGJRKJxbZCtrS0OHTqEvr4+nd2+NYU276RN9NQ5QtNIpVK0tLRg165diIqKwtWrV5GXl6ezxGVtbc1c5T18+BDJyckGmxtFy3hCoRAnT56c0NOpra0NfD5/TA0RIYQJDrKzs3H69GmDO3GPfq0lS5bg4MGDePjwodb1Wi4uLuDxeFAoFEhMTERSUhJ8fHwmrHfRB2traxw4cACBgYE4c+aMVv5ak0EIgYuLC1xcXCY8WSmVSpSWliJ66RysdbGG/NlP05xH4UFPP86XTY+UMxnOzs6IiopiHPWnkpbEYjE6Ojrg6ekJz//8H7T1PD8emD++hpGSfL3WQwjBe++9x1zMNDc346OPPmLqm1JTUyEWiwGossj6ZnOamppgbW2tUdZaH2QyGS5cuKC2Ts4QlJaWai0/FhcX4+LFi0ZclW4Yok7RxOzmlQie2Gy2VgcYMzMz7N27F87Ozjh9+rTG/ky6IBAImLql3NxcJCYmqg1eSktLmUyHh4cHDh8+jOrqaiQnJ2vttTQeOqNlyBM/m83G+vXrERkZicuXL+P+/fvMSV+hUKCiokJtfcPixYuxevXqCW0aDImDgwNiY2NhY2OD48ePM9kDTaEoisk+ubm5GWWNdKB36NAhVFVV4dy5c3r7Bi1atIjpuuzt7WVO7oDKfNXJyQkWFhYIm8vDVnchMw9PriT470yCs9pbXxkcpVKJCxcuID09XW1AOXq4LIvHg/BH76O/X2UxwGIRKON/C0rUp9danJ2d8dvf/pa5nZCQwEiKQ0NDjHkmi8XC/v37dW6QkMvlqK6u1mj2o75wuVzGZ8tY9Pb2Ynh4WOtyCYVCAYVCodMIH2Nimr338vNKyHb/+Mc/3v3Wt76l1XMIIfDy8gKfz0daWhpsbW2N3jYrFoshl8vh4+MzYZasp6cH7e3tY+RHeuRJa2srsrKy4O7urrNHkqurK/z8/BhTxYaGBoO9Zzs7OwQEBCA/Px9lZWXw8vJCQ0MDrK2t4ezsPOnzzMzMmPqw7u5u3L59Gx4eHgbP7ACqE5qXlxecnJyQnp6Ovr4+uLu7a3TSIISgt7cXq1evhkAgQFNTk0oSMkKtiEAgQHBwMIaHh3Ht2jWw2WytGxDGo1Qqce7cOVRXV2PBggWM/LhkyRImKzXHkg17HhcPO+S4/ESIASkbGbWAkA8smR7D8QmhHbydnJwmNVft7OxEX1/fmGCFK7SBGFbg1j4Am80Chw2I87LAC9upV/2Tv78/ampqUFFRAYVCAbFYzBSId3Z2wtHRkVknRVGoqKiAWCzWSn4vLy+Ho6MjHBwcdF6nNgQEBBitDlQf+dHFxQVBQUFGHxWkDbT8ePz4cZNs9xLzSmSe9DmpBAcHY+/evbh58ybu3btnlA4wmsDAQGzcuBGEEAwODqKiooK5j55fN5G7M5vNxoYNG7Bq1SqcP39eJydsGnrb9+/fR0pKCkQikc7bGg89G2/OnDn4xz/+gcLCwjGF1lPR3t6OpqYmg8mKk+Hh4YG4uDhIJBKcPn0a7e3tUz6nra0N5ubmsLW1hUKhQFpaGtLT0422RkIIli1bhpiYGFRUVODcuXPo69M9a8JisbB582Zs2LABhBCUl5fD19f3hSB1gQMX3w+xwVzL5yerX2UCf8oDZtIy7rXXXmMymOPH3NDy44IFC154nv3WXWi3WcjctpT3QHRq4gHC2vCb3/yGuSgoKSkZk9FLS0tjbiuVSnz11VdatdyLxWJ0dnYy7ubTBUVRaG1tNfgxsKmpCUKhUC/5cTbJZLT8aOLl5pUInvTF1dUVsbGxqKmpQUpKyrSkiPPy8pCens7U+ox2CJ+MoKAgHDx4EPfv30dGRoZe7cVhYWE4cOAAMy/PUAcnFouFNWvWYP78+WhsbEROTo7G2w4KCsKbb77JFLe3tbUZZE0TQfstrVixAhcuXEBeXt6k3li0/EjPr2Oz2di3bx82btwIAEYNuO3s7BAdHY358+cjPj4ehYWFOr/enDlz4OzsjIGBATx69GjSiw5HC4Iz+4Glo7JNZ8pG8Lt7wzMaQAEq5+nPP/8c9fX1zP/q6+vh5OQ0qfQ799//Gy19z5sZLGqyMVxwR6912Nra4o9//CNz+9y5c0wgOjQ0hNTUVEbq3b9/P7Zv367xtg0xv04XysvLER8fr/VAcHXQ8qM+wUZZWRn+8pe/jAlQZ4qenh6d5EcT/3yYgicNsbS0RHR0NDgcDs6cOYMBA5jrqWPdunU4ePAgzM3NIZPJUFlZqVF9A127MzQ0hDNnzqC/v1+n1+dwOEyrfX19Pb788kuD+UH19PTA1tYW3/nOd9DZ2Ylz585pvE6686eiogKnT59GQ0ODQdY0GYGBgYiNjUVdXR3Onz8/4Tpramrg5uY2RnJwcHBgJr/funULmZmZRguiWCwWli9fjpiYGJSVlU26Tk2ghyDT7vWTIeQDp/YAa9wBZysZtgUPAPxB/CpHCuUMBlB2dnbw9PRk5OaRkRE0NDTA19d30ucQNhv2P/kTentVF0WEEJCLH0LZo99ct/Xr1+PIkSMAVEXXN27cYO4rKytjMsTW1tZgsVgYGRmZsoats7MThJBpk+tG4+Pjg02bNhlk7iRNZWUlvL299TJfdXR0RHBwsMGNf7VFnTpg4uVjRoMnQshdQoh01N8/iIr/I4T0E0JaCCHfHvV4b0JINiFEQgi5TwgJVLd9Q8PhcLB161YEBQUhPj4eTU1NRnstNpvNpP0zMjJQWFiocRcYn8/Hrl274O/vj4SEBK2Ln8fD4/FgY2PDZKH0YfQBxsLCAvv27YOPjw/i4+NRVVWl8XZ8fHwQGRlpULPByaA73ebNm4f4+HiUlJQwgZBEIkFLS8uk8iNFUWq9lAyJnZ0dYmJi4O3tjdOnT6OoqEjrgK21tRWWlpZ44403sGXLFgCTZ84seMDnOylsCxKDy1aVCZlZiPDzLAnkM3Qe4/P52L59O6ytrUFRFB4/fjyh/PjC85xdQB34EYYlqmwtm6VE01/+oPd6fvGLX8DT0xOAKlAYHRylpaWNkcUvXryI5OTkSYMAWn401viiqeByuQgJCTHYzEWxWIyuri7m89EVBwcHbNiwYUZm8I1GE3XAxMvDTGee5gEQUhRl9uzvTQBvAtgIIAjAFgD/QwhZ8uzxJwDcBeAMIBFAwnQvmBCC5cuXY8uWLbhy5QqKioqM+noikQhSqRR+fn5adZwRQhAaGoodO3bg+vXryMnJ0fnKbM6cOdi/fz/Tkl9YWKizjNfY2AhbW1vmAEOvc8+ePbh79y4yMjI0qmnicDhYsmQJU9yemJhoUBfu8bBYLISGhmL//v3Iz8/HlStXIJFImO7HyXxpCCHYuHEjIiIiAKi6inTNCmmzzujoaBQXF+P8+fMa10IpFApUVlYiMDAQ5ubm4HA4kMlkSExMxJMnTyZ8joBL8MMl1hiRPa+BsrIS46eZQ5DKZlbDu3btGq5cuaLxiBy7sPXoclsF8cAwcq7VoOjLi2i5dk2vNVhYWODDDz9kAudLly4xxc0SiQS3bt1iHrtq1Sps2LBh0t9SfX09nJ2djd55OhVVVVV48OCB3tspLi5GUFCQwS4qenp6jF4PORkymQw1NTXT0v1oYnYwY8ETIcQMgJyiqOFxdx0B8AeKopopinoCVYAUTQjxALAEwH9TFDUA4I8AvAghM/Jr9fb2RnR0NAoLC5Genm608QUlJSVYu3Ytdu/ezZzMtKlrcXNzQ1xcHJqampCUlKS3X1JtbS1u3bqlk1xGH2Amqm9wdXXFkSNHMDIyovU4EpFIhN7e3mmpRXNycsKRI0cgFArx8ccfo6GhYcr6htGO7levXsXFixeNWgcFqExKDx06BG9vb41roaqrq1+QHwFM2aYu5LPwoyVCyGXPszt2wiH8LGcIQyMzE0BRFIWhoSGsXLlSK0lozr/8BKViN/T3qLK8RceOQTyqfkoXQkND8Z3vfAeAah+49iwgW7RoETZt2sQ8zsPDA+7u7sz6R0PLjz4+PnqtxRDU1NSMyb7qQkdHB1gslsHkx+bmZnzxxRdGl/Eng5YfDZWVMzH7IcY+iE/6wqqg5yaAdgBeUGWUvgOgBEAERVElzx73bagyUccB/IKiqNBR27gH4H8oikpR91rLly+nCgoKjPE2mHbxoaEh7Ny50yDSFk1HRweePn2K5cufOx8XFxcjPT0d0dHRWo3/UCqVyMnJQWlpKXbs2KHX6JC2tjYmYBgZGdH45FRSUgILCwu1XUK0rJeZmYlVq1Zh8eLFGl2ZKhQK5gTf1NSEOXPmGN0d/sKFC2htbYWfnx/WrVun0efQ09MDiUSi9+gWbejp6cH169fBYrGwadOmCe0nJBIJ7t+/j4iIiBc+t9GSo7pRQsNyCn8sHACL8/zqv6PPDL9cZQGh2fTWgLS0tKCjowOLFy8GoApaND2xyQYGkLlrF4aeDY62DgxEWGIi2Hq4eA8PDyMqKgrl5eUAgDVr1iA+Pn5COfHhw4dobGzErl27mM+9qKgI9vb2RvMQ0wapVAo+n69zxkipVCIrKwuhoaEGy6IpFApGojXkMVgTRCIRHj58iPDw8DGfCSHkAUVR+tnWm5i1zGTwtArAfwP4PoAmAH8CMBfAOgDeFEU1PXtcLICvQRU8vUFR1MZR28gA8AVFUfETbP9tAP8BwMbGxsbBmC60FEWhvLwcdXV1WLFihc7jFsYjEolgYWExZoekKAp9fX2MI7dcLtfK86i1tRUPHjyAn58ffH199UqZi0Qi3LlzB4sXL57yoK5UKjE0NKTxgW1wcBB5eXkwNzfXarbW4OAgMjIyEBgYyHS/GYORkREolUqw2WwUFRWhq6sLoaGhWn33dXV1EIvF09I5pVQqUVNTg/LycgQEBGD+/PljgqShoSHweDy1vyWRSISbN28iKCho0gJsBUWQxfaDmdXzQKWpk4UtrFJYcabPkXxwcBCWlpaMrJuZmQk/Pz94e3tr9HxFfT2GfvtbQC6HuRUPARsXombb21CydfcXq6+vxy9+8QtG8j548CB27979wuOqq6vR2dmJ0NBQcDgcKBQKSKVSnf3bjAVdBqDtRcrw8DAoijLaSJnpRiwWw8zM7IXs7Pr1603B00vMjAVP4yGEOECVheoBsJqiqKpn//8mVJmnEwB+SlHUmlHPyYEq83RV3baNmXkaTXV1NTIyMhAeHj6hp4w21NTUQC6Xq23h7ejowIULF7B9+3Ym3a8J/f39SElJgaWlJbZs2aLzQUwikeDmzZsIDw9nOssm4/79+/Dx8dEquJDL5cjJyUF5eTmioqI0fo8VFRWMwakxCrVlMhmys7MRFhbGZDOqqqpw8+ZNBAcHY/Xq1RqZ9t24cQMDAwPYs2ePUbNko+nt7UV6ejqUSiW2bNkCOzs79PT0oLKyEitXrlT7XIqikJubi5CQELVFsQqKwoeFIoyQ54p8ez8Px5ZbwcXS+BmoiooKsNlsRuJSKpW4evUqFi9erFXGrz4+Hp2f/D8sXu0OLo8NkZ0vLL+nXxH5n/70J/z+978HoCrATklJYY4VYrEYFhYWY36z9GceHBys8wxLY9DX14fExESEh4drZTMwPDyM3NxchIeHG9zoVqFQMAOYjW1oTDOROkBjyjy93MxkzVMUIWTjqH9xAcgBlAMYHXkEASgCUAEgkBDCevZ8FgB/AJq7yxkZHx8fREdHIy8vDzdv3tS5qHp4eBhPnz7F/Pnz1T5OIBDA1dVV6wOFUChETEwMrKyscOrUKY1MICd7/R07djCBU35+/oQWDh0dHWCz2Vpn5DgcDtauXYvNmzcjNTUV2dnZGhW9+/v7g8/nQ6lUIjk5GaWlpVq97lRUVlZi3rx5Y2QgX19fxMXFobu7W+ORPhs2bGDmGkqlUgwODhp0nRNha2uLgwcPIiAgAGfOnEFeXh4eP36sUbBPCMGaNWsYn63y8vIJvw82Ifi3JZawIM+zhQ29HBxMJGg0rsMHJBIJWltbx3Q/slgsbN++nQmcNN0vPQ8dgsuKZeDyVIGwZU8VJNfO6LW+d955B0uXLgWgCsK///3vY2BgAJcvX8ZHH32EwcFBJnCiM7sWFhazKnACns/E1HZAdXl5Ofz8/IwyIWBkZASXL1/WyyRYG+juR3XjpUy8vMxkt50dgI8JIQGEEGsAv4eqg+40gB8RQuwIISsAxAI4T1FUNYBKAD8khAgA/BxAMUVRL06bnQB93Je1wc7ODocPH8bAwAASExN1Mm7T9ABjZWWF3bt3M1erBQUFGs+3Y7PZiIyMRFhYGJKSknRqax/N4OAg7t27h+Li4jH/N0R7tbe3N+Li4tDe3o6zZ89q/F2OjIxgeHjYoP4vIpEI3d3dE1okWFhYYPfu3Vi6dCnOnTuH/Px8ta9NCGG+44yMDCQkJExLtxA9I+/w4cN48OAB8vPztZ6L2NjYiNTU1EkDUxYh+JdFlrBlm6GoUYAHTwVo6AcOnAdq9BvHp5bS0lIEBgZOmskrLi7GyZMnNXq/hBC4/vwDtHc9lxt5OWcgry/XeX0cDgcffvghE3SUl5fj/fffR2FhIaRSKVJSUpj9MD8/H8nJydNix6EtLBYLO3bs0Gpt/f39EIlEmDNnjlHWJBAIEBMTw8xqNDZ1dXVwcXGZ8e5HEzPDjAVPFEWdgqqTLhNADQA2gH8B8CmARwAaoAqm/o2W8AAcBXAAKmlvA1S1UFMilUrxxRdfGNQZVx1mZmbYs2cP5s6di1OnTmn1uroeYDo6OpCVlcUUpGpKQEAAYmJi8OjRI6Slpb0w2kJTrKysEBcXx0g/EokEFEWhrq4Orq6uWl+hjme8J5Qm2SQzMzMcOHCAyaq0tLTo3W1YXFystkaJEIIFCxbg8OHDqK2t1Xh476pVqxAeHj6t3ToWFhbw9/fHhg0bcO7cOeTm5mqclfHw8MD+/fvVBsWEELy90AJfDzIHj636vFpFqgCq1AiuEt3d3ZDJZGpnJdrY2DD1gprAsbSE+Tv/C9GAqpOTzSaQf/5zUFLdf0fz5s3Df/3XfzG3U1Ke97tUVlYyo1ocHBxw9OjRaZOgdIHuApwKiqJQXFxsdANJV1dXo2S1xkOrA7Oh+9HEzDBrap6MydKlS6k///nPWLVq1bTVltBUV1cjPT0dYWFhCAkJUftYur5hwYIFU9YQTURXVxfs7e1BCMHw8LBWQ2llMhlu376NpqYm7NixQy8XYdpuwNnZGQKBABEREQYd3Nne3o60tDQ4OzsjMjJSo5otuVyOzz//HI6Ojti3b5/Or9vU1IRly5Zp9Hh64GleXh5Wr16NRYsWaXTiaGpqQmtrK5YvX27UE01xcTGsrKzg6emJwcFB3LhxA/39/diyZQszjFkTxGIx8vLy1NaxZD8FvnEFkMgBFqEQNm8IPwoVYJGzYfZHiqKQlZWFpUuXatyUoE09XNuZL+FQfBEcjmq9IqE3LH+g+ww8pVKJ2NhYZGZmAgC2bdvG1PQJBAJ84xvfwOPHj5nuR4lEovcFiDG4c+cOCgsL8a1vfUttBqa5uRldXV1YtGiRUddD73NCoVCtq7y+PHr0CI6Ojmpr6Ew1Ty83M22SOS3Q89RYLBaGh4dx69YtnTMs2uLj44OYmBg8fPhwSj+olpYWWFpa6hQ4AaorVUIIhoaGcPz4cWhTJM/lcrF582asXLkSiYmJePz4sc4yHpfLxYIFC8Bms+Hv72/wiefOzs44cuQIuFwuTp06hebmqZVbDoeDPXv2YP369QC0nzenVCpRVlamVX0DbagaHR2NkpISJCUlaVTXVFZWhuLiYqNKeIODg+jp6WFkFysrK+zZswcrVqxAcnIy7t69q/HrP336FE+ePEF3d/ekjwnzAE7tBWz4FDYFDiJ4rgQnKwbwuN0wkurTp09hb2+vceA0PDyMCxcuoKysTKPHu8R8HS3kecOCZX8dpLd17+BlsVj44x//yBTeZ2RkMBd2EokEycnJCAoKAovFwoMHD/DFF1/onTU1BkuWLMHBgwfVBnZyuVzj8VL6QghBcXGx3lMV1NHf3w+xWGw0+dHEPwevRPA0mqamJjx+/NiobtTjsbOzw6FDhyCRSHDu3LkJT6CGPMDweDzMnz9fqw48mqCgICbY01XGI4Qwfiuurq4oLi7WavSKJnC5XGzatAnr1q3DlStXkJubO2Vtk7OzMyPZ3LlzB7du3dI4iKqtrdVZfqQNK93c3HDq1KkpDQY3btyI6Oho8Hg8KJVKowT6E83gIoQgMDAQR48ehUgkwokTJ9D4zOtIHYGBgXjrrbcYuWyy97Z8DvD/tsowz0ElgTlYyXG8YgAlHfplv2UyGWpra+Hn56fxczgcDpRKpVZNHXN/9h7aOp9f/LBvHoeyU/dSgDlz5uDYsWMAVPv/6Nl3T58+ZS60vLy8EBwcbPCLEEMgFAoxd+5ctRm8mpoaeHh4aJUJ14dDhw5h8+bNRtn2dMmPJmY/r1zwNH/+fHzjG99gfIk0nRenL/S8OR8fH5w+ffqFk1J1dTU8PT0NcoDhcDjYsGEDczIrKipCR4fmQ07t7OwQGxsLLpertds3MHZ+HaCSh548eWIUV20fHx8cOXIEzc3NOHv2rEajT2i5hv6bCqlUiqamJr3qG1gsFlauXIl9+/YhPz8fly9fnjSTQAhhJJDs7GzEqrM6GAAAIABJREFUx8drXdCtjvb2dnC53ElraczNzbF9+3asXbsWaWlpyMjImPL1aQ+iqqoqJCUlTer2vtmLhwDr535FjlZyfF7Wr1cAVVFR8UL341Sw2Wy8/vrrWlmKsM3MIPjmrzAkUr03LodA8refAErdumoB4OjRo8waqqurxzSYpKWlgaIo2NvbY+3atdMWfGgLbd1RXV39wn1DQ0Noa2vT2F/LEBjzc9JXHTDx8vDKBU8AmNR+c3MzPv30U6OmeEdDCMFrr72GrVu3IiUlBQ8ePGDGSLS3t6t13taVkZER3L9/H4WFhVo9j5bxVqxYgfPnz2sV/DQ3NzMHGEIIXn/9dURFRYEQAqlUavCA1dLSEq+//jpTTD6VFEMIwbp167Bu3ToAKu+jtra2SR9Pdz8a4sqflhxtbW1x4sQJVFZWqn28t7c3fHx8DGYoqI386OPjg6997WugKAonTpxATU3NlM+RyWSQy+Vqfyt75gvgb/U8gHKyluOzsgGUdmofQA0ODqK3t1enjjQ6cG5sbERycrJGWSjh4qXonbcOlFK11vbSWnR99ZXWr03DZrPxu9/9jlnLlStXmPuam5vH7LcdHR2MR9dsgsPhoLKyEq2trS/cN1X3o7HIzMzE3bt3DbrN6ZQfTcx+XsngicbOzg4BAQFTziYzNF5eXjh06BBKSkqQlpaGR48eGe0Aw+PxEBsbi8jISACqTJs2UkVwcDCio6Px4MEDjWQ8uVyOqqqqMe7eHA6HOfmnp6cjISHB4LMA6QHD+/fvx71795CamjpltoQ+Yd25cwfJyckTrqmvrw9DQ0NaFVBPBYfDQUREBHbu3Ins7GykpqZOGlC6u7sjPDwcgKrWQl/5U1v50czMDJs3b8bWrVtx584dpKamqq29CQoKQnR0NOOzNVkGaq+PAL5WzwuMna1l+LRkAGVaBlAlJSV6O7SLxWL09vZqXFM059v/joZeM9y/WYeS/BYU/vgYRvSwQlmyZAmOHDkCQDX+ZnRAffPmTea30dPTg+rq6mmzXdEUQgji4uKY3ylNd3c3FAqFXs0nukLblBiS6urqaZUfTcxuXoluO00cxpVKJfLy8rB48eJpGxsgk8lw8eJFlJaW4nvf+57RW5IpisLZs2fB4/Gwd+9erU44dDdeY2MjduzYMWk7eHl5OXg83hiTwtG0traip6dHL98nTdaamZmJ2tpabN26dcrar6GhIfT19TEFoLSsR1EUcnJyEBISMuk8N0OsNTs7GxUVFdi4caNaafD69euorq7Gm2++qVPtlVQqxb179xAeHq5TFk0mk+Grr75CSUkJIiIiEBQUpPY3lJqaCpFIhAMHDkx6YZBYNYRq0fOgpa2fi28vtEaA/dS/zba2NjQ3N2vc/aiO0bMRNUHa3o4727dD9syGwnXbNiz78591DuL6+vr+P3vvHR7Vfab9f84U9V4Q6g0VJHozHdMxYIoxmG4cO2tv9nXKbnbf5M1e+Tlls+3NxvntJhucuFCEQRRRhYREFb2pN1RQb6A+0oxGM3PeP+Q5lkAjzajhNbqvi+tC0jlzvnPa9/k+9/3cD3PnzqW5uRm5XM67774LQGBgIBs2bMDFxQVRFNFqtd/oydvYLspY/Th9+vRvXFuZgaC9vZ179+6xYMECsxe5o9V2327IP/rooxc9hmHHJ5988tFf/dVf9blNdXU1CQkJuLm5jdhKSSaTUV9fT2RkJMnJycPeVkAQBGQyGWPGjMHT09OifeVyOaGhodjZ2REfH49cLmfs2LE9Jov29nYePXrUZ0m+o6OjdH7Ly8u5d+8eAQEBQ5p1k8vlhISE4OrqKjVt9vPzM3kMpVIpVT1lZWVx9epVQkNDqa2txWAwDKtJoVwuJzg4mLFjx5KcnExtbS1+fn696neCgoIICQmR3KYtbT2TmZlJYGDggPUacrmcwMBAAgICSElJobCwEB8fH5OLDYPBgIODQ5/l3FHuSqpV0KjtquxzsDFwpUxHsKM1Hnamv5vBYODBgwdMnTp1SLyxZDKZ1LDW1ta23wlf4eCAQ3AwVV95NKkKC7H19cV5gG7TnZ2dqNVq7t27hyiK1NfXs2PHDjZt2iQFykZTVVEUefLkyTcuKMnKyuL48eNMnDiRyspKlErliDbA7g0Gg2FIhN1paWmMGzfOoqbDv/jFL6o/+uijTwZ98FF8I/FS0HbmiG19fHx4++23JS2IKbphKFFaWoqnpyezZ89m/fr1XLp0iZSUlGHVNERFRUmcfWFhocXO4pGRkRLlePr06R50U3Z2tlRebQ6qq6spLy8fcBub/hASEsLu3btpaGjg0KFDfZbSGyEIAlZWVshkMgoKCkZM3+Dn58fu3buxsrIyqS9SKBRS4JmVlUVcXJzZ92ljYyNqtXpI6EcvLy927NhBQEAAhw4d4v79+73es5GRkVLPr6amJpPWB5vD7Aix/5rCs7fW894ZQ59O5MXFxfj4+Ayp91FHRwe5ubkUFxebtf3Y5csJ3LYNAKWVHKvEvWiunrb4uMYKrg8++EAymS0pKWHv3r29bv/gwQNiYmJoaGiw+FjDCS8vLyIiItBoNDx+/HhYfZbMwZUrV4iJiRn05zx9+hSDwfBC6MdRfHPxUtB2QUFB4oEDB5g/f75Zq5CWlhYOHTrEggULho1e0mq13Lhxo4exYHt7O2fPnkUQBNauXTvspnhnz56lqamJbdu2WUzj6HQ6KfuwevVqrK2tKS4uZtasWRZ9jlarlUry6+rqhkV/JooimZmZXL9+nTlz5jBlypQ+7wNj3zbjSn+kXYTLyspITEwkICCARYsW9ZrZyczM5NGjR2zYsKHfazec9GNDQwNJSUnodDpWrFjRa0ZTq9Xy+eef4+vry9q1a01+1uH8NjIbOjiV7kxrh5wx9nBkE4Q8Ywg+WPqxL1hqRqlrb+fh9o1MDFdia6ekUwfyv/0TMjfTLufPoqamhqqqKqZNm0Z+fj4rVqyQ9He/+93v2LJly3NjfPToERMnThxxIbY5yMzMxNnZ+YW3lcnJyaG+vl7y+BsIBkM/jtJ23268FLTd/v37P5o0aRLFxcUEBwebZd/f2NhIZGTksAUwOTk5+Pj49GgVoVQqGT9+PA0NDVy6dAkfH58+u9cPFuPGjSM8PBwrKyt0Oh1tbW1m6ylkMhnBwcE4OzsTHx9Pfn4+S5YswcrKyqIxGCe/+/fvEx8fT1hY2JD3ihIEAS8vL0JDQ7l165Yk/DQ1VrVaTUFBAWq1mitXrjB+/PgR08FBl3fOhAkTKCsr4+rVq7i5uT3XUsTLy0sqMtBqtTQ1NZk8bxUVFYiiOCyTma2trSTYTkhIoKOjAx8fnx6TlVwux8HBgYiIiD6vbbS7EmtsOJUvp9MAbZ2QUAjLgsG122OYmZlJUFDQsOjQjBRgU1MTjx496rPVC4BMqcQhPBJ51lWUSjlyGbTfv47VgnVgxkJNr9fz8OFDpk6dikKhwMPDQ9LXQNdzsXXrVqysrLh37x4eHh7Y2to+R5l/U9Da2kpGRgZBQUEvnFb09PQkMDBwUOeppKQEKyurAdGPo7TdtxsvRfD05z//+aPf/e53lJeXc+vWLYKCgvqcDBUKBWFhYVLglJmZiYODg8WBgSm0tLRQVlbWq9GaIAgEBgbi5OREfHw8VlZWeHl5DcuLUhAEabK4cuUKly9fZsKECRZpSNzc3LCxsSE3N5f6+noCAwMHdJ48PDxwcnIaVj8Y40Tf2NhIcnIyzs7OuLu7P7ddamoq4eHhhIeH4+fnJ02gQ6WfMAcKhYLQ0FA8PDxISkqivr4eX1/fHoG/cSwXL17k2rVrTJw48blrp9PpSEtLY9q0acNmsmgMTsePH092dja3b9/Gw8Ojh7bKw8NDCpwKCwtxcnJ6LhsgCAIBzgIzvOFcAei+CqBuVuqY5yfgZivQ2NhIdXU148ePH9ZrcePGDR4+fNjrOX0WNt6+1OaV4the0UX7ihraauuxmtB/FraoqAhHR8ceQdqMGTMkN3q1Wo1KpaKgoICMjAw6OzslOqy0tJTk5GTCw8O/ERkoY3Py9PR02travhEl/aIo0tzcPKDFj1arJTMzk2nTpg3o/I4GT99uvBTB0yeffPLRBx98IFWAJSYmMmbMGLOEs62trZw4cQLoqnwZLERRJDU1laioqD6zWu7u7oSGhnLt2jWqqqoIDAwcVodhZ2dnHB0dLXYl12q15Ofn8+abb9Lc3ExycjIeHh6SqNlcKBQKaQJpbm7m1KlTQ65pga6MWUBAAN7e3ly8eJG6ujr8/f2loOTJkyc0NTVJE5LxHqmsrCQuLo7AwMAR7THm4uLSIwvl6ur6XBZqzJgxeHh49Kpnys/Px8PDAw8Pj2Efq5WVFZGRkTg6OnLhwgUaGhrw8fHpEXw8efKE2NhYrK2tTa7m/Zxguk9XAOXlpGXZ+GaSS/VEOCspzukKaIY7E+jv78/48ePNFgg7zpxH5ck4nGy7tF+KumJ0/tHI+6DvNBoNubm5TJkypcfkrFQqCQgI4PTpLv1Ua2urRGdXVVURGRmJg4ODlB0LDg7+RvS9q62tRaPRMHPmTKKjo4dssTkYpKSkkJCQwPTp0y1+f/bGDliC0eDp240Xv1wZQQiCwNSpU1m9ejXx8fE8fPiwX7G0o6MjO3fuZO7cuQCD7jdWU1ODjY2NWQ+km5sb27dvRxAEDh06NKwCUTc3N6ZOnQp0+bOY25olLy+PcePGYWVlxfz581m9ejWJiYlcu3ZtwEJwlUqFSqUa0L7mwtfXl127diGTydi/fz8VFRUYDAZycnJM6txsbGxGlL4zwtramuXLl7Ny5UouXrxIQkJCD6G+o6OjtMqvqqqSKrba2tqoq6sbkqDfEowbN449e/Ygl8vZt28feXl50nNmbMw8bdq0Pj9jjh/812odq6NbUMrBz1XLbx+2olJ6DJttRHcolUopcC4qKur3XhZkMtz/93/Q3Nj1zMhkArp9v0TUmi5WycnJMdn7ceXKlSxbtgzoEscbvZ1EUZScx4OCgti9e/eAJ/ehhF6vJy8vj6ioKPz9/V84ZWdEREQES5cutXi/lpYWmpubpU4UoxjFs3ipgicjAgIC2Lp1K5mZmf0264WuLJBcLqezs5PDhw9za4COwnq9nvz8/B4Gkv1BqVSyatUqpk2bxpEjR/p1pB4KGKvg+qvkamlpoaWlpccLJiAggF27dlFfX8/hw4cHZOjn6+vLO++8I9k2FBcXD0sFopWVFStWrGDJkiWcPXuWY8eO4e7u3uuL39fXl7feegtbW1tJUD7SxRaBgYHs3r0bhUJhsiIvNzeX9PR0tFqtxdWPQwlra2uWLl3KunXruH37NnFxcVLrnODgYORyOR0dHVKg1xuWBskJsP86WA1w7+RMqw8lTSN33mtqajh58iTp6en9bmvr60fH4rfR6bruVVu5DtXej3rdtrGxkY6ODpMFEoIg8Ktf/aqHuawRxkbM0JVJ1ev1lJaWWvK1hhzFxcX4+vpK462pqSExMXHYKmnNhZeXl8VSBGP142DNV0fx7cZLGTxBFx2ybds2NBoNR48e7dFTyhRkMhljx47tV0RqCkVFRT1eMOZCEAQmTZrExo0buXbtGpcvXx7Wl9KECRN45513cHR0RBRFqqqeb37a1wvGzs6ODRs2EBkZyZdffkleXp7FYzBO+FVVVcTFxZGRkTGwL2MGQkND2bp1K4WFhdy7d89kLz/j98zPz+fcuXOUlJQM25hMwdrammXLlvHaa69x+fJlzp8/3yMLtWTJErZt20ZLSwuiKL7wrISPjw+7du3C19eXmJiYHrYGBQUFXL9+vde2HvCVc3WkPd7WXz8v/u5a/u2+asQCqLFjx7J+/XqmTJli1vZjNrxFtegj/exYn4/menyPbcydnAMCAvj+978PdAnYu7cdSkpKkjLDt2/f5vjx42b1dRwOaDQaKisrexjjqlQqioqKaGzsw29ihGAsADF3sWMJOzCKlxcvhVVBXw7joihy+/ZtMjIyWLdunUU+OIWFhbi5uZllbKlWq7lz5w4LFy4cVCZArVaTkJCARqNh7dq1w1qNB1203Llz59i0aVOP3nvV1dXU1NRIVJ8p1NbWcu7cOXx9fVm8ePGAdBAFBQWEhIQgl8stNoY0FxkZGbi6utLc3My1a9eYOXMmM2bM6PVYoihSUlJCUFCQ5ET+IlaoWq2W69ev8+jRox7u5EazR51OR21tLW+++eY3Qn/S0NBAcnIyHR0drFixgjFjxlBfX9+vHkulUrE3rZ7ObhnB0nprfjrTgQDnkTvvxpYf/T1zOpWKhp9sZYxHV7ajUwfyH3+CzKXLxqG8vJympiYmTpzY7zE7OjpYvnw5RUVFKJVKdu/eLdF8s2fPZuXKlajV6hFvvtsdDx8+xMfHp0cWzWAwYDAYzKpsHm6kp6eTnJzMnj17ei0Q6Q69Xk9KSgqzZ88eNEU/alXw7cZLIxg35TAuCAL+/v64uLhw7tw57OzszDJD0+l0nDhxgqdPn5pVVZKRkUFISMiggx2lUklkZCTt7e1cuHBhQOJsS+Di4iKVmRsDBYPB0KO8ui84ODgQFRVFcXExt27dwsfHxyKXXuiiTY0l+UeOHEGpVFrskN4Xulc/enl5ERYWxv3798nOzsbPz++5l6ggCLi6uiIIAi0tLRw+fBhPT88R0eJ0R3d38osXL1JVVYWfnx8VFRXSytlgMBAaGvqNoB9sbW2JiopCLpeTmJhIW1sb48aNQy6XU1FRQW5ubq8ak9TUVF4N86Kqw5q2rzKuLnZ6zhYYmORhhbPN8H83URQ5duwYeXl5TJw4sc/zKbOyQucZAulXUCjl1FU0U1vVjvucuXR2dpKWlsbUqVPNEjAbK3+PHTuGwWCgpaVFyvBUVlYyfvx4XFxcpCzJSAfyDQ0N1NbWEhkZ2eO4xm4GMLJVqr3B2dmZ0NBQPD09+x2HsRJ0oOxCd4wKxr/deOmDJyPc3NwIDg7m4sWLNDc399syRCaTST5JSqUSnU6HIAi9PpwNDQ08efJECkAGC0EQ8PPzY8yYMZw/fx6tVouvr++wvKCMVKUgCKjVak6cOEFLSws+Pj5mO+4aJwAbGxvOnz+PIAh4e3tbPF6tVktRURGBgYFDFjCKosjDhw+Jjo6WKpZsbGyIiopCo9GQmJiIjY0NY8aM6XW8bW1tFBcXExkZOeT+VObCycmJiRMnUldXR2JiIrW1tSxatAhPT0+Cg4MRBAGVSkVzc/MLF/IKgsCYMWOIjo6msLCQa9eu4ezsTFFREYWFhURHR/cIKurq6mhpaSEsLIwpHlY8ajDQ3i2AOlMgMtlDOewBlCAI2NnZERwcbNa9Z+PrR03OY4ov3yEvtYan9x/iOXcuZa2tjBkzpt8MSHcEBATw+PFj8vLyaGxsJCQkRLpXnzx5IrVDysrK4sqVK/32HBwqiKLIgwcPmDRpUq/+cAaDQdI9jnTRQncolUqcnJz6PSdqtZq8vLznqh8HitHg6duNl1bz1Bs8PDzYvn07jY2NHDt2rN8u687OztjZ2SGKIufPn+fcuXPP8eqiKA5J5/feYBRnl5eXc/z4cbO7wg8UarWapqYmampqTDb+7Qvjx49n27ZtPHr0iBMnTpilM+sOW1tb3nzzTYk+LCoqGnRVXnV1Nba2ts9NiDKZjJkzZ7JlyxbS0tI4efJkr8cyVkQaJ8OioqJhba9jCkqlkkWLFjF+/HipWrL7eC9cuMCJEyf6LY4YKdjZ2fHaa6+xatUqrl27RktLC2vXru1BLxoMBnJzc6XqR0EQeCfKAQ+rryfqIA8NP0jqpLJl+MccEhIiGY2aU3Xr+79+Qrt7eNcPBgMPfvQjnpSVDSiQ+PnPfy5lrZOTk6Xfl5aWSloohUKBUqk0q0p2KFBeXo6Li4vJbLpMJsPHx2fYG56bg+bmZm7dutXnucnJySEyMnJYLWFG8e3BaObpGSiVSiIiImhqauLy5cv4+fmZRTO1tLT02gS1vLwcmUxmsX+SubCysiIqKoqmpiaSk5MZO3bssNFHxkqziRMn4ujoSHFxMQ4ODha9bIxZHaNRZW/u2X3BGIBqtVqOHj1KY2Mj4eHhFn8X6NI3pKamMm3aNJP0o52dHRMmTJDakDg5OT2n0TGOqba2ltjYWGxsbPDx8ent44YVzc3N1NfX88Ybb0jjtbW1xdPTEz8/PwICAr5xIlhnZ2cmTpxIU1MTly5dQiaTkZeXh1arpbW1FVtb2x46REEQmOJpRV6DAbVez50SOx5W2HChGFaNA0fzDPIHhfz8fE6cOEF4eHifjvyCTIbHnDmUHz+OoaMDXWsrjgYD/qtXm+U+3h329vbY29tz6dIlNBqN5DK+cOFCpkyZglwux93dnaioqCFplNwfOjs7SU9P79d8NSgo6BvRE+7JkyckJiaabIw91OwAjGaevu146QXjfSE/P5+LFy/y6quvSg2DzUFNTY3UpuL69evMnz9/RF5oxcXFXLhwgenTp5sUOw8GDQ0NPHr0iNmzZ9PS0sKnn37K9OnTWbhw4YA+r7y8nPPnzxMWFtajx5+5ePr0KXZ2dtjZ2aHT6ZDL5RZ95/z8fORyudm966qqqkhISMDLy4slS5b0akxYWFgoleKPpP5EFEVu3rxJdHS0lEUz0ni2trYsX75cmjTy8/NRq9VmV5CNFOrr60lMTOTu3bssWLAANzc3k/eFQRSJK9DykwvWaL8qPA1whn3rn++FN9RoaGjgxo0bLF261Cyqtio+ngcffohMLjB+mjeeCxfi+L1/svi4er2eNWvWkJmZiVKpZPHixXz++efPbadWq4fd3ys7Oxt7e/seRSSmYDAYqK6uHlCLk6GCXq+no6Oj1+sliiLXr19nypQpQ1qAMyoY/3ZjlLbrAxEREWzevJmbN29y5coVs+mYGzdukJycTG5uLiEhISMSOEEXrbB9+3YKCgo4efJkjxL2waI7/QhdOps33niDV155BWBAVJW/vz+7du2SGjHX19dbtL+x5Ycoipw7d47ExESzy5HVajXV1dUW0Y/GsntbW1v279/P48ePn9vGKIDu7OwkNja2Vy+m4UB1dTX29vY96McxY8awfft2/P39iYmJITU1FYPBQH5+Pnl5eS+EXuwL7u7ubNu2jT179pCamkp9fb1JSw6ZILAp3Jo/rQHlV2+xsmbYctzAldLh/V5ubm68/vrrZmvcxq5ahd2COSx4bRwhkR441uXQkXbd4uPK5XL++Z//GUEQ6Ozs5MKFC1y9evW57S5evMjZs2cHbehrCiqVSmrFZA4ePHjA4cOHX6htgVwuN3m9ysvLcXV1HfbK5VF8uzBK2/UDe3t7xo8fT0ZGBtnZ2QQHB/cbDI0bN46xY8dSU1PDxIkTpazISMDa2pqoqCjq6uq4cuUK3t7eQ/JSKCsre45+dHFxQaFQYDAYOH78OCqVymJHXiNNKpPJOH/+PEqlckC9/JqamnqlTU0hIyOD0NBQi8+NscLN09OTCxcuUF9fj5+f33PZEY1GQ05ODn5+fsOu+TA2l+2NfpTJZPj5+REaGsqdO3fIyclh0aJFTJ48GaVSiV6vN1no8CIgCAK2trZYW1uj0Wj44x//iJeXl8nrGuIK0Z6QUAQiIsvHt1Cs0lDeYMVUL5ml7JhFUKvVxMfH4+Dg0CdVXlxcjO3kqVg/vIq9Xdd7QJd9F8X89SC3LNvq7e1NdXW1ZJKZk5PDjh07egicjYJ8S6tazUVqaqpFBRIuLi74+voOqEhkKNHY2MilS5dwd3eXxm4u/TgQjNJ232680MyTIAghgiBcEQShTRCEHEEQXv3q9/8gCMLTr/79otv2boIgxH+1fbYgCPNHYpy2trZs3LgRb29vDh48aNLUzwgrKyuqqqqIjo7m7t27fPnll2g0pts0DDXkcjmvvvoqixcv5tSpU326OJuDzs5OiouLiYiI6PXvBoMBJyenAVdyGU1At27dSlZWFqdOnbJI/C4IArNnz2bGjK4MeVVVVa9ZISPq6+vp7OwcVDlyYGAgb7/9NgD79++nrKysx9/t7e3Zvn07oaGh0piGS6xdWFhIQEBAn/obNzc3tm7dSkREBLGxsaSmpqLT6Th37pzU7uObAKOB5PTp01m+fDnh4eFcv36dkydPmjSBXBoCxzbDqigV3s463O31lHY08fNrnXQOo8G1XC6nsbGxTxf9jo4OysrKCJ88GdnGD+n8imO0ketRffbrAR33xz/+sUQZ5+bmcuzYMRoaGjh27Bg1NTW4uLhIBQxDfV3r6uokfZW5sLe3Z9y4cS+8ebFCoaCsrKzH9Xr06NGIsgOj+PbghWqeBEG4D5wGfgu8CfwG2A3sA5YD7cBF4MeiKJ4UBOEAoAd+CLwG/A4IEEWxzz4iA9U89YaCggKSkpJYuHAhEyZM6HWb2tpaysvLmTFjBsXFxRQXF7N06dIXsupqbm6W/KtWrlw5oAai2dnZODg4mJ2mLy0txdra2mTrib6g1+u5ceMGOTk5rFy5ckDGfydOnKCpqYm33377udWkKIqkpKQwbdq0IVuZFxUVkZycLGm3nn0Rt7W18Ze//IVJkyaxePHiITmmEQMxX21paSE5OZnW1la8vLzw9PRk+vTpQzqugaKqqoq6ujpJj2U0W7x//z4PHjxg1qxZJrMEN6o0XK1RIfvqMdMZoOKJA/+8yAbXYeqbq9fr+8xYpKen4+HhIWXOyn721wQoa4Cue1G35f+gjJ5l8XF/+9vf8h//8R8A0jkxGAyMGzeOHTt2AHD16lU0Gg0rV660+PN7g8Fg4Nq1a7zyyisWv0d0Oh0ZGRl4enoOW/GMOeiuQ1SpVDx8+JAFCxYMy7t5VPP07cYLWwoIgjAN8AB+JYpimyiK+4AtwE7gT6Io5oqiWAr8N7BVEARrugKsfxRFsUkUxS+BJ8CSkRx3WFgYb731Fvfu3SP5pbyfAAAgAElEQVQpKek5Tcaz5dUhISEsW7ZM8tpJTU0d0VW+s7Mzb731Fi4uLhw8eLDXVit9wahvMJZo9wdRFLl69SoXL14c0PeUy+UsXLiQ1atXk5SUxOXLly3O2Kxdu5aNGzdKou3uPfrKyspwd3cfUkojNDSU3bt3o9FoOHDgAJWVlT3+bm9vz+rVqyV92FAiOzub8ePHW7Sqd3JyYuPGjcycOZPHjx+jUqno7Oykvr5+SHVylqK33o8ymQyFQoGTkxPW1tYUFxf3eo4B5vnY8FaoM3pD10SokEGQl4ofX2njUf3wPHPGwKmiooL8/Pwef2tubkalUvWovPT+3//K06dd5fKCIKA/8n+h03JrgQ8++ECqYisoKJD0a4WFhVLWVS6Xo1Aohux98/jxY7y9vQe0AJPJZNy9e5fCwsIhGctA0T1IGi4LmVG8HHiRedTpQAGw7yt67h6gBSYBqd22ywbGAeGAShTFil7+NqJwd3dn+/bttLe3Exsb28NPp68XTEZGhuRpM5IYDI2XlZXFhAkTzH7BCILA5s2bWbt2bdfkoNcPiK4yeli1trYSExPD06dPAcjMzHxuknoWVlZWUkn+rVu3iImJQa1WS/TjQK0N+oKtrS2rV69mwYIFnDlzhmvXrvX43mFhYZK4/cKFCxQWFvLo0SMyMzMHPLkZBdUDoR8FQSAqKoq3334blUrF559/zqeffkp8fPxz22ZnZw+oP6Gl6It+NF7T9evXM2fOHM6ePcuFCxeeC/ZCnZV8L9oFQ+fX5zRsrJr/SG3l4uPhCaCMlY53796VrqWRfnz22VE6OVEUufxr+k7WSdu+f7X4mPb29vzd3/0d0KXl6V6YkJycjMFgQBRFJk2aNCTBgZF+NLcy9VmUlJQwZcoUXn311UGPZSDIy8sjJyeHlpYWDh48yO3bty2mH02hurqa+/fv99tIfRTfLrww2k4QhJ8C/wTsAk4A7wA/o4uqe0cUxetfbTcP+Pyrv+8TRXFct8/4M/BYFMXf9PL5fwX8PeDi4uLiERcXN+TfQRRF8vLyKC4uZtasWXh4eNDW1mYyqyGKIq2trZK4tLOzc8S59ra2Nu7evYu1tTXTp0/vUyej0+nQarWDcs5OS0ujoaGBhQsXDqjPlSiKlJaWkpmZSWRkJBcvXgS6Vt7m4MmTJ1RXVzNx4kQ6OjqQyWTD3udNo9GQmppKa2srM2bM6CEYN/aj8/X1JS0tjZKSEv7hH/5hQBOcSqXCzs5uSLQk1dXVXL9+nbFjx/LKK6/0OEefffYZGo2G733ve4M+jimIotjns2PcRhAEDAYDOp2OnJwcKioqmDBhAoGBgT3O4b6YL/F4/Xu4+38tNK9tUWBTX88K1/IhF5JrNBoUCoV0j3d2dqLT6Z5bRBkMBn75y1/yK29rFkTYSd/rYfR6WsdYRlHr9Xp+8pOfUFVVhb29Pdu2bZPuheDgYD755BPWrFlDdHQ0oij26m9kLtRqtWTCORCcOnWKrKwsfvazn72QXpD79u2jra2NDz74gBs3buDj40NISMiQjOP69eskJCTwj//4jz1aOS1evHiUtvsW40V3bcwURTHmq///URCEj+ii8rq/ceyARqDpmd93/9tzEEXxE+AT6NI8DdeKZ/HixTx+/JiEhAT0ej0LFiwwq+KrsLCQxMRENm/ePOImcqtWreL69evk5+ezZs2aXsdr1DfMmzdvQGl6I/z8/KitrWXevHmDGTKNjY3ExcVRWlrKu+++a/EKtrW1lZs3b+Lk5MSsWbOGXby6cuVK8vLyuHLlCp6ensyZM0eaWJcsWYJMJmPHjh3MmTOHhQsXWlzpU1paikqlkujhocAbb7zB9evXKSgoYOzYscyfPx9XV1d+//vfs3z58mHNGty/f5/o6Oh+nwW9Xs/p06dxd3fnhz/8ITU1NVy8eJHa2lqWLl2Kp6cnoijyhz/8gUnFV/COfJeStq5iDS8nHSprVxLagvjlIjnWw/D2MxgMlJWV8fjxYxYuXPjc4qSyshJbW1va33yPp5f+goenDYIgEJ2dgM3GA6C0LLD/zW9+w549e2hrayMrK4tJkyYBXcJuJycnVqxYwf379/Hw8Bjw9WtubiYrK4u5c+cOONhISkpi/PjxuLm5kZWVxbZt20Y0gDp8+DBBQUEsWbKEwMBAdDqdyQIYS1FQUICbmxsrV64cpQBfIrxI2q4EkJYxQtddJwMSgO5K7Cgg/avtXQVBcOvlby8UwcHBrF69mvz8fB4+fGhW+tbd3Z3Q0NAX0rpALpezaNEili1bxpkzZ7h9+/Zznj/FxcUD1jd0x7hx46TAqaGhoQe1YQlcXV0JCQlBoVBQWVnJo0ePLNo/OzsbhULB3bt3R4Q2FQSB8ePHs3v3bhoaGjh48CA1NV1CYYVCQUVFBY2NjTQ2NpKYmGjRZw8X/Whtbc3SpUtZvnw5Z8+e5U9/+hNZWVloNJoeOqShhpF+NGcRIZPJcHJykrIoY8eOZdu2bYwfP56jR49y5coVysvLUalUhIeNY2u4A3PG2GO85dq1Mo7lyNgRB/XD0M3ozp07fPLJJzg7O/ea1TXqkYInTkS//B10X5UDquubqL92yeLjLVu2jDlz5gBIPl7QlWH29PQkICCAtWvX8tprrw3o+5iiHy1FdXU13t7eWFtb4+DgMKLVx9BlqOvp6SnRjyEhIUPmg9XQ0ICbm9to4PSS4UUGTwmAjyAI7wmCYAv8L6AU2A98TxAEX0EQIuiqrDskimIbXZV5HwmCYCsIwruAI3DnBY1fgiiKVFRU8Dd/8zdYWVlx6NAhGhoa+tzH1dWVVatWoVAo0Ol0pKSkjFhPKiNCQkLYsWMHpaWlHD9+XOo119HRQUVFxYD1DaaQm5vL/fv3Le5pZ0Rqaipjx47lO9/5DikpKSQkJJh1zmpra1EqlSxZsoTdu3dLRpLD3QsQurQp69atY9asWcTFxXH9+nX0ej0ZGRkIgsDGjRsliwVzkZ+fT2ho6LBRvqGhofzbv/0bixYt4s9//jPt7e3DFjw9a77aHwRBYOnSpUyePBnoopZlMhmTJ09mz549aDQa/vCHP6DRaKT7d5GvLW8EO6HTyTmf7YjOIHCvCtYdhrynQ/t9IiMjiY6Olsb3LEpKShAEgYCAALw2bKFc7UbOg2puJBSS9qt/QWfhsyEIAj//+c+Bruf23r170t/8/f1xcnLC19d3wNR7VVUVDg4Og6L89Ho9dXV1eHt7ExERwbp16wa9KLMEarVaCiZzc3MJDAxk7969pKcPzbrbGDyN4uXCCwueRFFsBFYA7wO1wAbgTVEUz9MVQGUCKcB/iqJ45avdfgBEAg10BVtviqI4jE4u5qGyshIHBwc8PDxYsWIF06ZN48iRI2ZXllRWVnL//v1eK4iGG46OjmzevBkfHx8OHDhASUkJubm5hIeHD7lp3Ny5c9m5c6eka7E0A5SVlcXUqVMlZ3KZTGay8soIY/WjsdO8MXAqLCzkL3/5S7+eXUMBozh7165dPH36lIMHD3L37l08PDxYsGCBlHF58OABpaWlfX5Wa2srjY2Nw17ubWyN4ubmhkqlIjY2VhLtDyVKS0sHXP1YW1vLZ599Jl1/Ozs7Vq1axdixY2lvbyctLU1ytY5wteKn0134wSw5xvxARSu8EStyoWjodJ9FRUUsX74cuVzea2BfUlKCt7e3pI3x/um/U1apRRShvbycvN/+1uJjTpo0iY0bNwJdz4gx861UKrlzp2ttqVariYuL67fYojt0Oh2PHj0iMjLS4jF1x5MnT9Dr9T2sS9rb20dMYF1XVwd0FXW0tbURHBzM5MmTB2Sl0hsaGxtHg6eXEKO97QYJY9Zo3rx5PUS21dXVnDlzhsjISObPn9+vxqa5uVla3bW1tQ3YcHIwKC0tJS4uDltbW957771hdUXPzs4mKSmJbdu2mV0tptVqUalUPV5UhYWFJCcnM2HCBObMmfPcmAsLCzEYDM9RXCqVitu3b/Pqq68OSMg+UIiiSE5ODpcvX8bf35+1a9cil8vR6XTExMQwZswYkxSLKIrcvn2biIiIEXtZNzY2cuDAAUJCQigrK2Py5Mm88sorQ3LOOjs7B9X7sb29nYSEBBYvXtyj4bFaraa8vJzm5mbu3bvHlClTmDVrljTmi8XwYQK0dcIEHzWhHh2EOzjxNzMG50j+9OlTioqKeOWVV6itreX48eO89tprPbzKGhsbaWhokMxTAcrj4kj78Y+7fhAE5h0+jJuF2cjy8nIWLlyIVqslIiKCRYsWAV1Vij/60Y+wsrIiNjaWCRMmmPSnexZ5eXlYWVlZ1MKoN2i1WkpKSvD09MTV1ZWmpiY+//xzFi9ePCL9FTUaDUVFRZSXlzNnzpxBZdF6Q1FREXK5/Lk+f6M+T99ujAZPg0RfL5j29nbOnTuHKIqsWbPGrIDIqI9ZvHgxEydOHI4hm4Qoily8eJHq6moUCgWrV68e8heNEW1tbTx8+JB58+YNWrzd1tZGYmIi7e3tvPbaa1L5sUaj4fbt2yxYsKDPQNA4ic+ePXvE6ASVSkVSUhItLS2sXLmSsWPH0tHRIXnzaDQarK2te+goampqqKysHHFDS2N1VGtrK+fPn6etrY1ly5YNOvuVlZWFk5OT2R5i/UGn0z0X1LW0tHDlyhWePHnC4sWLpec0/yn85IqWmSEtyARoVsvQtjvxT68qsBlAXGg0X50+fTr29vbodDqSkpJ45ZVX+g10RVHk7rvvUvdVn7qIeeGE/vK3yIPMb0YO8Otf/5r//u//RhAEtm7dSlhYGCtWrJCCN0uq3Nrb27l37x4LFiwY8uIKURS5f//+iGo+KyoqqK+v70GnqlQqqTvCcGA0ePp2Y7Qx8CDQ3t5ObW2tyc7idnZ2bNq0CR8fHw4ePGgWLefk5MTkyZMH5Kw9WFRWVjJmzBh27tzJuHHjOHTokMXCbHNhb28vvZg1Gg3nzp3rk8b74x//yJkzZ0x+1saNG5k4cSJHjhzh4cOHiKJoNv1YU1NDenr6iFB40FXafOjQIVavXs3MmTOJi4sjJSVFCpz0ej1xcXE9PJcMBgN5eXlERVk2oQ4GmZmZfP7555I2TKVSUVNTg7+/P/Hx8b36LJmLoaYfb968ydGjR6mrq+OLL76QzGCdnJxYt24dS5Ys4fLly5w6dYrm5mYiPOCHc/SSG7mzrQEn52b+6ryWugFI8kpLS/Hw8JAWSAqFgtdee61HcFBdXc2RI0eea4AtCAKTfv1rHMe4MG9lKOEhNnTu+zXoLBM0f/jhh7i4uCCKotTiqPt7RBAERFGkvLzcZMNlIwZivmoKDx8+5NKlr8XwgiAwc+bMEQuc0tPTOXbsWA/dnl6v59NPP2Wwi+rW1lYSExMlanAULw9Gg6dBIDs7m6ioqD5fMDKZjPnz57Ns2TJOnz7NgwcP+qw2UygULFq0CAcHB0RR5Nq1a9TW1g7H8HtAp9NRUFBAZGSk9HLbsGED165dIykpadg6tENXtVVJSYnJ4EmtVnP+/HmpWq03CILA5MmT2bp1K3l5eezfv5/6+nq8vb37Pb6/vz/vvfeelJVobW0d2BcxE9euXSMlJQUrKyuioqJ6VORVV1cjk8kIDw8nLCxM2qeoqAgfH58RFdrevn2buLg4iY728PAgKiqKuXPnsmfPHuRyOfv27SM3N9eiCkpjBddQuju7u7vj4eFBfn4+R44cee4aBgcH8/bbb+Pl5UVMTAy3b99mnpeS1wMdpUo8a4VIpG8L309Wk2XBXKjVann8+HGP62VEZ2cnly5dorCwkJycHL744oteCyZsfXwI++ADnN26rq+NqEZz7A/mD4KubgI//OEPga6F3d69e5/ru1dZWUlsbCy5ubkmP+fp06eIojhkFiqJiYnExMQ89/u6uroRcRyPjY3lypUrPWQVcrmcVatWmRT2m4vKyko+/vhjSkpKBjnKUfxPw2jwNEA8efIEURTx9PQ0a/vQ0FC2bdtGTk4OZ8+eNatKTK1Wk5eX12eT26FCQUEBQUFBPV4w3t7e7Ny5E61W28Ple6jh6+vLe++9h5+fH9CVCeo+GWdmZqLT6cyiq9zc3Hjrrbfo6Ogwy43cCKNgub6+ns8++0zqWj/UEEWRzMzMHs7Pxoq82bNnc+rUKVJSUpg0aZKk0zK6I3fXyYwEcnNzGTdunKRHUiqVLFu2DDs7O6ysrAgNDWXdunXcvXtX6idoDmpra7G2th7SzENERATLly+nqKgIQRB6zQYrFApmz57Njh07qK2tZf/+/di3VLMnwgXZV69CmQBTAtr4t3sqzj0yLyDMz8/vcZ66QyaTUVlZSV1dHSUlJSgUCpM+cD6736Os8WuTRaucKxjKLcv8vv3229J5bW1t5T//8z97/N3X15fVq1ebFIEbNXlD6R9WXV3dqzj71q1bXLp0aVjbVbW3t1NVVdUrNRwRETFoh3FjQUJ3zd0oXg6MBk8DgMFgGNALxsXFhW3btmFtbc2hQ4f6DUbs7OzYtWuX1BOtqalpQK1O+kN7ezt1dXW9Nv61sbFh9erVzJgxg9jYWNLT04flZWf0xKmvr+fLL7/sUXJ9//59rKyszBa6VlVVMXv2bLZt28atW7c4e/as2fSSi4sLU6dOHbRI1hSMXd2fXfEKgkBkZCS7d++mublZonkNBgOxsbFSN/uRglarpaCgwCRNmJ2dzfHjxzEYDOzcuRN/f38OHTrEvXv3nvMM6w69Xk9eXt6wWR/k5eUhiiIFBQUmt3F2dmb9+vUsWrSI5ORk7l2MZ2egHHv512KnKB8NZ8ta+PiOgb5u95aWFpqamqTA/1nI5XK2bdvG3LlzKSkpISAgwOR1FASBMf/wrzQ87bpXZTIB7Re/Ar35z7yVlVUP64vPPvuM0tJS0tPTOXXqlOQ/Zkrw/yz9OBQwFTwtWrSIXbt2Das/UnZ2tslFrpHCtLTfZ3eMBk8vL0aDpwGgtLSUMWPGDOgFo1AoWLFiBTNnziQ2NpacnJw+t7e1tUUQBHQ6HceOHeu199hgkZWV1Sf9KAgCEyZMYOvWraSnp3PmzJlhayDr5ubG0qVLJadk6NJMTJo0yay2KjqdjsLCQiIjIxk7diw7d+7Ezs6O/fv3m5VaNzYmtre3l2jTocy4GTNapgJBOzs7Xn/9debNm8eZM2c4d+4cM2fOZOvWrQAj1lS6sLAQnU5nMkMRFRUludPL5XJmzZrFtm3bKC0t7bMBdXFx8bDRj6IoSgGKOdYHoaGhvP3223h4eHDycAzT1Y/wsf06exTo3kllZzM/uKBH3QtrbfSo6s9A0hioFBQUSFYZpmAfGIRq0hr0uq4A1EavQnNyb7/fpTusra2lYEGn0/HnP/+ZkydPkpaWRnFxMdAloD5x4kSPxZiRfhxK89W2tjZUKlWvwZOLi8uw0tBGdkClUpnMMCUkJPRYqFkKo59ff9d1FN8+jAZPFsJYdtubvsESTJgwgTfffJPbt2+TlJTUb0ZJoVCwcOFCiw0V+8OTJ08AzKIf3dzc2L59Ow4ODhw4cICKiop+97EUgiAwadIkbGxsMBgMHD9+HHt7e7PbuzxLPxrNMVetWsWFCxdITk42219GpVKRk5MjTThDAa1WS2hoaL9arPDwcHbv3k1BQQFFRUU0NjYiiiKJiYmkpKQMexDV0NCAvb29ycyTTCaT9HEtLS08ePAAFxcXNm3axMyZMzl9+jTJyck9nKQ1Gg2VlZXDRj+2trai0+lYs2aNRNv1lQWDrvtj7ty5bN++nbrqKjpvHidU8fX94WKr53aVgc3HoEbVc9+amhqsra3Nyjq0t7dTVlZGa2trv9fO/93vUfr068yQVXoShsqiPvb4GgaDgZqaGl5//XXp52ebBouiiF6vp6mpqYfO0Eg/DqV1h1GnaMpTqa6uThK3DyWM7EBISIhkkPksBEFgw4YNrFq1asDHaWxsxMnJaUTtTkbxzcBLY1UQHx+PtbX1oEvvMzIycHV1HbIqoY6ODhITE2lubub11183ewVz//59RFFkxowZA057GwwGUlJSmDFjhsVZtKKiIpKSkpg4cSJz5swZll5xGo2GkydPEhUV1SMTZQptbW3cv3+fhQsX9npONBoNly9fpqqqilWrVpnVg7C9vV3K/rW2tuLg4DBibRjKy8tpamrC3t6e5ORkAgICMBgMuLi4MHfu3GE/vsFgMOu6Xr9+ndTUVPbs2YOjoyPQpddLSUmhuLiYxYsXEx4eTmpqKt7e3maJ+AcKURTR6XQolUry8/O5c+cOW7Zs6dGwtS8UFRVx5coV5AFRVLuHcTHfgYK6rn3H2MOf18KUsV30Y0pKCrNnzzb7s43Uqzk6m9aCAnS//1+4enQ5g2sUTtj89FOQ9z9Ja7VatFotP/jBD0hISHiuafCmTZukZsHG37W0tJCRkcG8efOG9P4WRZHGxkZsbW17zTI9efKE48eP8/rrr5v1PJqLx48fo9FoiIiIoL6+Hisrq2GxXWlpaaG5ubnX+WDUquDbDflHH330oscw7Pjkk08+cnV1pbCwkIkTJw745dDS0kJZWdmg+zx1h0KhIDw8HL1eT0JCAq6urmb5wqSmptLa2kpERMSAx1JSUoK1tTU+Pj4W7+vm5kZkZCRpaWlkZGQQEBBg9iRiLhQKBQEBAfj7+yMIAlVVVQiCYJK+S0tLIyIiwmQgqFAoCAsLw9HRkYSEBNrb2/H19e0zQFAqlQiCQEdHBwcPHqSpqWnAeiiDwWD2tdLpdKSlpTF16lQ8PDyYMGEC5eXllJSUMHHiRNzc3CTKYLjatJg7Vn9/f8LDw3tMTkqlUsqwXb16laysLORyeQ+h/HBAEARJU6TRaKitrSUsLMzsc+Tm5sakSZPQ1ldTn34DT2d3MlpcEBFo64S4PAhxBVljEU5OTmYbvEJXYYCdnR2iKNLW1tYnDW3t7k5tcS2OraXIZAIKQwcdLSoUkf0XTcjlcqysrIiOjubAgQN0dHSgUCik7E9NTQ0zZ85ELpdjMBhoaGiQbDCGmkYTBAFbW1uT59/e3p5p06YNaWCj1WrJzMxk6tSpyOVy7O3t+3w3ZWZm0tDQYHbxT3f0tSD/xS9+Uf3RRx99YvGHjuJ/BF4a2m7dunVS12tRFC2mPYaqQWZvEASBadOmsX79ei5dusTVq1f79GERBIE1a9awevVqBEGgvb39Oe+Y/jAU9KODgwObNm2SPKEsaf1gDnQ6HXv27CEmJga9Xk98fLxJzdeTJ08QBAEPD49+PzcsLIxdu3bR0NBATEyMWR4tRiGuuaL13hAfH8+7775rVluaR48e9aAfra2tWb58OatXr+bq1aucPXuWo0ePcvr06SGn8Gpra3n//ffN7v0lCIJEXWVnZ3Pt2jVpTL6+vuzcuRO1Wk1mZib37t3r12NooNi7dy8HDhyQfvbz82PTpk3Y2tpadI6MVXm7Nr/BFCGTbdbxOCq7xiyTGfhTejsnczosCqJPnz7NZ599BkBKSgoHDx7sVzcY8MEPKe3mztF5Mx69um966+bNm+zduxedTse4cePYuXMn0LWwMNqNNDY2kpqaCnRpfj777DOUSuWwiJ4vXrzIiRMn+txGJpNJGcOhQF5enkQ/5ubmcvjw4T7PdXZ2Nnl5eQM6VlxcHGlpaQMd6ij+B+OlCJ46Oztxc3OTVhZ37twhKSmpXz1Ed9TU1GBrazuswkAfHx+p/9nRo0f79BsSBEHi2S9dusSRI0cs6hWVl5dHWFjYoLl6oyfUxo0buXHjBomJiUPWsyo3Nxe1Wk1wcDByuZwNGzawbNkyoCuLY5wQB1L9aG9vz/r165k+fTrHjh3jzp07fd4PxgDXuHp/+PChxd4umZmZaLVaidoyhba2Np48edJruX1AQAC7du3CwcGBpqamIaU6jMjLy6OsrGxABRG1tbXU1tb2CJCqq6uZM2cO77zzDuXl5f32IxwoUlJSnjM6FQSBzs5Ozpw5Y/EkZ6zK++7KaN62iiPAtoX1k5uZF9rOPYK5WWF+9eONGzekYDQiIoLp06dLFaamIFMo8PjRb2isV1OSX8+1k1nk/cfv+tzn/v37JCcnS8/13/7t32Jvb49Wq+XBgwfSdsYG1ZMmTcLX13dIrQm648qVKz0MMnuDwWDg4MGD3LhxY9DHa2lpoaWlRap+TEtLY9++fX1Wqm7YsIENGzZYfCxRFPniiy8GJTgfxf9cvBTBU1NTE9nZ2dLPer0evV5vdgZJr9eTn58/bOXV3WFra8sbb7xBcHAwMTExZnk8LVq0iJUrV0pZiv5W2cYXzFBOvMbKNoADBw4MiVv3gwcPkMlkUv8rDw8PidJMSUkhPj4eg8FASUkJY8aMsbhzvLGKcMeOHZSWlvLll19KVFhf0Ov1ZGVl9bin+oMoimRkZJhFG/dnIGllZcWrr77Krl27KCws5NSpU9y8ebOHMHgwyM3NxdraekAu94sXL2bjxo0oFAoMBgNarZbCwkIiIiJwdXXljTfeYM6cOZw5c2ZQDuXPoqGhgfr6+l4zqUZ6aqAZr6CgIP72nY3smVKPq13XZ0wLUPOfqe3cM7PKvbS0VAqGvby8mDlzpll6MqeoaJqmvEnm3Ur0epHizz6jto9gpKKioof+xsPDg+985zsA5OTkSAub5uZm0tLSaG9vZ+bMmcNW9VZTU9NvA16ZTEZQUNCgTTl7M1998uQJTk5OfVKkNjY2A2IT1Go1Wq12tNLuJcVLETy5urpy+/ZtLl++jF6vZ968eaxatQpBEFCpVP36fBQVFeHn5zfkmh5TEASBV155hTVr1nDhwgVSUlL6zIo4OjpKFUyPHz8mNjbWZPXKcNKPVlZWrFy5kgULFnDy5Ml+szn94cGDB4wfP77XDIhRgKrT6SgtLR0U/ejs7MzmzZuJiori8OHD/brAy+Vytm7dyooVK4Cul2h/lENFRQVNTU399vCMCAwAACAASURBVCusq6tDJpOZRT8aM5UeHh58/vnnnD9/fkgovJycHLPa2vQGY0ZUr9dz8uRJjh49SnBwsDR5CYJAREQEe/bsQaFQsG/fPrKysgY9bmPg2Fsln0wmk7KMA4VcLmf39EBsNF+X3E0LbOff77b360be2NhIc3Pzc5nEyspKjh071q97f9B33sXrq4wrQOrf/z1qE4uT8vLy58TL77//Pg4ODuh0Oh4+fCj9/tq1a1RUVBAcHMzDhw8HTF2ZgsFgoLa2tt/gCWDBggWDXpxWV1dja2vbg36sr68361m6deuWxZmvUY+nlxsvRfCkUCjYvn07jY2NHDt2jPb2dilwuHr1KnFxcSYdv9VqNVVVVcNmmtgX/P392blzJzU1NRw9ehSVStXvPp2dnRgMBpMCzerqauzs7IZ1tRQeHs6OHTsoKSnh6NGjZml8nkVjYyOFhYUmrRlmzZrF4sWLyc3NxcvLa9DtEQRBYOrUqWzbto2CggJiY2P7dMy2srJCqVQiiiKnT5/mxIkTfQYAGRkZAH1WDRoMBnJzcy2iUORyOfPnz+eXv/wlzs7OxMbGUltbO+BgpKOjg+Li4kH30JPJZCgUCjo7O02ary5ZsoQNGzaQlpZGbGysxbq97ugreIKvxe/V1dWcOHFiQNRydWUFix1UeNl8/WxNC2znlzfbKegjYVlaWgrw3HkwGAy0tLT02w5IEASm/Mu/YPNVENLZ1ET5//d9DE97Lvra2tp6Nex0dXXlu9/9LtAz+2R8LuVyObm5uUNqyQFdz3BnZ6fZonqdTjdgw0q9Xs+jR4+eC8Dq6+vNqm5sbm62+D01Gjy93HgpgifoylRs2LABX19fDh48KPmPLFmyhNdff13SHzw76eTk5AxZg8yBwN7enk2bNhEQEMDBgwelF7EphIeHs3XrVpRKJXq9npycHOk7mXrBDAecnJzYvHkzQUFBxMTEWCwmt7Gx4Uc/+hELFiwwuU1LSwsqlYrS0lISExN7eAoNFK6urmzZsoWQkBAOHTrUr6O6UQs1efLkPjN5AQEBrF+/vs/KxpKSEry8vCymH6Grlc6OHTsIDg7m5z//OX/84x8HRFO1tbWxaNEipk6davG+3SEIAl5eXqxduxZBEHrt5wZddO/27dsJDw/nyJEjpKSkDKiPorW1NdOnT+9Xp6XRaGhsbDQ5HlPo7OykqKiIqMgIdkY44W7dM4D6P5dbKKjt/f5Tq9V4eHg8Fzz5+/uzZ88es9rUWLm6Mu3jj1FYK5g2P4BwHx0d//1T6HaN6+vrcXFx6dXt/Lvf/S7Ozs7odDpSU1MRBIHAwEBmzpyJIAhs2rSJ1atXm3s6zEJDQwNyudyszBN0UfGxsbFmta56FqbYgYaGBrMyTytXruS1116z6JjG4GmUtns58dL4PHXvnl1QUEBycjILFizoUT1VWFjIgwcPWLduHba2tjQ0NPDo0SNmz579Iob9HMrKyoiPjzfbWykrK4vExETeeust/Pz8yM/PRy6XM27cuBEacReqq6uJj4/H19eXxYsX9yuUNQeiKHLz5k2io6NxcHCgvr5eWuF2dnYOSfl+fX0958+fx8bGhhUrVuDk5NTvPsXFxWg0GoszNx0dHdy8eZMFCxYMSsQviiJJSUnk5eXh6urKihUrLCqnHyrU1dVRWlrKzJkzaWpqIiYmhrlz5/YZlKlUKq5cuUJ1dTVLliwZNjNNvV5vMSWZnZ2Nvb29RL1p9SKf5TbT1Pk1XXuvQMbfBFXz6gzLFlsGg4Hs7Gyio6P73a/sd/+EX+M9ZLKuQL09dC52u/6+xzaiKPYayH/88cf8+7//OwqFAm9vb+Lj458L3Do6OlAqlUO2WDQWdphzvhsaGmhpaSEgIMCi46vVau7cucPChQuf20+v16PVas3WdJk6d73BYDDQ2tqKvb19r8/sqM/TtxsvTeapO8LCwtiyZQv37t0jKSlJWqEbdStGOsbYeuGbAmOlVWVlpVk0XnR0NJs3b8bPzw+1Wk1FRcULoR+9vb3ZtWsXMpnMrEorg8FAYmJinzROdXU19vb2uLi4oFAopAAhOzubL774gubm5kGP293dne3bt+Pv78/BgwfJzMzslw7LyMjgwYMHPbI+Rn+wvvbNy8sjPDx8SKofV6xYwYcffsi0adP4/PPPSUxMNDub09TUNGj90bP0o5OTExMmTOhXgO7g4MDatWtZvnw5V69e5dSpU2ZdR0utR+RyOaIocuvWLbOoKpVKRX19fY/MkZVc4J3xTjh2u14zwwz8qdSHvfuPWuS+X1JSwoULFygsLOx3W//v/4SKxq/Fz7aFN9Bm3OyxjanJ/91338XFxQWdTkd5eTnnz5/v8feGhgb+8pe/8OiRZc2I+4JMJjM7UHVzcyMoKMjiwK0vdkAul5sdOCUnJ3P69GmzjyuTyXB2dh51F39J8cKCJ0EQ/lEQBM0z/zoFQdgpCMI/CILw9Kt/v+i2j5sgCPGCILQJgpAtCML8gR7fODG2t7dz5MgRWltbiYyMZMuWLSgUCoqLiyVH6W8S7O3tefPNNyUar69qPEEQpG7it2/fJjU1ddDaoIHCysqKFStWsGjRIs6cOcONGzdMiskfP37Mxx9/3EPc2h19VT+6ubnh4+MzZNdNJpPxyiuvsHnzZtLS0jhx4kSfGpXXX3+dTZs2IZfL0ev1dHR0cOPGDd5//32TQWNzczMqlWpAZqWmIAgC4eHh2NracuvWLfbt29fvtRdFkb/+67/mD3/4w6CO/fjx4x70o0wmY9GiRRK9UVtb2+f+QUFB7N69Gy8vL2JiYrh7926fFGR6ejrbt283K/gwQqfTUVBQYFY1qzEr9GxQYi2X8W6UE3bdXL/HuMs4Lr5O3Lkkzp49S3NzMx9++CHnzp0z+fnBwcG89dZbZhU9CHI5Y372/9NQ30URCoKAGPt/Edta+PTTT9m713QfPEdHR95//33p59///vc9KDJXV1ciIyPNohHNwYkTJyRvK3OhVqu5d++e2fqjhoYGk7qq0tJS/uu//svsyl9nZ2eL9EsXL17k6NGjZm8/im8XXljwJIrir0VRtDH+AxYAeUAj8P2vfp4O7BAEwWjC8XugDvAFfg0cEwSh/26xJmBtbc26desIDQ0lJiaGiooKyRcmOTmZrKwsqffbNwkymYw5c+ZI1XjXrl3rs6rNWH4fFhZmFv8/nAgLC2Pnzp1UV1dz+PBhSTfQHUY/GlPVUYWFhQQEBPRK/3l7e7NmzRrkcjlarZbbt28PiSmjp6cn27dvx8fHh4MHD5qsEJPL5VLQcOnSJb788kvS09NxcXHp1Rqit/LqoYJSqeTNN9/k7/7u71iyZAkXLlwgISHBpD1AbW0tTU1NvfpLmYuOjg7KyspMUsO5ubkcPHiQ8vLyPj/HaFS5fft2Kioq2L9/P2VlZb1uW1hYSFNTk0UO0Uqlki1btrBkyZI+t+uvpYqNXMZfRTlhI8gpb1RyPtuJghY7rri+jZ2TO3v37iU9Pb3PzJ8gCPj5+SEIAmq1ut8s2v9j773jozqs/O3nzmikGfWGKhJCQh0JSSCDBQJMNRgwmF6MwXaydmJvdvPZ5N04mw2JEye/3/s63pR1EicORXSM6d1U0VFBEuoF9YbQqI+kKff9Q8y1hNqogWN4/rHRzNw2d+4995zv+R6lmxuGRd+nva0jU25hJtL8p/+HhIQ7/QquY2JipCC2rKyMvXv3Ul9fz/nz5zEYDMyePXvIlgFGbt26xb179wb0mba2Nq5cudLrd90ZY3WgtwaLwsJCTpw4YbKGKjo6munTp5u8rdeuXeP8+fMmv/853y6+EWU7QRDMgO3A94EVwF9EUcwURbEI+DOwRhAEi0ev/ZcoinWiKO4BHgB9X/36XzeTJ09m/vz5HDt2jMTERLKzs5k3bx4rV66ULshDabkfKby8vHj99deprq5m//79PT6tGW/O0dHRvPrqq5JuZyBPd8ON0Zk8MDCQPXv2dAtEkpKS8PHx6fEJWKPRUFFRYZIHUX5+Pjdu3JCaA4aKXC7nxRdfZPny5SQmJnL48OE+S6eBgYEEBQWRnp7e61iS8vJyqfw4Enh6emJjY8PYsWPx8/Ojvr6e7du3k5mZ2e0mbWxVDwoKGvT6+is/+vv7M3PmTJM9xuzt7Vm2bBnTpk3j9OnTnDx5spvYOy8vD2dn5wGP+DD6+zQ1NXH9+vVux8NU81WlmYx/GW/H1FG26Awd33FChYxttS8yMXoKWq2WlJQUcnNz+wyMampq+Mc//mFS2cx5/iKqbL7W1VlrKnmhobDP49rU1ERLSwvvv/++9LczZ87wxz/+katXr0omns3NzSQlJQ25fFtVVTVgvZ29vT3f+c53TJJLFBcX4+Dg0KvprLHsb0q3XWdMbTypq6t73mn3DPONCJ6Ad4AiURSvAOFAcqfX0oFxQADQJIpiaQ+vDZmxY8eydu1akpKSOHXqFB4eHtINuqamhu3bt5s0xuNJY2lpyfLlyyVTzcdLF8XFxTg5OXUpY9XX13Pjxg0yMjKe9OZKCILAxIkTWbVqFYmJiRw9epSWlhY0Gg3p6em9Zp3S09MJCQkxSRcRHBzMG2+8Id1QhsuQ0cXFhQ0bNuDi4kJcXBzp6ek93mi8vb3x9vampqaGsWPHkpaW1uX1J9n92NbWRklJCR4eHixZsoRbt25x6NChLpqioZhjgmnlRzMzMyZOnIhMJkOj0fSbgYKOc8Xf359NmzZhbW3N9u3bSUpKkh5o8vPzh9QEkZOTQ0JCQrcs6P3793F3dzdJM6Myk7EuTOCnnZpDLxfBX4r9sHNw5LXXXuPatWscOHCg12y2o6MjgYGBJmeHPX/0Kyo6yQJfH2tBgFXv+htjluaNN96Qskt1dXVSZjY+Ph69Xi8NR66pqTFpO3pCr9fz4MGDQTUrmNKYodVqKSgoIDAwsNf31NTUYG5uPqAS/smTJ9m3b59J71Wr1c877Z5hnnq3nSAIMiAfWCWK4h1BEHKBzaIoXn30+lRgK7AZ2C6K4rhOn/0bcF8UxY96WO53gR8B9vb29s6HDh0yaXvq6+vJzs6mvr6eKVOmYGNjQ319PcnJyUyePHnEnHiHg5qaGu7cuYOHhwfjx49HJpPR1NSEtbV1t6xHc3MzKpUKmUxGa2srFhYWIzqwtS/0ej3p6emUlJTg6urK4cOH2bRpU7cbol6vp7W1dVBjQ9RqNVeuXCE6OnpYtUVqtZrExERUKhWRkZHdbAYSExM5dOgQs2bNoqWlhXnz5kmGkcZywnB0H5qCVqvFzMwMQRBob2+noKCA3NxcgoKC8PPz469//SsKhYK33357UMtvbm5GqVSaLBBOSEigrKyMBQsW9OkA/TgNDQ0kJyej1WoJDQ3l008/5aWXXuq3BNcboijS0tLS5bwSRZGmpqZ+R+n0xJeVPhyp8iHcU8OUsc1cjc/nRxM0IBq4f/8+mZmZeHp6EhISMvTvvrqSiTd3YGPbsRx1s0jKy98HedduU51OR3t7u3R+nj59mri4OBQKBevWrZO2IyAgABcXF1paWga170bUajUff/wxS5cu7dWrrTdEUSQlJQVzc/Neu1ZbW1uRyWR9njf79u2jrKyMH/7whyavu7S0lLa2jrmF/V0Pf/nLXzJp0qReLR5eeuml591232K+CcHTEuCnoihOfvTvO8AHoiiee/TvuXTom94Ezoqi6Nnps7uAq6Io/rmvdTxuVdAbVVVVlJaWEhUVRWpqKtevX2fu3LmMGzdOamE11tmDgoK+kV0WGo2Gs2fP0tjYiK+vL56enj2aFBrRarXExcXh6enJ/Pnzn+CWdsfYAeTi4sK8efO6BCKiKBIfH09UVNSgxOCtra3Ex8czbdq0YQ+A9Xo9t27dIiUlhdjY2C76pcbGRtLT04mOjqahoQEHBwdEUZQCgJ7aq0cajUbDnj17CA8Px9fXl6+++oq2tjbs7OxwdXVl8uTJA15mWVkZNTU1TJgwYUDbUVNT080R2xREUSQrK4szZ85QXFzMhg0biIqKGvByHicvLw83Nzfy8vJwcXEZVKAtivDrG62YqTpKunoD6DS2/PdUcwShY7+vX79OdnY2U6ZMYcKECV0CTmOTQUhIiEkeSdUHd+F0dz9yuYxWjZaHQYvw3PyO9LrBYODKlStdHv5aW1uZOnUqlZWVRERE8MILLwAdZbP33ntP2p6BtO53Jj8/nw8//JB///d/H9A5YeTkyZOoVCpeeumlbq81NTWRlJREbGxsn9v24Ycf0tzczG9/+9sBr78/2tvbWb58ORs2bGD16tU9vue5VcG3m29C8HQIuCaK4v/36N9xQJIoip88+vcPgFDg3+nQOI0WRbH20WvJwPdFUbze48IfYUrw1NMFpry8nOPHjxMcHMzUqVORyWSSyHLevHn9jtp4Whg9kA4fPszmzZv79BwyPuU5Ozv3aK73pGltbeXChQtUVlayYMEC3N3dgY7OmaampmEZYGoMxMLDw4c17V5dXc3p06extrZm7ty5vT65Jycnc/ToUTZt2tRnYDtS6PV6zp8/T2hoKJ6enoiiSEZGBleuXCE4OJiYmJgBZYJ0Oh3x8fHExMQMOpNSWFiIpaXlgMXKRn+srKwsYmJiCAsLG3QwqtFo+Nvf/saYMWNwdnYmJiZm0NnY+jY9f0qrRy7vKC3qDGCus+XHk78+rjU1NVy8eJHm5mZmzpwpCfVbW1vZtm0b0dHRJo+UKfvv72FWns3dGyXoRDNiDx3C9lFJKz8/H51O163EtX37dj744INu2aclS5YQGRnJlStXaGhoYNGiRYM6BiPFzZs3CQgIMKkrcDDBX3t7O/X19f02IIiiiMFg6DXT+jx4+nbzVDVPQsdZPR3o3LKwC/ieIAiegiAEAv8G7BZFsRk4CmwRBEElCMJbgA1wazi2paCgoJu+wcPDQ+oMO3jwIC0tLXh6erJ69WpJ0NjfTLOngSAIyOVy3n77bW7evMnZs2d77fYRBIGIiIguU8hTU1OHZUbaQKmurubXv/41AQEBTJ06lcOHD3Pjxg1pZEhAQMCwrEetVpOammpSm/pAcHFxYf369bi7uxMXFyc5Jj/uVyWXy3FwcBhUxmU4kMvlzJs3T9KClZaWYmZmRkxMDBqNhu3btw9oVEdf3Y+mYAzmrly5MuDP1tfXM3XqVFasWEFmZia7d+8e9FBqlUrF8uXLcXBwGPLsRzsLOe+E2KDTdVxizWTQLm/gl9dbqdN0/LacnZ1ZsWIF06ZN4/z58xw+fJja2lqUSiWbN28e0Cw+t598TFaFkvZWPYa2NhLffx+dRiNp3XrShK1ZswZPT09J0G7EqH2ysLBAqVQ+lWuBkcc75aqqqlAoFCbbKQzmOzx37hwHDx7sd7+N19nnPJs8bcH4eEAFpBr/IIriaWAHkAbEA38URfHSo5d/AAQBtcB7wApRFIfch97a2kppaWmPFxhLS0tWrFiBm5sbO3fupLy8vEtb8Y4dOwbcjjvSGC8wgYGBrF+/Hr1ez65du/q1XRBFkcLCwmGfcWUqiYmJJCQkSNu+YcMGysrK+MMf/oCzs/OwuIZDhzB306ZNREREAB2lteG6QRg78lasWMHp06f53e9+18UwURRF6urqePPNNyXR9EBH1wwntbW1fPHFF/z+97/nf/7nf1iwYAFz587l4sWLHDt2rF8j1paWFqqqqgYtMoeOY/baa68NKsPxi1/8gl/+8peMGjWK1atXExERwZEjRzh79uygGwTs7OywsrIacpdmQfpdUj/9CW3ajhu4mRzMVU384nY9/5ugo1XXcQMeN26c1Niwd+9eLl26JJ2PDx8+7HXId2d+8KMfUTJ7NvJHD39N+fncf9RV2duQZwsLC/7t3/4N6BCTGwMVtVrNvXv3mDx5MnPmzBlUALJt2zY+/vjjAX+uM3fu3OFvf/ub9OBnNF81xb2/qamJLVu2kJyc3O97HycqKoqXX365z/dkZ2fzu9/97htpZfOcJ8NTDZ5EUUwTRdHy8QBIFMUPRVF0FEXRRRTF33X6e5UoivNEUVSJohgpiuLd4diOvi4w0OGrFBsby6xZszhy5AjJyclSOnjUqFHDZio3HDx+gbGwsGDBggVER0dz4MABadt7QhAElixZwsKFC6Xg0OgR9SRISkrC2dlZMva0sbFh3rx5ODs7c+XKFe7evTtsQY5RRK/RaNi1a9egsh594eLigqOjI3Z2dly9elXa9qKioi7djwkJCZw8eXJYHNEHg6OjIwsWLECtVktdf0aTSnt7e3bs2NHnfL/09PRhmf3o4OCAUqnEYDBw48YNkwIfrVZLUVGRFLgJgsD48ePZtGkTZmZmbNu2jZSUFJNtRnQ6ndT9eOHCBb744oshdWiWlJTQWFnMOm857dqvj4+rrY46oY7vnWtmd5qIztDRhRgdHc0bb7xBe3s727Zt4/bt28TFxXH9ep+qBCkzK7i6EvKf/yn9vWrPP9AV50ql755YuXIlXl5eaLVaaXg1wNWrV7sEcAOdBZiRkTHk4NPT05PIyEipG7Cn6kBv1NTUcOvWrUHZsbi7u+Pj49Nn0FhUVNRlOsVznj2edubpqVNXV4dGo+nzAmNk3LhxrFmzhrS0NE6ePIlMJmPx4sWSqDQ9Pb1H08cnSUFBAR4eHt0uMKGhoaxZs4Z79+5x9OjRXm8KnTtYLly4wN69ewc1gX6g6PV6kpOTiYqKki5aRnH+q6++ytq1a0lPT+fLL7/sNxsyEJRKJRMnThwWLdXjpKenExMTw5o1a8jIyGDPnj2kpaV1KT/GxMSwatUqyaPoaZRIHB0dqaurw9/fn8uXL6PRaFAoFMTGxrJq1SoyMjLYu3dvt6fsmpoaDAbDsJkqQseN+tatW+Tm5vb73uLiYvR6fbeMsVKpZNasWVIpb9euXf2aR0JH+XHMmDGYm5szefJkFixYMKTmgqKiIuzt7fF3s+dHUfY4makwPPp65TIIcNNwrLCZeTvhdF6H0NzKyop58+axbNky7t+/T3t7e79axIqKCkRRZPTo0XivXo293xjCp3gydaYHAXdP9BkEKBQKvve97wEd56tRhlBTU0N2djYtLS3s2LEDUxpuOlNZWTnk88LDw4OpU6eiVCppbW3ttfzYE4P1eDJSW1vb5zn4fCjwc57p4MloIDkQfYODgwNr167FzMyM3bt3S5kZozPu7du3R3KT+8RYfuxtoKqjoyNr167Fzs6OuLi4fudvxcbGMnfuXCmYGskbe05ODs3NzV10HlVVVZibm+Po6IijoyNr1qyR9ERGQ8ehIggC0dHRkrfO7du3h0ULVVNTQ0VFBeHh4Tg5ObFmzRrkcjlpaWldsiFyuVzSHhUVFbFnz55hDQ5NwXgsXVxcSE5OpqioSHrN2dmZNWvWEBoayhdffMGVK1fQarWSyHy4Zz+OGjWKN954g/Dw8H7fm5+fD9DrDdVYyps4cSJHjx7lzJkzvZbAjOVHo2jbzs5O+h0N9vsoKiqSsqgqMxnfCbNiU6A95nR06bZpBZJLLMlXw7+cgGX74eajn6SrqyurVq1i6dKlXL16lcOHD/c669H4O/b09ESmUOC/+XW8/RwRBAFbQz3tSZf73M5Vq1bh5uZGe3u7JEEYNWoUZmZmWFpa8sorrxAdHW3yfut0ui6DuoeCKIqUlZVx9+5dAgMDTdYYDTV4Sk5O5uTJk71mlurq6lCpVCiVykEt/zn//DzTwVNZWRm2trYmmbJ1RqFQMH/+fCZOnMjevXvJysrCwsKCtWvXMnPmTKAjmHrSWYTMzMx+LzBmZmbMnDmTOXPmcPz48T5nzNna2krztoqLi9m/f/+A0/emotPpCA4OlnRIer2erKysLvoGuVxOTEwMS5cu5caNGxw/fnzYjC+N25CVlWVS1qM/ioqKMDMzk4KL5uZmPDw8eOedd8jPz2ffvn3dboZ6vR5BEJ64BUZmZiYKhYIXXniBt956S3IYN56/giAQHh7Oxo0baWhoYPv27Vy9ehVnZ+dBeW71h9G1Wa1Wc/ny5V5/R/n5+SiVyj6zxoIgEBISwubNm7GwsGDbtm1dDDaN9Ga+WllZyeeffz7gYF0URYqLi7t1U3pYmfHvEXbM8bTGRrBCLny9vuRK2HDYwOYjIpkPvp5PaJzb98knn3Dp0qVuDtjGmYnGINzptXWUVHwttDYc+yv0UV5SKpXSzLu0tDQSEhJ4++23paA0ICCgm39ZXxgzksMRPNXU1LBt2zZyc/suPz6O8aF2sMHTpEmT2Lx5c6/X0vr6+gHfN57z7UK+ZcuWp70NI85nn3225bvf/W6Xv+l0OqlMNNiOCVdXV7y9vfnqq69Qq9UEBARgbm6OwWDgyy+/lDrEnoT5pFqtpqKiguDgYJPWZxwCevfuXdLS0vDy8urzKaq6upqSkhLCwsJGpMPE1dWV+fPnSx1b+fn52NjY9OhzY2Njw/jx46msrOTChQs4OjoOy5gEmUxGcHAwvr6+yGQyKVAczP56eHiwfPlyKaOVlJREcHAwjo6OhIaGotfrOX36NHq9Hnd3d2QymdTlpVAoMBgMVFRUDMmo0FRCQ0OZMmUKLi4u0vGvqanhiy++wNvbWypdmZubExAQgLW1Nfv370elUjF69OgB2RoMhMzMTJKTkwkICOjx3LSzs5MMPvvDzMwMHx8fxo4dS2JiIomJiTg7O2Nra0tNTY30+30cS0tL2traCAoKGtB+arVa9Ho9ERER3byiBEHAw8qMFz3NWDu+w8Yg/QHoRZjq14ynUzP/e1tOaqWckFHgZG1GTU0NUVFRtLS0cOHCBRQKBS4uLgiCQG1tLSqVSprLlpObS6uowvFhNjK5DDN0aDFDPrb30nRwcDC7du2isbGRiooKxowZ0yWrWF1dzZUrVxg7dmy/+ja1Wk1hYSEzZswY0LzBnlCpVFRUVDB//vwBBeqZmZmoNyJjdQAAIABJREFU1WpeffXVQa1XqVT22T16/fp15HI5c+bM6fU9v/jFLyq2bNny2aA24DnfeJ66z9OToCefp8zMTJRK5ZC6hIy0trZy5swZmpqaWLRoEba2tqSmpmJmZjYiWprHEUWRa9euER4ePuCnIVEUSUhIICEhgVmzZvU57sAoku8sSh+OwLC9vR29Xt/FwO/mzZvExsb2G7gUFxdz5swZxowZw4wZM4bNrVsURfbs2YNMJmP16tVD2s/KykrKy8u7mTg2NDRw7tw5mpubmT9/fpcn9Tt37hAfH8/GjRufyjDnBw8ecPr0aRYvXtxN15GWloaVlRXl5eWkpqYyZcoUIiIiht3ssyfn7+Fabk5ODpcvX8bDwwO5XM706dP7XY8oioiiOCKmpqUN8IcELQ729RhPtfwH5ty+b8WyIDnvTRJxtup4obq6mkuXLtHS0sKMGTO6XMNaWlq4c+cO06ZNo/id1xjr2bGtOlGO2U92gLL3DNIf//hHyVBy7NixXL58Wfr9FRUVceLECZYvXz4sGSVTKS0tpba21qQy7nBTUlJCeXn5oExj4bnP07edZzLz1NzcTG5ubq/DWgeKmZkZgYGBtLW1cebMGZycnAgODpYEk/fv36esrGxYhbWdKS0tRRRFSV8xEARBwNPTEy8vLy5cuEB1dTXe3t49Bi3GY5WVlcWpU6fw9PQcFsHkzZs3+cEPfsDkyZNxdHQkNTUVHx8fkwJBOzs7QkNDKSwsJD4+HhcXl2FJpwuCIGW+Bpr6V6vV/PSnP8XDwwNnZ2eSkpKIjIzsVo6zsLAgODgYhULB6dOn0Wg00s181KhR2NvbSxqckaK4uJgjR47g5eXVRRxtZWVFeHi49LfGxkYsLCxoaGigqKiICRMmMGbMGHx9fUlISCAlJQUXF5dBub/3hiAIUrYnJSWFmpqaLjPZUlNTcXR0HLCFhSAIODs7ExYWRnp6OklJSTg4OODm5tZrYCSKIidPnqS0tBRfX99+11FbW4tOpzM5mLe1AG8HHfn1WowFRUcrPUFurSRVwO+uK9DqBexai7FSKoiOjsbGxobLly+TnZ2Nu7s7lpaW3L17F39/f6ytrWm3GoU88xoKczkyQUTb2Ig8uHftUnBwMHFxcbS1tUkNBIGBgVRWVkqdb0+yVGWsDkRGRpKTk0NdXd2gy3CDIT09ncTERCIjIweVfX6eefp280xqngYyXNZUjMLjJUuWcO7cOeLj4yVdRVpaGklJSSPS1qrT6cjLy5N0KoPFzc2NDRs2IIoiO3fu7LPNOCgoiFWrVkl6jt4MOE0lMTERhULBmDFjUKvVtLW1mTSWwohSqWT+/PnMnDmT48ePc/ny5WExL/Xx8ZF0Hzk5Ody5c8ckHdu9e/fIzMzEzMyMgoICPD09ey2JCoJAaGgoGzdupK6uThLym5ubS2UTtVrN6dOnR6TrMSkpiT179vSoezMGyykpKWzdupWamhrS09O7NFg4OTmxatUqIiIiOHToEBcuXOhmbDhURFEkNzeX/Px86finpqby85//XNL7DBYnJyfef/99SktL2b59e6/NAoIgYG1tjZWVlUnnwI4dOwY8IzDQ3oJ3Qh0Y7/h1wKWQw4u+LbwSVsfeLC2vnRjF/zlbhdbQ4Q+1fPly/vznP/Pzn/+cgwcPotFopFKZy6zZFKu/Dojld7+Cut6Hm9va2vLmm29K/969ezefffYZn332GTU1NSgUCkRR7Ka5epxPP/2Un/zkJwPa957Izc3Fx8cHCwsLkpOTB+Sn97Of/Ywvv/xySOufOHEi7777brfgXBRFfvazn3H5ct9C/Od8u3nmgidju/VQa/G94enpyeuvv05lZSUHDhyQSnnLly9HLpej1+tpbGwctvUZLzDDoTsxNzfn5ZdfJiYmhkOHDvUaLAiCILlj19XV8fnnn5OTkzPo9SYlJREeHo5cLpcmvw8mIzhu3Dg2btxIfX09O3fuHLTTdE8UFBSQl5dnkmdQWloaSqUSLy8vysrKTMpUWFlZsWTJEmJjYzl+/Djnzp2TgpDKykru378/ImL9zMxMnJ2d+ywN+vn5ERkZSVtbGxYWFt30ZZ39lbRaLVu3biU7O3vYGiaM/mOLFi2SzouCggLkcvmgsq1GsrOzGTduHM7OzixbtowZM2Zw4cIFDh8+3KPlyIwZM5gyZYpJ52ZPYnFTsFLIWDTGhvX+djgrv852OFrpWTqhnimBOo7XhzA7Dg5nQXlFpWTkW1paSlJSEnfu3EGn0yEIAk5v/wfqmo4OQ5kgoj38lz7X/9Zbb0ni8M5GodeuXQPg+PHjHDlypM9lFBYWDjnQb25uprq6WjqGS5cuZenSpSZ91mAwkJycPCiPp85YWFj0mHFqbm4mISGh1+7H5zwbPFPBk8FgGJH26sextLRk+fLleHl5sXPnTkpLSyU9xdWrV4mLixuWLjHjBWa4SztBQUGsW7eO/Px8vvjiiz6DPQsLC0aPHj3okmR5eTlVVVVERUVRWlqKnZ3dkEoDlpaWLF68mMmTJ3P48GGuXr06LBm/+fPn89prryGXy9FqtX0ek3v37hEUFEROTs6A2qsB/P392bRpE6Iosm3bNvLy8ggODubNN9+UghZTHKdNJTs7u0+dG3QYisbExJCbm4uvry/FxcU9vk+lUjF//nwWL17MjRs3OHjw4LD5npmbmyOXy2lra+PkyZNkZGTg7e096IeGhoYG6uvru3go+fn58cYbb+Dh4cGePXuIj4/vMQioqKjg1KlTvQbSxk67oYzf8bJWsDnInpc8LFHIvg7WAl3bWButpq5Nzw/OwHcvu9PkGIFCoWDevHls3LiR8vJytm7dSlZWFk6TJ1PW/nVgrChMRizK6HW9jo6ObNy4EaCLO3dqaip1dXX4+voSEBDQZ2BcXV09ZInC49UBKysrkx+o6uvr0ev1w9JEkpOTw4kTJ7otH5C82Z7zbPJMBU+FhYW4uLgMqO12sMhkMmJiYliwYAEnT57kxo0bGAwGwsLCeOGFF4ZkvmdkKFma/rCzs2PVqlWMHj2aXbt2kZeX1+P7VCoVixYtkrRPd+7cGZAreWJiIgDh4eHk5eX1eyM3BUEQCA4O5vXXX+fBgwcmjaYxZZlG/crFixfZtWtXj+WLxsZGCgsL8fPzG3D50YhSqWTevHksWLCAy5cvc/z4cSkAvHfvHv/4xz+oqakZ0v5ARzmwqqpKchbvi/z8fDw9Pbl9+zaHDx/uM/g3Zl/HjBnDnj17uHHjxrDNgGxubqaoqIjs7GyTuux6wujv1tNvx8zMjBdeeIHXX3+dxsZGKQjpHCzU1tZSUlLSa2ajtraW5ubmIQ99lgsCk10t+U6wPUH2XweJdS1yGlo7Lt3FrbYUv/AbPsoKxODoj6OjI6+++irz58/n9u3b7N27F/OVmygvqgOguqyBvANH+1zvv/zLv2BhYUF1dbVkLiqKIjdv3iQ0NJTIyMherzkGg4Gampohicp7qw4YTXL7y2gO1aagM83NzTx8+LDLb72uruNYPg+enm2emeCpvb2doqIiybfoSTFmzBg2bNhAcXExX375JUqlkkmTOhowamtrOXHiRL8agp548OCBJHwdKWQyGS+++CKLFy/m0qVLnD17ts90fHNzM3fu3CE9Pd3kdURFRfHOO+/Q3NzM2LFjh7Xt3dramqVLlxIVFcWBAwe4deuWyaM6+mLixIm8+OKLPeqYmpqaiI6ORqlUDnm4rLe3Nxs3bsTW1pbt27eTnp6Ol5cXISEhwzISqLy8HJVK1W/AqtFoKC8vx9fXl+nTp7N06dJ+g3+5XE50dDQbNmygurqaHTt2dDHfHCyOjo689tprtLe3m1QO7YnKykpUKlWfmQkbGxsWLlzIokWLuH37Nvv376e6ukMvFBoayqZNm3ptljBm5oZr8LOtuZylY21Z5WeLk4Wc9yOsWeJajIXs63M5vcmJRfvM+PFXUNXUce5s2LCBsLAwLhYUkCGMJj2hnFsXCsn6898o2rev1/W5uLiwevVqoGNQuJGkpCQ0Gg0Gg4GcnJweg8eHDx+i1+sHnXnqqzpgMBjQ6XT9auqMwdNw/EYiIiLYuHFjl9+6MfP03F382eaZ6babOnUqo0ePfionvLm5OSEhIdTW1nL+/HlcXV2xs7OjsLCQ1NRUQkJCBtRibzAYSExMJCIiYtiG5faFra0toaGhFBQUcOPGDdzd3XvsqjI3N+/ik9TY2IiZmVmfwnxbW1u8vLzIz88ftu7HzgiCgIuLC4GBgSQlJZGSkoKnp+eQso+WlpZSRqm6uprbt2/j7e2NTCbDxsYGPz8/7O3th+XmKZfLGTNmDF5eXly9epWSkhKmTp2KSqWira2N9PR0XF1dB3XcXFxcWLlyJaNGjerz82lpaYwdOxZbW1vMzMykJ+7CwkIqKir61A9aWFgQFBSEra0tX331FRUVFXh4eAwpSLa0tGT69OnY2Nhw9+5dfH19Td5/vV7fa/djT9ja2hIWFobBYODs2bPU1dXh4eGBUqlEFEVSUlKwt7fvsiyFQsHo0aOJjIwcNusMAAcLOVHOShyVchaOt2dlqEBGfgllbTasiKrHRmngQr4ZO1IE2vQwwU1gtLsL4eHhVI9yI//8DRSPsibVly5hHx6OVS8lf39/f7Zu3Up9fT0+Pj5YWlpiMBgwNzfHycmJffv2YW5u3u0cb25upr6+nilTpgzqwe7+/fsolcpu3ljQ4QUXGhra7/f24MEDysvLB+wN1RM9nVdlZWXk5+ezaNGiPq8jz7vtvt08E8HTn//85y2zZ88esRKXKQiCwJgxY3BycuLUqVPodDoiIiKYMGEClpaW0hgCU/Q+9+/fR6VSDchxd6iYmZnh7++PSqXi1KlTiKKIu7t7t+Npbm6OTCZDr9ezd+9eSkpKeu0ELCsrIzU1lcrKSoKDg0fErdqIhYUFISEhiKLIqVOnAHrc/oGSkZEhlYAUCgVqtZrMzMwhma/2hLW1NePHj6elpYUzZ84AHeNrLl++zLhx4wZ97ARB6PMY1NbWUlVVRVBQUJf3iaLIxYsXKSoqIiwsrN/j6OjoSHh4OA8fPuTs2bMIgoCbm9ugjr8gCNjb21NcXCyVek19iMjLy8POzm5AmRHjto4fP56ysjK++uor5HI5crmcY8eOYWlp2eVmb2lpib+//7AGTp23xYhKrqct4zgzo5zQK+W42+kIdG2lqV3GyRw5BzIErMwhzE2Oj88YPF9+mZKzZ6GhAUSRyq++wiVmMkq37tcRe3t7cnNzyc7ORqfTSV5SDx48IDY2Fl9f327nBHRok6ZOnTqowKm9vZ20tDQiIyP7fODSarXIZLJezx03N7dhCZyMpKWlcerUKenhbvTo0SxZsqTfB7DnwdO3m2fCJDMoKEi8efPmNybN2tjYyMmTJ5HL5SxcuBBLS0vy8vI4cuQIr776ap/DL9vb27l27RqxsbFPfIyHkYaGBikAWbBgQa8BX3Z2NkqlslftR1xcHHFxcfz0pz9lxowZI7a9j1NXV8fZs2fR6/XMnz9/yOn91tZWlEolLS0tLFy4kHXr1vG4o/1wolarpW68yMhIqcRhNDE1BYPBwI9+9COWLFnS67EXRZGrV68SERHRo9O5VqtFp9MNWL9nzMC2tLQwZ84caayIqZw9exaVSsW0adPQarUmZ7E0Gg23bt1i+vTpQ7Ipqamp4eLFizQ3NxMWFtZlmDV0lLpGjRo14P0aKLm5uXz66af4r36PBouu+puqBjOu5VtR1ahgnCP8ZCrMHgttD6q5vGwZ2qpKAiPc8PR1hHd+h6VvdznDvXv3mD9/PjKZjLVr10rByKJFi7rMoOyMTqdDLpcPKig2+nb1NQi5rKyMgwcPsnz58hE/vkby8vLIzMxkzpw5Jp/rBoMBuVz+3CTzW8xT1TwJgvAjQRBKBUFoEgThkiAIQUIHvxMEoV4QhHJBEN7p9P6xgiBcFQRBIwjCLUEQ+le60qHd+aYETtChp1i5ciVubm7ExcVRUlKCr68v8+bNk3QcvQW1mZmZ+Pv7P7XACTpKGStXrmTs2LHs2rWr17lfgYGBUuCUmprKjRs3uuxXYmIiTk5OAxo6OhzY29uzcuVKgoKC2Lt3LwkJCUPSQhn1EPv27aOmpmbEdXUODg6sXLmSyMhI4uPjuXLlCvfv32ffvn0m2xkUFRWRkZHRp5C7pKQEe3v7XkfEKBQKVCoVoihy9uxZk314HB0dWbFiBZMnT+bYsWN9Duztif3793Px4kXJRNNgMHD58mUKCgr6/FxGRgZBQUFD9ndzdnZmxYoVTJs2jeTkZI4ePUpZWZmkAfrtb3/L/v37h7SO/tDr9Vy7do179+4xSV7L4jHW2Ci+3i9XWx2vRdYzK7CRiiY9bx2DNQchR3QhZts2XpgzDv/xLlhamlH///4rJdnZ3dYxfvx4pk+fjsFgIC0tTfq78Xecl5fHgQMHuvx2Pv300y5eUabS0NBAQ0NDvwGRs7MzwcHBfY6S+uSTT/jwww8HvA29MW7cOBYvXiwFTlu3buU3v/lNn58pLCwctvU/55vJUwueBEGYDPwQiAWcgFvAp8CbwBwgBJgP/EoQhMhHH9sBXAZcgS+APaasazg624YbmUzGtGnTmDdvHidOnODWrVuEhoYik8lobW1l9+7d3X6A9fX1NDU1PbEnrr6QyWS88MILLFu2jBs3bnDy5Mk+he/GESVGGhsbSUtLIyoq6ol0Pz6OIAhERkaydu1a8vPz2b9//4C6BHsiLy8Pe3v7QY9zGAhGbyXjsN79+/dTVVVlcmBgDHh7K6lqtVry8/NNMl81epcZhbSmbn9QUBCbN2/G3Nyc7du3k5KS0m8QaxSvd+600+l0FBcXU1pa2uvnamtr0Wq1g+p+7G37x40bx6ZNm3B1deW//uu/+P3vf091dbWkExpJ8vPzpSHSoz09GWclsGmcNTGuKuSdkj5Ga4MorxbulIss3gsf5PnzcM7XmVF3VyU1H/4rx48c6fYdfv/73wc6HtqMzSK1tbWUl5cjiiJarbZLwF5dXT3geYzG7kdTGiwsLCyYO3dun510xcXFw2rnYcRoBpyfn9+nh5yxOek5326eZuZJD4iAMYUiB2qBDcD/FUWxTBTFNDoCpNWCIHgDkcAvRVFsAD4GfARB6Pfq/rR0TqYwduxYqRvv4MGDNDc3SxepzjqOvtqrnyZubm6sX78ec3NzKYvWE3PnzmXJkiUIgkBra6tUdpo3b94T3uKuODg4sGrVKgICAoaUhaqurqa0tFQKBo0DXIfqvt4fVlZWLFq0iBUrVqBUKqWZZ/0FgllZWVhbW/cozIUOf5uxY8eapCUyMzNj2bJlxMTEAAzIlsDCwoKXXnqJ5cuXk5mZye7du/u8Md2/fx9RFLsET+bm5qxevVoajPs4oihKth7DjZmZGVOmTOGHP/whgYGB/O///i8ajabP0tNQMQaQer0euVyOra0tW7duJe1uEtM9rPhOsAOBnawNFHKYPLaFFVF1CIgczYEllcv5b8cfUi90lOImjFXifHw/O+PiuHz5smRFMXXqVCZMmIBWqyUtLQ2ZTMb777+Pp6cn48aNY926dV2CpaqqqgF32lVUVGBpaTmg6kBdXZ1kGfA4arV6WDrtOpOQkMCf//xntFot9fX1fdoUGKsDz/l289SCJ1EUE4CjQA7QCnwP+AAIB5I7vTUdGAdMADJFUWx79HkDkPXotX9qrK2tWblyJZ6ensTFxVFbW8u6deukDFNubi4FBQVYWVl9o8qPRszNzZkzZw6zZs3ixIkTXLlypdsNVBAE6UYcHx/Pjh07sLGxeSKDk/tDEASioqKkLNS+ffsGlIUyGAzS7DXjANPi4mLu3bs3bCaR/eHn58fmzZtRqVR8+OGHfPzxx33uQ3Z2do+CX+iwW3j48OGAfIqMAt7Gxka2bdtGRkbvRow9YWyPj4iI4MiRI5w7d65HLyljae5xjyej7qm+vp5Tp051CVpLSkpwcHAYcEZkIISFhUlBeEtLCwkJCUMeHdMbGRkZBAcHU1FRgZubG9bW1oSFhUnfl72FnGVjbVkzzpZRnVzK0Vsg0vF9t+thuxhLrN1f+LvFYrRyc8Lt6pjZVE9bWxvbtm0jISEBvV7P9773PaCjzL5nzx7pd2w8d4zZJ1EUefDgwYCmN+j1erKzs03yGjOi0+nYsWMHt2/f7vaaKIrU1tYOe/Dk7u7OxIkT0ev1fQZP36TqwHNGlqcmGBcEYTawG3gZyAX+LxANRABjRVEsffS+9cAbwHZgsyiKczot4xywVRTF3T0s/7vAjwB7e3t750OHDo3wHg0P1dXVJCQk4O3tTUhICDqdjlOnTuHs7ExMTMw3KuvUE62trSQnJ9Pc3MwLL7zQo5i8paWF8vJyRo0ahZ2d3YCEziONKIrk5+eTmZlJUFAQfn5+/ZbC2traaG9vJysrCw8PD+nCaRSSP/7/I01lZSUXLlxg9OjRREVFdbOVMBgMfPHFF3h7ezNlypRun29ubkapVA6qW1Cn05GYmEhAQMCgHZ7b29vJyMigrKyMkJAQfHx8pPPj9OnTJCYm8sEHH/R4zpSXl5OYmEhsbCz29vaIokhTUxPW1tZP5BzbunUrqampvPfee2RnZ+Pk5ERYWNiwlab1ej2tra1YWVnxpz/9CVtbW8kRvCdEoMrckWpzB0KbCshvsmVvhS/5LV1v/t76Sj5r/i0OhWmkuU2mJfoF7qWn09jYSHBwMH/4wx+kUS3r1q3jlVdeAZAsHJycnAgJCeE3v/kNCxYsYOrUqSbtj9GzaaCdieXl5djZ2XXrqGtpaeGjjz4a0DYMlC1btjB58mQWLFjQ7bXOv52XXnrpuWD8W8zTDJ7+CLSJovgfj/5tC9QDOiBEFMXcR3//Dh0aqB3AB6IoTu20jGvAr0RRPNXXusLCwsTU1NRvzA26P1paWqSn54ULF5Keno6dnR0hISG0t7ejUCi+0ftiLDHGx8czefLkLt1Ioihy/fp1xo8fj52dHWVlZZw/f54lS5Z8o7JqarWas2fPYjAY+uzIa2tr4/r16312PxYWFnLkyBFee+21YTNO7A+jF1h8fDzOzs6sW7fOpCaD6upqiouLJSPXodJfiaMvqqqqOH/+PACzZs2S9EoajaZPHWPnQDU9PR1ra+shu32bSnJyMomJibzxxhuIosidO3e4e/cuERERREdHD8nfytj9GBkZibW1Nbdv30ahUBAZ2SEJbW9v59atW4SEhHTTBHV+QBFFOJkHn9zSE+zZyNV8a2qbzXA21PFF4wdYlmRTYRvO+A9/TXlNDZcvX+b27dvSoF03Nzdu3Lgh7cu9e/ewt7fH1taWgwcPMnXqVAICAvrdn+HqfuyMWq3mL3/5Cy+//LJ0XIYLg8HAgwcP+OSTT5g1a1Y3yUF5eTnV1dVEREQAIAjC8+DpW8zTDJ7+DyDvFDxZAg10lOw+EkXx0KO/fwI8APYDtwFnURQNgiDIgGpggiiKfebHR48eLV6+fHnQ4xyeBqIokpCQwI0bN3B2dmbNmjUIgsDBgwextLRk4cKFT3sT+0WtVnPq1CkUCgUvv/wyNjY2lJWVcfr0aerq6njvvfeoqqri2rVrLFu27IllZkxFFEXu3r3LjRs3mDRpEpMmTep2kU9JScHZ2Zny8nK8vLx6FCRrNBquX7/O9OnTn4ipaWdOnz7N4cOHiYiIYPHixXh6etLe3t7jTdxgMHDlyhUmT548LE0WeXl5HDt2jOXLlw96gK9Rr3T16lX8/PyYNm2ayduWkJBAdnY269ate6oPG/X19cTHx1NWVsa0adMICQkZ1PYUFRXR2NjY62xOjUbD559/zpQpU/oNfkVRZE9eA8VNWvQGuHnfitQyJaP1DzjY+BPMq0vIKJYT8YdPsfLxITU1lZUrV0ri8I8//hgXFxdkMhkzZ84c8L5ARxlw9OjRgx7lUlJSQn19/YjPKjVy69Ytrl27xrvvvtvtHNTr9cTHx/Piiy9KWbTnwdO3m6cpGD8ObBAEIUoQBBWwBTgNfA78hyAIjo868tYDB0RRzKNDH/XDR+//GXCvv8AJOmrzxiexfxZfK0EQiI6OJiAggKqqKi5evIhOp5Pcpv8ZcHBwYM2aNdKA5Hv37pGTk0NxcTGpqamYm5vj7e3NmjVrUCqVGAwG7t69OyyDfIcDY0feunXrKCoqYs+ePV1myhn1Da6urnz00UfSk/njqFQqZs+ejUKhQK/Xc/369SFPnTeV+fPn89FHHzFz5kyOHTvG2bNn+dWvfsWPf/zjbu+9f/8+7u7uw9ad6u3tzcSJE4dk5mrsKnzjjTdQq9W8/vrrkklrXxjLSU/yXNJoNHz55ZeUl5ejVqs5duwYbW1t2NnZsWjRIhYtWkRycjK7d+8esB5Kq9VSUFAgjdKpqakhOTm5S4erSqXirbfeMilr+KBVT3lzhy5MLoOpfs0sDmtArXJig83PaVE60JyfT/zSpVSePcuECRN4550O1xgHBwcyMzOJj4/n+vXrtLa20tLSwqVLl3qd9/c4Dx8+RKvVDmkGXnp6OtevX+9yLozk9d3f35+XX365x3J2Xl4eXl5eI2KM+pxvJk9TMB4P/ISOjFIlEAS8BfwNuAsU0WFH8G/GEh6wEVhJR1febDq0UP1iY2PD3r17SU9PZ/v27b1Ohf+m8fDhQxwcHHj33Xdpbm5mz549+Pn5ERYWBnQ8iZrqrfO0kMlkTJkyhWXLlnH8+HHy8vJISUnpMlzU+N/CwkLOnz/fr1/Pk8be3p4VK1YQFhbG/v37uXnzJjqdTmqvLigooLW11STxe2lpKbdu3Xpi56AgCDg6OhIUFMTkyZO5du0aFy5cQCaTdbnRtLW1UVxc3KdB60AxNzeXsm06na6LVcVAUalUuLi4UFdXR25ubr9deQ8ePGDmzJmsX7/+iWWdiosyA31uAAAgAElEQVSL+dvf/kZhYSEtLS2UlJR0CbY9PT1Zv349ERERHD9+nBMnTpgcbOTk5ODr6ytlLo26r8c7zoyBb2NjY5+BhIvKjM1B9riqvg4ERjtoWTWxDr2rK69b/heNBgt0TU0kfO97pH/0EW9s2IBSqUStVkvr1Wq13L59mwcPHvDZZ5+xatWqfgMYYzZxqBmj2NhYNm/e3OX7/fLLL1m5cmWfg6sHi6OjIzqdjnfffZf8/Hzp7xqNhoqKCsmF/TnPBk/VJFMUxa2iKI4TRdFOFMUloihWiaKoF0Xx+6Io2oii6NVZDC6KYo4oipNFUVSJojhdFEWTzDQsLS1ZtGgR586do7S09BtXHuqJzu3VSqWSxYsXM2HCBPbu3cu9e/cQRZG0tDQSEhKGbWL9SGJra8ukSZMwMzOjtLS0x24UX19f1q5dK93A+xsA+iQRBIHw8HA2bNhAeXk5n376Ke3t7djZ2ZGamgogBbV9MWbMGDZt2iTt40hbGXRGoVDg4+ODhYUFWq2WQ4cOSb4+mZmZBAYGDutImc7Ex8dz4MABmpqaBr2M/Px8VCoV77zzjtSVd/bs2W6ePgaDgczMTCIjI5HL5bS3txMfHz/ix9ro7ePt7Y2npydvv/12t/NcEARCQ0PZvHkz9vb2xMXFce3atT4zkU1NTdTW1nYpfZaVlSGXy3u0BaiqquLzzz8nuwfjy844Kc3YGGDPi65fZxqVCpH5IY2MifFh6+rtaOUd5d2Czz8n61//leWLFwN0Mc28fPkyDx48wNfXFy8vr36D1eLiYpycnHqcjzkQrKysupXB1Wo1bW1tI3aNLy8vp6CgoEvZOz09nZCQkGHTbT3nn4Nn5tsePXo0b7/9Nn5+fly9ehWNRkNWVlafxo5Pk8cvMIIgMGHCBFauXElCQgInT55k9uzZrFixAjMzM6mF9ptKRkYG4eHhKBQK7OzsKC8v5/z5891uGh4eHgiCQENDQ4d3TaeL9DcBW1tbFi9ejK2trSSKT01NxcPDw+T2aGMX2sOHD/n888+fWKZt/Pjx+Pn5YW5uzqZNm3B0dGTXrl2cP3+exsbGEZ2VOGXKFBYuXDikG+b9+/fx9vbG3Nyc0NBQNm3ahEKhYPv27SQnJ0v+XI+XH8vLy0lISOjTRHM4KC0tRS6XS8fReIPNzc3tlgkxNzdn6tSpvP7669TV1bF161bpoehxevJ3Ky8vx93dvccbtouLi8nlUrlMYIaHFev97bA1/3pZQW5tTJjvw7H3D2AQZLh52WJWlsX4K1cQBIG8vDxpnwwGA6mpqSQlJaFQKPrMPBnLj6YIyk2huLiYL7/8UnqANNoUjFS2MSsrq8t19uHDh+j1+gF7Wz3nn59nJniCDj+lVatW4eTkxN///ncOHDjAnTt3nvZmdaOvC8yoUaNYv349CoWCXbt20djYCHSMTNixY8eQnuxHipqaGukCY2VlxaxZs/jud79LW1sbO3fu7LGco1Qq8fX1/Ub6peTn5xMbG8ubb77Jw4cPuXDhwqAE0cbhzoMZojpYsrOzUSgUFBQUkJuby4oVK7h9+zbp6elDKqv1h0qlkowDq6qqpLb3gZCfn9+l6UOpVPLSSy+xcuVKcnNziYuLo6CggJKSki7lRx8fH958880RL6sUFxfj6enZJXtXV1fH8ePHSU5O7vEztra2vPLKKyxevJjU1FR27tzZJcirqqpCoVB0C8wrKip6DY4EQSA2NnZAXY5e1greDLIn1OFrzY6tyoD/VDcu/PhLImO9iX7Jh/GOAhMeDf7u7OVlMBgwNzenpKSEDz74oNeydHZ2dpfy41DR6/XSaBfoyDwN1iLDFGxsbHBycsLd3X1EzVef883n6Q1Ie0rI5XJmzJiBu7s7hw4dwsrKClEU0ev1T3VeXGeys7Px8/Pr9QKjUCiYN28eOTk5HD58mEmTJjF+/HiUSuWQU+HDjSiKZGRkEBUVBcCyZctYtmwZAAsXLiQ7O5sjR44QFhbGiy++KN14zM3Nu7QC37lzB29v7yEJTIeDlpYWKisrpfbqpUuX4unpyaVLl7h48SJTp041uR3d0tKSV199Ffj6OAUGBo7oeThx4kQcHBx48cUXqa6upr29nQULFqBSqTh27Bi+vr7ExsaO2EgjURQ5c+YMgiCwYcMGkzMEGo0GV1dXSTDdGWdnZ1auXElOTg7bt2/H398fjUbT5bdgDCTKy8vRarUjYl1QVlbWbbnGOYq9Obkb8fDwYO3atWRnZ3Py5ElcXV2ZNm0amZmZ3cb9iKJIeXm5ZMjaGw0NDZIVgClCZqVcxmIfG3xtzTla2IQgiGRXWXBJE4ab9Rv8RBNH1DRvNur03M19QEZGBhEREcjlcsrKyiTbBG9vb86cOYOjoyPTp0+XTDMbGxtRq9XDGmz4+Ph0CYpra2tH1N29tbUVBwcHzMzMKCoqGpby43P+OXlqVgVPkkmTJokJCQnd/l5bW8vRo0dxdnZGrVbj6+s7YsZqptLY2Mjdu3eZNm2aSTeW+vp6Tp48KdkBWFtbU1dXx4ULF5g7d+6IuiqbQmFhIc3NzYSGhqLRaLCwsOhWamhqauLs2bM0NzezYMGCbpmY1tZWtm/fTkBAAC+99NKT3Pxu3LlzhzFjxnRL02s0Gi5evEh5eTnz5s0bcCaqoqKC3bt3M3v2bMknZqTR6XTEx8fj6+uLTCZj1KhRXLt2jdzcXGJjYwfdUt8fdXV1yGSyHg1Uh7rclJQUZDIZaWlpREdHExUVJQXkoiiye/duDAbDgAI3U2lvb6e5ubnXzIdWq0Wv1/erx9FqtSQlJXHmzBl8fX2l0TtGDAYD+fn5WFpa9pmZraioYP/+/bz66qsDnrVX16bnf5Ja2JlohVbf8Xv9z5YdvNt2CFEUmbc3hYymdmbMmCEFtPb29ixYsICAgAD0ej0pKSncunULHx8fYmJiyMzMJCAgYNjdv6HjuxVFkV27duHq6jpiY5+OHj1Keno6sbGxtLe3Exsb2+tD7nOrgm83z3TwBB0XvHPnznH16lVee+01oqOjn/DWfY0oity8eZPAwMABXWAMBgM3btwgLS2NuXPnIpPJOHfuHKtWrXqqxpPt7e1cu3aNadOmoVAo+PTTT7l16xZbt27tFkB1NtaMjo5m4sSJXd6j0WgwNzdHLpfT2NiIUql84p5JNTU15Ofnd8kEGLOXxot1QUEBX331FT4+PkyfPn1AwlWjkF4QhBFxXVer1dTW1uLj44NcLicrKwuFQkFCQgIajYaNGzcik8moqKjgq6++wsLCgjlz5ozIzQ46vvOkpCQCAgKGHOSLosi1a9cIDw/H1tYWtVrNpUuXUKvVzJw5E19fX6AjG6NQKJ74sHC9Xk9cXByurq49OlM/TmtrK5cuXQI6NFxTpkwhPDx8wIL+/gxF+0JngHdPwNlOkry4xi28aMjgdF4V75zPw9HRkRUrVgAd5cLvfOc7uLu7o1arsbGxQa/Xk5CQQHx8PE5OTmzYsGHYj31tbS0HDhxg9uzZw9ot2hv37t1j586drF+/vs8mkefB07cb+ZYtW572Now4n3322Zbvfve7Pb4ml8vx9/fHxcWFmzdvYmtrS11dHRqNZtifjPujqqoKjUYjXehNRRAEvL29cXNz4+zZs8hkMhYvXiylk/Py8rC3t3/iRoEZGRm4u7tLN9+//vWveHl5MWvWrB73wdXVFX9/fxISEkhPT8fT01O60CoUCmQyGQaDgf3795Ofnz9imZGeEEWRxMREIiIipLKcKIr8+te/RqfTERsbC3SIwcePH09JSQkXL17E3t7e5ODD1tYWQRDQaDQcOHAAe3v7Qbtz98SFCxfYsmULc+bMQS6Xk5OTw4QJExg3bhz+/v7SsbaxsSEsLIz29nZOnz5NW1sb7u7uw96J19jYyPHjx5HJZP1m6n7/+99z7tw5ZsyY0ePrZWVlGAwGaTkqlYrg4GDs7Oy4dOkShYWFuLm5YW9vL4mac3JycHJyGpZz6N69exw+fJiAgIAeS2RGawhvb2+THmjS0tLw8/MjMjISHx8fkpOTuXPnDjY2NtTU1JCYmIiXl1e/34nxAaOpqWnA7uYyAeb5QWIFlDaAvUqHRbAfaePnsqw+gWPZFVTWd/icGa+VaYmJ+AUEsHPnThwcHHB3d2f06NE0NzdjY2PDpUuXJP3jcJ1PFhYWVFdX4+HhgYWFxYjLL+RyOVZWVkyePLnPc+cXv/hFxZYtWz4b0Y15zlPjmRKM94axDX358uVcuXKFrVu3cuXKlSdqqGkwGMjKyhrQgMzHGT16NK+//jpNTU3s3buXhw8fUlZWxpEjR564H1RDQwN1dXWSoWd1dTVlZWX9jkywt7dn1apVjBs3jj179nD37t0u34NMJiMmJoZJkyY90WCwqKgIZ2fnLrO0KisrefjwYbenT2PG5pVXXuHKlSscO3ZMcmY2BWN5Z7hbn3NycrC1tcXNza1Le7VKpZJu6FevXpVKX1FRUWzcuBG1Ws327du5f//+sG6Pra0t69evJyYmpt/3ZmRk9Gp4qdPpyM3NJSgoqNtrvr6+bNy4EU9PT/bs2UN8fDzt7e3k5eVx/PjxLn49QyE1NZVDhw71eeOOjIw0SWtlfHgzCsJHjRrF8uXLmTFjBlevXuVPf/oTH3/8scnBR0pKCn//+98H1Y2rNIP3HiXj5wQ1onK1otXcmlTfWfxlth8eVuZcv36dtrY2Ghsbidu3j02zZ+PyKFCEjmysr68vCxcuZM2aNTx48IB//OMfpKSkDIuJqUwm45VXXqGtrY1ly5Zx8+bNIS+zN9566y0+/vjjLj51z3k2eR48dcLV1ZUNGzYwYcIENBoNjY2NaLXaJ+IGnZ+fj4eHx5BT2iqVisWLFxMVFcW+fft48OABS5YskQzpnoTjcucuFOMFxthtZMq8KUEQmDRpEqtXryY9PZ2DBw92MRMcN26cJBJNT0/n2rVrUpv6SNDe3s79+/elbjEjxoC0N7O/0aNHs3HjRuzt7dmxY0evreiPY2try4YNGyQ9S0VFxbAE8llZWQT+/+2deVxU573/388AsqMCAoqKC6sbokZRwQ23xo1oYlxjFts0tzbtzS+3Tbdb23u75N7e3vR2S5tUjXEhxiRqTFTcUAziBoigIAgKsu+yM8w8vz+GmYDsCAL2eb9evBzPmTPzPOfMOed7vsvn6+NDUVERer3elMhrxNi7Ky8vz/R9dnZ2rFixgpCQEE6fPs2RI0dMFZ7dgbGsvKqqilOnTrWoxVRXV8f9+/dbrZZLSUnBw8OjVc+Kubk506dP54UXXqC8vJydO3dSV1dHaGhot7VsyszMxMXFpUPnb0xMDGfOnGlxnTF8PWHChCY3ZyGEyRDUaDRotVpOnjzZoWMxZswYpk+f3mXtozENKVzXs76eW8y4FYwbP4HzayfxmpcdEeHHOXLkCGVlZSQ+eMAPf/97Xlu6lKQGiQjjfnZ0dGTFihWEhoaaEvyTk5O75fedl5eHTqfr0VSFnJwczM3Nqa6u5ssvv+wz3RAUjx9lPD2EtbU1zz33HBMnTmTv3r3s3buX/fv396gQZU1NDVlZWd12IRdCMHHiRNatW8eNGze4efMmtbW11NXVsWfPHq5fv94t39Maubm5WFpaNglXxcTE4Ojo2KkqJycnJ9avX8/w4cPZu3dvi8ZHTk5Oj+v3JCcn4+np2SzHKiEhATs7uzZDTubm5gQHB7N69WpiY2M5ePBgM1XoljDeOAsKCggLC+PatWuPNAej4rW3t3er6s4ajYaVK1eycOFChBDU1dWZ9vfo0aPZsmULTk5OfPjhh1y9erVbbxzZ2dncvHmTgoKCZuvu3buHXq9v0XiqqqoiPz+/QwnRdnZ2PP300yxfvpyYmBiuXLlCfn4+FRUVHVb6bo379+93uG2S8fta2n9ZWVk4ODi0mjKgaZAJmDFjBra2tuzevbtdkU17e3tmzZrV5dYhbnYGD9SdAkvKawy3jCqpIWHNf2Adsp5tU0by+pxJ+AwY0OSGciY9naWrVnHoD3+gvKSk6We6ufHss8+yYMECLl++zN69e00io11BSsknn3xCRUVFjxlP1dXVPHjwAC8vL6qqqsjIyOjT2nqKnkUZTy1g7Cu3cuVKcnJy0Gq1Paoe21Pqzo6Ojqxfv97k+TCW1vakDopOpyM5OZlx48Y1Wb5s2TK++c1vdtrVbWzvsmbNGq5du8bhw4ebhMBCQkJYvXo1Go2G2tpa8vLyumUeRozhx5bKn2tqavD39+/Qb8PV1ZWNGzcyatQo9u3bx5UrVzrkLXN2diYkJKRD6uVtcfv2baSUDBw4kCFDhjQJPzbGzMzMtC/379/PhQsXTOssLCyYPXs269at4+7du+zZs6fTPdpaw9PTk61bt7ZY0m8MF7b0cNEVdWdjm5Tx48fz6aef8stf/pJDhw512fuh1+vJzMzssPEUFBTEqlWrmp3vbYUfjRhlCkaMGEFwcDCbN2+mrKyMHTt2EB8f3+ZvKisri4sXL3ZsUo3QCBg1EPRSEN/I+3S5qA79vOfhO38gonIA059/ni/27WN6g5yIi4sL8xYt4tNr1wicMoU//vjHTTyLQghGjRrFpk2beOqppzh9+jQff/xxm213WkMIgYeHBzY2Nj12fbtx4wYWFhY4OzszduxYXn311R4rplD0fZTx1Abu7u5873vfY9CgQXz88cekp6cTGRnZrSGikpISampqcHNz67bPbIy5uTlz585lyZIlnDlzBnt7e1M46ObNm10SK2yLtLQ03N3dm4UIJk2a1OXu62C4EG/atAlnZ2d2795NUlKSqSLN6BG6cOECYWFhzdp1dJXGPbhaMvp+9KMf8eMf/7jDn6fRaHjqqadYv3496enp7Nu3j/z8/Da3MebjWVpaotPpOHXqFCUPPcV3BB8fH37+859jbm7eIXXnAQMGtNqE2tHRkTVr1hAYGMjRo0c5ceJEt/QSs7GxAQwFDo1zDgcPHkxwcHAzUciCggKklM3Cjx1Bo9EwceJEXnrpJaZOnUpeXh7Xrl3rkjetrKwMa2vrDhtPGo3GFKpsLDSZkpLCqFGj2kzsLisro6qqyrQvHBwcePrppwkNDeXWrVvs3r271dy0e/fucf369S4dq9EN9sitXEs0GM6Fklo9KWV1SMeh5BcUMmTIECbNnctnMTH8/nvfYcXy5YwePZqQkBAq9Hp++8EHzB4/niO7dzcxVIUQ+Pj4sGXLFnx8fDh8+DBHjhyhuLi4U2O0srLCwcGhR5rz1tTUcOfOHSwtLbG3tzcdQ8U/L70qVSCE+BfgB8Bg4CNgG6AF/gdDk+BK4JdSyncb3j8a+BCYCsQDL0opb7X3PW1JFXQEvV7PpUuXOHLkCA4ODnz/+9/vlnLbh8ure5rq6mrCw8N58OABS5Ys4ejRozg5OZmEGh+VmpoaoqOjCQ4ObvJUnZCQYOrp1R3k5OSYRPgWLlxouulWV1ebQlPAI5f75+TkkJOTYxL47E6MhllkZCTjx49n5syZ7UovFBcXs3//fubOndulpqrx8fEMHjy4wzf5xuTm5uLi4tLMw1NbW0tUVBRJSUnMnj2biRMnPvJNJSIigqysLNauXdvqPtHr9URGRjJt2rRWvWidoaioiLNnz1JQUMDSpUu7pEau1+s75QGLiIjg+vXrbN26FYCrV68SHBzc5mdIKSktLcXc3LyZvIOUkjt37nD+/HkcHByYM2dOEz2yurq6Jg8bneG/voI/N1xCvzOrEr25wQBztzVnk9dALl26hLOzs0EqQK/j/P/8mLNVNqbt09LS+Oqrr0yGm/+wYWz/3e+Y3kL1pFarJTY2lqtXr+Lp6UlgYGCHro+XL18mISGBoKCgbmv/YiQmJgaNRsPZs2d55plnGDVqFHfu3OHixYusXbu2RYNXSRU82fSa8SSEmAUcBBYBucAnwGkgG/ge8A3AETgLLJJSxgohIoHzwNvAq8BGKWW7ioKPajwZycjI4NChQ0ydOpXAwEBKS0txcnLq8udlZmZSWlr6yCGZzmBsKHzhwgWmTJnCpEmTsLGxoaamBiHEIz21xcTEMGzYsGZetB/96EeUlJTw7rvvPurwTdTX1xMVFcXNmzdZsGBBs4tldnY2586dY9myZV0yTHU6HZGRkQQGBraYaLtz507S09P5xS9+8UjGQmVlJWfOnCE/P59Fixa1W7LfWLentra2w8drz549WFpa8uyzz3Z6vCUlJezatYsZM2a0WhmXn5/P6dOn0ev1hISEPJInVUpJfX29SVKgsrKymYpzeno6NTU1j1Sd+jAJCQkcOHAAJycnRo4cydy5c3s0LFNTU0NlZSVOTk5cvnyZ0aNHd8mL9jA6nY74+HiTQOXs2bObGFpSSmpqajr1AHggEf7tlOF1qK8ed9di9A23jm+PG8wgy68fluq1WpI/3UF2ahJR9V/nH2m1WuLj44mPjzeF7+b4+rL9//4PnxYerKqrq7l69Srx8fGMHz+eGTNmtDvmEydOkJqaymuvvdZtqRYlJSUkJSURGBjY5NzJyMggOjqaxYsXt5hnpYynJ5veDNuFAvuklIlSyiLgD8BaYBPwX1LKLCnlDWA/8LwQYiQQgMET9QCDd2qUEKL1BIFuZuTIkbzyyitkZWXxzjvv8P7773c57FVfX09qamqL7SZ6EmMYaN26daSmpnLixAmqqqo4ceIE+/fv73IScHFxMbW1tc3ap9TW1pKYmNjt3htzc3PmzJnDihUruHDhAkePHm0SjqitrUWr1XZZSPPOnTsthh+NXLt2zfQk/yjY2tqyYsUK5s6dy/Hjx9sNgRlvHiUlJezYsaNJ2Kc1CgoK+Nvf/tbl8Q4ePJhFixa1eQxdXFxYt24d/v7+HDp0iNOnT3e56bbRO6LT6Thw4AArV64kPDzctL6uro67d+82q358VMaMGcPChQt59dVXGT58OGFhYZw7d67deXz00Ue88847nf4+KysrnJycKCgooKampkOGU1RUFGFhYW3mZ5mZmREQEMBLL71kSiqPjIyktrYWwBQW68yDs10jx0qNVoOb9deSDJn5RVy5csW0n26npGAzfTELv7+dqY5fb2hhYcHUqVNZv349EydOxMzMjPNJSYQsXsy316whKzOzyXdaW1sTHBzMli1bqK+vZ8eOHVy8eNE0j4fJzc1l8uTJvPTSS91mODWuHNbpdE322ciRI3tdiFjRe/Sm8TQAaHwWCGA0MAlo3EUzEfAE/IFbUspaACmlHkhqWPfYsLW15dlnn2Xq1Kk8ePCg1RO5PW7fvs3o0aM7LVzXXRiTyZ2cnNi9ezdDhgxh2rRpXUpabys3KCEhAa1W2yGJgq7g7u7O5s2bsbOz44MPPiAlJQUwVIdt3rwZa2trpJSmHKmOUF1dTXZ2dqtipVVVVaSnp3cpdNYanp6ebNmyBQsLCz744IN2x2tra8uYMWNabQ7bmOjoaJNuU1cx9k7U6/VcvXq1xepTIQQTJkxgy5Yt6PV6du3aRWJiYpcTsXU6Hbdv36a+vr5JG5KkpCS8vLy6XQzRxsbGpApv1Liqra1l165dbSZjx8TEdLlSTK/X8+WXX3L58uUO5bJduHCBL7/8skNGsKWlpSmpvLKykp07dxIXF4enp2enQ+iaRl8nAWvzrxck3k7l3//936moqKCyspKCggJGjRqFsB3Ism1vsX5hEC7mX+87KysrZs6cydq1a/H29kYCn0dHM3vmTN761rea5TrZ2dmxcOFCNmzYQElJCTt37uTatWvNfoPf/e53OXjwoCmM3x3cv3+fgQMH4uDgwMGDB1m+fHmzysaelElR9F16sxPuKeAdIcR7QAXwOmAGOACN6z/LG5Y9vLzxumYIIb4F/BswaNCgQaZWB92Jv78/f/rTn3B3d+fBgwf4+/t3SBFar9dTVVWFnZ0dd+/e7fZxdRYnJycOHz7M0KFDyc3NpaSkhHv37jF58uQO3aDq6urQ6XQtltMfP36cyspKU7uMnsTZ2Zl//OMfODo64u/vbwppZWVlER0dzcyZM9tt0AoG42jAgAFERka2uP727duUlZVRW1vb7XMy9pj78MMPsbGxYfLkya3m9FhaWppkJzIzMxk2bFiLxm9ERARSSrKysh65GjE/P5/IyEhu3brVZu6UhYUFLi4ufPTRR2g0GgICArqklq7Vak3VbAUFBaZzp7i4mNTU1EeZSqvU1dVx4cIFhg8fjre3N66urnz66accOHAAf3//Zh6i+Ph4PD09u/RbqKurw8HBgbKyMq5evdqupzQ2NhZzc/NOf5eVlRVubm58/vnnVFVVMWHCBAoKCjrsiUwscwYMDwv5BYWUOxdhaW6PuaznZsINKioqiI+PN/WvPHfuXKOtzfCeEczQ5FjSC8t4gGGO9vb2TPH14vbt2wBopeTDL77g4LFjPL1gAU+vX9/M82tjY8PQoUM5fvw4Bw4cwM/Pj5EjR6LX68nJySEvL49PP/2UtLQ0pkyZ8sgeqPLycuzs7IiIiCA2NpbKykqioqJM6zMzM4mJiWHp0qU9kqiu6Lv0mvEkpTwqhAgAvgLqgEMYPE82QOPAtg1QApQ+tLzxupY+/+/A38GQ8/QolV5tsWzZMvbt28ft27fZtGlTh1qrXLp0iUmTJjVrgNubrFixgtOnT5OZmcnw4cOpqKggODi43QuCVqvlwoULzJ8/v8UL/5EjRwgMDOyxRp0Ps2rVKr766iuSkpIICQnBy8sLKSWzZs0yPA0LgU6na9XDVlxcTHJycrP8hsbcvXuXgQMHsmHDhh7rkRYaGsrVq1e5du0a06dPb/NGYKwUGz9+fDPv0u3bt6mqqmLy5MksXLiwW8Y2b968Zo2RWyM0NJQbN24QFRWFg4MDM2fO7JRYY1RUFF5eXkybNo1Tp04xZMgQgoKCejRUYvSUjRkzxtQrzSjqeP78eQYMGMDcuWNz5ysAACAASURBVHMZNGiQSTZjzpw5na4mNfZ+XLFiRYe9aO+++y5Tp07tcuVqaGgod+/e5ezZs8TGxhIcHNyhfp7aNOCu4bWTkzMvB3597frduXJGjRrFhAkTuHfvXoufFx4eTo7Gjtfe/DbXjx7gfNJdqqQZ/2/ri0yztuPPBw+S3pAyUK3X88mpU5yOjOT1bdt4adu2Zh76VatWkZWVxYULF7h37x5+fn7Y29sTGBiIt7c3eXl5+Pv7P1JO6s2bNxk1apRJQywmJoYRI0Y02ff5+fm4ubkRGBjY603YFY+XXgvbCSHGAJ9JKd2llKOBG8BFDKG4xvGQccB1IBnwE0JoGrbXAD4Yqu56DVtbW1555RU2bNhAeHg4d+/eJT09vVVRzfz8fDQaTZ8ynMDwZLps2TICAwNNuVgWFhbo9fo223IYw4+tPTG//fbb/OAHP+ipYTfDwsKCefPmsWLFCiIjIzl69Cg1NTWMHj0aIQQVFRXs2rWrxbYc7UkTGBkxYgTLly/v0eayZmZmzJgxg3Xr1pGWlsbevXtbza9zdXXlueeeY/LkprUT1dXVZGVlUVJS0q3VR0bDqaSkhC+++KJNgUaNRoO/vz8vvvgiWq2206G89PR0Ro8ejVarJTs7G41G0+M5JkIIFi9e3KTJrLGc/sUXX8TV1ZV9+/Zx/vx50tIMXXNb0gFrj4fDj8XFxRw+fLjVnLfq6mrKyso6FKptC6O2UmVlJfv27euQLEDjG4XuoUNXWFiIs7Mzt27dajUcOGHCBMMDlpUNM55/hde/9wZrZ05i2NRZrP/97zl55Qo/nT8fdwtz5s+fj5OTE6W1tfzyf/6HwEmT2L9rV7N8THd3d9auXcv8+fOJioqiqKiI6upqRo8ezauvvvpIhlNlZSWFhYVNRH0fPHjQzEBycXFh0aJFynD6J6Q3c56mAh8LIVyFED7AW8B7wF7gTSGEoxBiBrAR+FhKmQrcBt4QQlgDPwMSpJTdo9L3CGg0GoKCgli2bBmHDh3inXfe4dKlS83ep9fr27zA9AX8/PzYuHEjmZmZfPzxx1y8eJFPP/20RTHEiooKioqK2lQNN4rKPW4ezoUyhgYABg4c2GIIKTMzk8GDB7d7IQwJCeG1117r9jG3hKOjI8899xxTpkzhs88+IyIiokVjZeTIkWg0Gmpqavj0008pKiri5s2b+Pn5sWfPHl588cVuH1tBQQGZmZlUVFS0+15ra2sWL17MypUriY2N5aOPPmpX4wpgzZo1LF26FBcXFyZPnmyq9ntcVcJJSUl8/vnnpu+zsLAgMDCQF154gcrKSg4ePIizs3On5R8ePHhAeXl5k1wunU5HTk4ORUVFLW5TUFCAubl5h0LP7WFhYcHrr7/O22+/jZubG2FhYZw+fbpVjbTGe/vhx4rCwkIsLCxwdXVtNd9o2LBh+Pr6mjy+loMG47dktWm9tasrr+3Zw29/9D28vLxYs2YNCxYswMHBgbzyct78yU+YM3UqXzyU6G4U2gwODsbW1paUlBQOHjxIdnZ2l/aLkYSEhGaNxysqKppVfRrpzpZFiv5Bb0oVaID/A9YD9cCvpJT/J4Qwa1j+AoZQ3Q+llPsatvHGoPM0CbgCbJZStpup2V1SBR2hqqqK3bt3Y2dnx8qVK7G1tTVdMNLS0qirq2tTQbivoNfruXbtmqncedmyZc1CXpcuXcLT07PVJ7ywsDD0ej0bNmx4nENvRlZWFidOnGDIkCGEhIQ0ucAnJCQwZswYLCwsuHDhAkFBQW3mnZSXlyOEaPUi2pNUVVVx7tw5MjMzCQkJaVFxu6ioiE8++YQZM2ZQXV1NYGBgj46prq7OFFLpqM6RXq83hfJ8fHyYNWtWu6G827dvo9Fo8PT05MqVK5SUlLBo0aIeFyq8fv06t27dIjQ0tMUx5uTkEBERQX19PfPmzeuQESWl5OLFi4wbN66ZF62+vr7NEJ5er0ev13drsnx9fT11dXVcunSJW7duERAQwNSpU5uEyk6mwdbPDa8XjoZ/rPx6+6SkJK5fv86aNWvaHJdWq+XWrVsMHz68RQmI+spyfv+7/6aar0PqxobpMTExJsNu3MiR/PzttwmaM8f0vtzcXC5dusScOXPIzs7m1KlT3Lt3jw0bNjBtWufUAvLz81sMPx45coQBAwawdOnSJstv3LhBeHg4W7dubfJQpqQKnmx6VSTzcfE4jScwXByvXLnClStX0Ol0TJ8+nYCAAKKioggODu72KqGepKCggC+++AJHR0cCAwM5cuSISZgyIyOj1QuTlJIXXngBHx8ffvrTnz7mUTdHq9Vy8eJFbt68ybx58/Dx8aGiooIdO3YQEBCAk5MTtra27fZIO3DgAB988AFhYWG95qrPyMjg1KlTODs7s2DBgmaGnFarJTo6msmTJ3P06FFqa2t7xPPUmLi4OJKTk3nmmWc6XEFaXV1NZGQkaWlpBAUFNWkkDQbDpK6ujiFDhnD58mXmzJmDRqPhwoULlJSU8PTTT3d7S6OHkVIipWxXuDI5OZnIyEhcXV2ZM2dOm6HF7Oxs8vPzm4VZG39eWloao0aN6vH5FRcXc/DgQebPn4+XlxelpaV89dVXZGZmMnPmTCZOnIhGo+HEHfjWUcM2i8bA7xbryKrUopNQlJmOl7N9u6HLqqoq/va3vzFr1ixmzJjR/A1Skn/tPGdOniS5rukDTH19PYmJicTFxZkqnGeMH89Pf/vbFqtIa2pq+OMf/4hWq2X8+PHMnj27Qx5wo/jqU0891eGqvZKSEtLS0hg3blyTUL4ynp5szLZv397bY+hx/v73v2//1re+9di+TwiBu7s7rq6uhIeHY2FhYeq91N80QWxtbZkwYQJ5eXmcPHkSIQRTpkwhMTGRgICAVr002dnZhIWFsWrVqm5X++0KZmZmeHh4MGzYMM6dO0dGRgZjx47F19eXYcOGcffuXfz8/Nq9WX300UcIIXj22Wcf08ibM3DgQCZOnEhxcTHh4eGYm5vj6upqMjzu37+PRqNBo9Gwfft2qqqqWL58eY+OqbS0lLKyMnx9fTtc4WRhYcHYsWNxd3cnOjqaxMREXFxcTMbg3r17+fOf/4yvry+enp4mY3XEiBF4e3uj0WgeWUW+PYQQpibJRuOo8W/+zTff5P79+yxevJhJkyZRXl5OeHg41dXVuLm5NXtQ0ul0xMbGEhAQ0OpDVHZ2NgcPHsTe3r6Jbtpnn31GVFRUt2qmWVpakpeXx4gRI3BwcMDKygpvb2+GDx9uap5sZ2dHEY4cTTHsZ8/B4ONaxxcZFaSU1ZGTk8OM0W7t5gBaWFjg6+vL2LFjWz5mQmA7bBQTZs/Dk3JK7t+jVBr2kUajwc3NzRRKKyoqIiM3l/379xMXFYXz0KFoNBqT58fc3JzZs2cTGBhIZWUlp06doqCgAGdn5zbHmZaWhrW1dbO8MqOyu4WFRbPft7W1NcOGDWt2LfzFL36Rs3379r+3uVMU/RbV264H8fDw4D//8z+xsLDgyJEj5Ofnc//+/d4eVqcxNzdn3rx5rF69GjMzM44dO2ZKEG2tG31MTAxAq0/XvcWwYcPYtGkTjo6O7N69m/z8fJKSkvDx8eHAgQOcP3++1W31ej03b95s1vS4NzDeHJ5//nmSk5PZv38/+fn5aLVa0tLS8PHxYeDAgabmxT2Nr68vK1euxMzMDK1W26QBbHsMHTqUDRs2MGHCBD777DNOnjxJdXU1aWlpuLm5odPpmhgRRoOmurqagwcPkvmQuGJPUFJSQlxcXBM9J2MTbCMWFhbMmDGDF154gerqanbu3Mn169eb6AClpqYyYsSINqtY3d3dWbVqVTMdscuXL3Pjxo1unJXBKFm2bFmT3CsANzc3nnvuOebNm0dUVBRfXWgq22HR2EOYX9BhCYzBgwe3b+xqNAyfv5IXfvBzNgWMYaj4WqR0wIABPPXUUzwTGmpadiY6mg3r1/Ps8uXNpF+MPQxffvllBg0axP79+00tqh6mtraWjIyMJoUCRmpqatiwYQOffvppi0PWarVkZGQ8tlw8Re+jjKcexsrKihEjRrBgwQL+8pe/tKsO3JcZOXIk69atIzs7m4iICI4dO2bSGXqYuLg4XFxcuiW5tbsxNzcnODiY1atXExERwYULF7C0tGTkyJFtjjcjI4PKyspuFcd8VJycnHj++eeZNGkSn3zyCfv27WPEiBFYWFhw//59k+dQr9eTkJDQo789IQRSSg4dOsThw4c79V1CCCZOnMiWLVvQaDTs3LmTGzduYGlp2eb+rqmp6ZamxO3h6urK1q1bm+Qr5ubmUl9f3yzPyc7OjiVLlvDMM8+QlJTE7t27uXv3LtXV1eTk5HSob56npycajaZJhVlOTk6PNRCvq6sjPj6+WTK2UWx2dKNwdk1tLRaNHLTmlladaisTFxfH4cOH232fGGDJ2JWb+eabP+Y5b2ecGmkqT7fUMa2R504C9wsLmRMUxJuvv05OTg56vZ733nuPixcvYmlpyaxZs3jppZewsrLiww8/5OzZsyapCTDkbnl7e7foETQWRbQWqk9KSuLjjz/udDNjRf9FGU89THZ2Nvb29syZM4ef//zn2NramlqiPI6LfneTlpbGxo0bWbx4scn9rdPpqKysbPKEbW9vT3BwcJ/uPD5kyBD8/Px46qmn2LdvHwMHDjQlYaempjarMExISADoU8YTfG14rFmzhsLCQs6cOUNqaqqpwtDHx8fUiqct2YnuGsu4ceOa5S91FGtra0JCQliwYAH5+fmUl5dTVvawNu7X7924cWOTRtA9iTGcmJ+fT0FBgen30Vquj6urK2vXriUoKIjTp0/zl7/8BTc3tw6HNY09Be/cuUN9fT0FBQWPLFPQGqmpqZw8ebLFqlqNRsOoRj0Xs7Jzib16xfR/M0urTvUANCa9d9Q7KWzsGbf+O/zLttdZNdya4Zpanv3337Dv3DneCQlhYsP+dHV1RScl+z/5hJkzZvDzH/8YT0/PJtp71tbWzJkzx5QDuGvXLs6fP09ubi4VFRWtPjwZq+laM57GjBnD6tWrH0uDd0XfQCWM9yD19fVERkYya9Ysk5u+rq6OM2fOEBERgZubG6+99lq/UaYtLS0lMTGRWbNmmTSTwsPDqaiooKKiwhRu6C8Yb0o+Pj4UFhZy4sQJLCwsCAkJ4ejRo1hbW/Pcc8+ZjIDc3FxiY2NZunRpnzQKo6Oj8fLyMuV4pKWlUVhYyF//+lfAkAvV2ZL6R6WwsJBBgwZ1ukgiOjqaN998kzfeeIOCggJGjhxJcHBwq1WO6enpxMTEsGrVqh4tyNDpdOzYsYNBgwZhZmbGe++9x0cffdTuTTM/P59jx45RWVmJt7c3s2bNajchWafTcfToUaZOnYoQgq1bt/Kv//qvPSI4q9frTYKPLXEhAzZ+Zng9Y5iOZWZfkjN6JgA1hblsX/QYHyhqqsDq631XGh/P397YRk1wCIWFhVy9epWMjAwArCws2Pryy/zL977XojzJgwcPiI6O5uTJkyxevJjg4OAWqyrj4+P54Q9/yG9+85sOpyKohPEnG5Uw3oOkpKQwePDgJi5tMzMzPD09GTBgAHFxcQwZMgQ3N7c+eTNujJSSmJgYU48zMOQf+Pr6IoTg6tWruLu74+3tTW1tbZ+vKKytrSUhIYGAgAA0Gg02NjZMmDCB6upqwsPDGTduHEFBQVhaWpqekB0cHPDy8uqTxyovL4/KykrGjh1rSii3tbU1JVS7ubmZihVKSkoIDw/Hw8Ojy42TO4JRtsMoXNgZsrOzmTRpEkuXLiUgIID8/HxOnjwJ0KL3Jj8/n7t37+Lt7d2j/SI1Gg3u7u5MmjSJBw8eYGZmRkhISJvbGM+dkJAQpkyZQlZWFqdOnQIM3pLWPFEajQZfX18cHBzIzc0lLi6OJUuWdFjdvTM0lt9oKQm/tAb2GxyvDBwAi90KyLFsuK6ZmWGZnYyri0unqgONnvdOVxSaN/3NWrq4EH87gSozK2xsbPD09GT48OGUl5dTUlbG5WvX2PX++9TX1jJp8uQmvw9LS0usrKxwdXWlpqaGs2fPotfrcXloLqmpqZw/f55nnnmmVS/bgwcPuHnzpql4QyWMP9koz1MPUVVVxZUrVwgODm714lhSUsKXX36JTqdj4MCBfbo/UlZWFoWFha0mH5eUlHD8+HE0Gg3Hjh3DysqKv/zlL93W3by7aSsny2hc1NfXs2TJEi5fvkxubi4eHh5Mnz69z6kJ6/V6zp8/z4wZM5pVEpWUlHD69GkqKytZuHAh7u7upKWlER4eztq1azsVbukKiYmJjBw5slP7rKKigtjYWIKCgprcxIuLi4mIiKC0tJT58+c3M8g6qjPVXRgrsAYPHtzm++7du0dFRUUTcdzi4mIiIyPJz88nODgYHx+fVo1yY4Wep6dnj1frxsTEkJqa2sTjCnC3FOZ+YHjtOKCO46tL2ZOrQdsQqffKvkrh/XvMnDmTCRMmtHsciouL2b17NyEhIUycOPGRxqyrKufU//6ca1o7tKKpIZaVlcWVK1dMgqyDrK3Z9t3v8tKrr2JlZdUsOlBcXExUVBT3799n2rRp+Pv7Y2FhQWZmJl999RVPP/10q17GpKQkvvjiCzZu3Gh8IFaepycY5XnqIa5fv46np2ebYorW1taMHz+e+Ph4jh07xvjx4zuVePm4qK+vb7e82jiXBw8e8N577+Ho6EhoaGifNJ5KS0vJyspqpiBsxDgXKSXHjx/H3Nyc8vJydu7cyZw5c3rc4Ogsd+7cwdbWtknIJS4ujrfeeovp06cze/ZsrKysCA8Pp7i4mPHjxzNt2jTTb7OmpqbHPIUuLi5YWloipeT27ds4OTm167mLiYkhLy+PAQMGNNnX1tbW+Pn54eDgwNmzZ8nIyMDV1dVkMBpFXE+fPk19fX2PK9ufPn2aiIgIxo8f36q3S6vVcv36dQICApp4MqytrfH19cXFxYWoqCgSExNxdHRs8cZcVVXFkSNHsLS07FIbmM5QVFREZWUlo0aNavKbsBsAf7tmaM1SrTPjtZm25NdoKaszWE/T/LyY5uVhkjewtbXF0dGx1WNtZWWFlJLRo0d3WE+pNTQWlnjODiHAqhp95m1ydRbIhu91cHDA19cXZ2dnSktLKXnwgPNffcW+99/HrLYWq0GDcHZ2Nl13ra2t8fb2xsPDg5s3b3L+/HnMzMwYO3YskyZNavPh1t7enokTJ5p+d8rz9GTz2O5sQohIIURgw2shhPi9EKJMCJEthPh2o/eNFkJcEEJUCyEuCSH8Gq3bLIS4L4R4IIT4qxCiT8aGioqK0Ol0HXKvm5mZsWXLFn7yk59w7tw5zp8/T3p6ep+qyEtNTcXDw6Ndr5hGo8HS0hJXV1e8vLz4/PPPKS4ubjEJtbeQUpKQkNBu/zohBP7+/mzevBkLCwsuX76MmZkZNjY2psTxvkBtbS2ZmZnNyquTk5PJy8tjyJAhCCHw9fXlxRdfxMzMjA8++IDk5GTTvti1axelpaU9Os579+5x9OhRkpKS2nxfXl4eQgh27NhBZGRki+8ZO3YsW7ZsYejQoezfv5/IyEhTyxopJUVFRa22OOkuKisrefvtt9HpdG3e/JOTkxk7dmyr4dERI0awadMm/P39OXr0KJ9//nmzY2FnZ4dOp+Py5cvdOoeWmDBhAsuXL292rptrwM/562vSry7ACOuvz58zWZXU2jmZ5A2io6PZt2+fKffoYYQQzJo165H6zxkpKSnhWtx1NJMXsPTffs13500mwLwc0ega6uHhwZo1a1i4cCFmZmYUVlXxi3feYeOKFZz461+peai9ypAhQ1i1ahWhoaGkp6fzpz/9iYiIiGb99RpjaWnZ73T8FF2nx40nIcQSIcS7QFCjxS8DCzE0/V0C/KcQIqBh3W7gHOAKHAT2N3yON4a2LWsBT2AisK2nx99ZGjeX7QxeXl688MILJCQksH37dmJjY3tohJ2jqqqKvLy8dpW3jcTGxmJtbc2bb76Js7Mzv/71r3n33XeblAT3JllZWdjb27eYPNoS9vb2hIaGUldXh5mZmakhbFvNcB8nt27dwsfHp1neSEpKCkOHDm3iybCysiIkJITQ0FDi4uI4cOAAAwYMYMyYMT0eivTw8CA0NLTN1kTG3o8ODg7odLomVVIPY25uzowZM9i8ebPJK3jz5k3MzMx47rnnTD3weoqsrCzMzc2ZNm0aQogmlaZGysvLKSkpaTdJXwjB+PHjefnllxkyZAj79u0jIiKiSTVuamoqhYWFVFVVtdp0vDspLy9vZoB+Y1ih6fUnt+BSipbaMkNpvrlGYG2maSJvMHXqVE6ePMnBgwdbbWpdXFxMSkrKI401MTGRn/70p4Z+dhYDGDjvGVb+4FdsCx7PRLMKaGRE+Tg5YNvIACrSavnV/v0836jVS2Pc3NxYvXo1VVVVbN++nZ07d5KQkNDi8QZDR4YzZ848lmOk6F0eh+cpCDAHGncP3QT8l5QyS0p5A4OB9LwQYiQQAPxSSvkA+B9glBDCF0MPvE+klFFSynzgf4F1j2H8neLevXs4Oztja2vb6W1tbGx49dVXWbFihamVSGsn6eMiMTERPz+/Doff4uLi8PPzw9bWlqCgILZt24adnZ3pZtCb86mvryclJQU/P7/239yIiooKSkpKeOaZZ/D19TU1cJVSUlNT0/4H9BClpaVUVVW1WL6enJyMl5dXi9u5ubmxYcMGvL29OXXqFFZWVuh0Ourq6nrMSyiEMClLV1ZWEhcX1+w9aWlpDBs2zNTUtSMGu729PU8//TTLly8nJiaGsLAwCgoKAMON+ejRo50S7OwoRmHO4cOHk5eXx44dO5oJRSYmJnZKssHYdHjLli1otVp27drFtWvXqK+vJzc3l8GDB7Njxw56On9TSslHH33EmTNnTMu0Wi3e9Qk86/v1+RuXA3Hv/Qp7tKzzHIirzdeBgMbeTi8vLw4dOmTyRDfGWOnWlkenPYyCl028PhaWOIY8x+q3fs1rIU/hZW44T9eGLid87142+/nR+ArdWHSzJYy/36VLl5KYmGgy1h++npWVlXHjxg2l9/RPQI+HvaSUPwMQQixstHgS0Ni1kojBE+UP3JJS1jZsqxdCJGHwNE0Czjy0TXMp2F6krq6O9PR0goKC2n9zK5iZmfHss8+Sn5/P4cOH2blzJy+++GKTZNPHRWFhoanypKOsXr26ieE4atQo3njjDS5cuMAf//hHAF555ZVucdd3lpSUFDw8PDpdjZWSkoKUkoCAAPz9/U0J1zU1Naawa1eM5UfBGHKbNGlSs5tzSUkJhYWF+Pj4tLq9RqMhICAAb29vIiIi2LVrF9bW1pSUlLB169ZHzkNpi7i4OK5evcro0aNNHsCamhru379PcHAwkZGRmJubN1O9bgt3d3c2btxIYmIihw4dYsyYMQwbNoz79+9TVlbW7flPxhY4Q4cOpb6+HgcHhybHITc3FwsLiy7lx9na2rJo0SICAgI4f/480dHRFBcXm9oLtWYUdxdCCBYvXtzEO5uUlISn51h+NUzDnVL4hieMKrzGO8X5rHYTuNm0fCsxMzPD398fPz8/YmNjCQsLw9PTk5kzZ2Jvb8/s2bOZO3fuI/XwM+qAtZjIbW6BS/Ay1jwVQsnt67hNnI4EntZo+I6zM+/96ldExcSw6a232vyOqqoqbG1tGT58OM8//zwZGRlERUVx6dIlZs6caUr4Hz16NNu2bevxnoSK3uexVdsJIe4C66SU0UIILTBaSnm/Yd1GYAvwAfCSlHJho+1OAjuBl4APpJR7Gpa7A+lSyhbvhEKIbwHfAnB1dZ0aFhbWY3MzUl1djZmZWbeVSldUVPDZZ59hY2PDvHnzHrvBUVFRgY2NTbclfaempnL8+HGmTZvGlClTerSk/GH0ej1VVVVtJvC3RWlpKXZ2dqYk2vr6ei5evEhSUhJLlixhxIgRj1XCQKvVUl9f32KfrrKyMiIiInjqqac6rPCen5/P1atXkVIyb968HjUG9Xo9FRUVTW521dXVmJubY2FhwQcffEB5eTnbtnUtKl9XV0dSUhL37t3D09MTHx+fbi9cCAsLIzs7mzfeeKPF9RUVFdja2nbLbyI+Pp7333+fmTNnsnjx4sd+HXj43NFJgZmQnD17ltOnT7N9+/YOFxzU1dWRnJzM3bt38fDwwMfH55ErjI8ePUpsbCw/+9nPOjwGvV5vklzpSH/Ed999FysrqyYNtqWU5Ofnc/PmTerr6/Hz88Pd3d30WfPnz1fVdk8wvZVwXQo0vurbACUtLG9rnXF5i0gp/w78HQxSBfPmzXvkQbdFeXk5cXFxzcqrH5Vly5Zx584dTp48SW1tLUuWLHksF8/09HSqq6s71cft+vXrODs7t+oxmDdvHps3b+b8+fOkpaUxdOhQlixZ8ljkGa5cucLEiRO7tZpx4cKFZGdnEx4eTl5eHm5ubsydO7fHjShjefXcuXNbNUC7Ila6Zs0arl27xtWrVxk+fDgeHh493hcvJSWF0tJSBgwYwMyZMxFCMHPmTEpLS5v0s+ssixcvpri4mHPnzpGeno6Liwt+fn4t9i3rCkIIysvLaXxd0el0REdHU19fj7e3d7c1xB46dCiZmZl84xvfMOUHmZubM3Xq1B6tvqusrOTcuXNIKZk1a1YzL1pCQgLDhw9n4cKFrXxCyyxevJiKigouXrxISkoKXl5elJeXM3fu3C5d2y5evIiHhwftXePz8/M5d+4cNjY2LF68uFMaZ3v27GHs2LEtfoeUkvT0dKKiosjIyGDMmDEd7vWn6L/0Vh15EtA4o3occB1IBvyEEBqAhn99gPg2tmmXx+Fd60gFV1cQQuDp6cnatWuJjIxkz5493fr5LaHVak2Cg53h7Nmz7N27t833WFpasmjRIvz9/QkLCyMqKupRhtohCgsLkVJ2yXCSwDNhnAAACBdJREFUUnLs2LFWk1qHDRvG5s2bqa6u5v333zfl2/QkXQ0/toeZmRnTp09n48aNXLt2jV27dvVo4quUktjYWI4fP94kN8hYsfmoODo68swzzzBnzhyOHTvGiRMnHvkzjcydO5fly5c3WWZsHB0fH29q89Md+Pj48N///d8sWLCAl19+GTc3Nz777LMmTYl7AnNzc27evEl1dXWL4cfvfOc7vP/++136bDs7OxYtWsT69espLy/niy++6HKe0MaNG1v1ADbG3NycW7duMWTIkE6Lw27evJmlS5e2uE4IwZgxY9i4cSNBQUHExMT0eF6aovfprbDdt4HNwArAC/gcmC2lTBFCRGOosvsz8ANgvpRynhBiPHAWmIfBC3UE+JOUclcHvrsAuNfe+7oBZ6Cw3Xf1D56kuYCaT1/nSZrPkzQXUPPpKh5Syr4n3KfoFnorbPceBqmBexgMoe9LKY2P9i8AHwL/AVzBYGQhpUwUQvwEOI0hfPd3DDlS7fK4fsBCiKtPSoz7SZoLqPn0dZ6k+TxJcwE1H4WiJR6b8SSlHNXotQ74TsPfw++7Dcxo5TPew2B4KRQKhUKhUPQKfa93hkKhUCgUCkUfRhlP3cuT1MfoSZoLqPn0dZ6k+TxJcwE1H4WiGY8tYVyhUCgUCoXiSUB5nhQKhUKhUCg6gTKeugkhxHghRK0Qwq3Rsh8IIQob/n7Rm+PrKEKIbUKIHCHEAyHEUSHE0IblQgjxeyFEmRAiu0Fuok8jhNgihLgrhKgQQpwTQvg0LO93c2mMECJSCBHY6P/9dj4NjcNThBCVQogDQoie7VDcQzQ+Jv38eDQ7Z/rrfFq6lvXXuSj6IFJK9feIf4AZEAVIwK1hWQhwH/ADPIBUILS3x9rOPCYARRiqHe0wNGz+sGHdKxjESt0xyEwUAgG9PeY25uKFoRn1NzBIW/wIiO2Pc2k0pyXAuw2/s8BGy/vrfAYDxUBow+uPgXd6e1yPekz68fFo8Zzpj/Np7VrWH+ei/vrmn/I8dQ9vANEPLdsEvCulvCWlvAf8FVj32EfWOeYAR6WUl6SUFcBewNiReBPwX1LKLCnlDQwXo+d7aZwdYQFwVkp5TEpZDfwO8BdCONL/5mIkCIO8SMVDy/vrfFYCMVLKQ1LKEuC39P1z5GFaOib99Xi0eM7QP+fT2rWsP85F0QdRxtMjIoTwxtC0+OGulJMwPLUZSQS6p7FWz/Ee8E0AIYQTsBGDRw3633wOAd9t9P+pQBXwgP43FwCklD+TUm7F8ETdmH45H5qPOwFwFUJ0rXtzL9DKMemvx6O1c6Y/zqe1a1l/nIuiD6KMp0egoffeP4DvSSkrH1rtAJQ1+n95w7I+i5RSK6WsE0J8E4M7OxQIa1jdr+YjpcyTUt4FEEKswdAC6C0pZT39bC4doL/Op8m4pZS1gPH49Gf65fFo7ZyhH86njWtZv5uLom+ijKcOIIT4sRCi/uE/QAvckVKebGGzUgx5A0ZsgJLHMd72aG0+Qoh/gEnJ3Qb4IXBICGFBH51PW3MRQjgLIQ4DfwBekVL+qWGzPjkXaP/YtEKfnU87NBm3EGIAhhBYfxh7W/TX40Er50y/nc/D1zIM4dV+ORdF30IZTx1ASvlrKaX5w38YTsb1QogaIURNw9vvCiFWAEkYkhaNjAOuP+aht0gb80kTQmxqeE81hjwtRwxPZn1yPm3M5XXgAoZwiq+U8kijzfrkXKD1+UgpX2ljsz47n3Z4eNx+QErDb68/0y+PhxDClpbPmX43HyHET1q5lt2nn81F0TfprcbATwRSyjWN/y+EkMAoKWWuEEIL/FEIcQBDtcf3MeRG9WUKgLeEEFFANvCvwA0pZZEQYi/wphDiHIaqnI3A7N4bartsBHKklC+3sK6/zaU9+ut8DgH/LYQIxpDv9GtgX+8OqVvor8ejtXOmP86nxWsZBiOqv81F0QdRxlMPIaU8LoTYjeGErQd+K6WM6N1RtcsODE//URhc21GA0UB8D0Np7z0MbvzvSylTemOQHWQSMLchvNqYsfS/ubRHv5xPg1G+BdgNDMEgVfCb3h1Vt9AvjwdtnzP9bT6tXcvS6X9zUfRBVHsWhUKhUCgUik6gcp4UCoVCoVAoOoEynhQKhUKhUCg6gTKeFAqFQqFQKDqBMp4UCoVCoVAoOoEynhQKhUKhUCg6gTKeFAqFQqFQKDqBMp4UCoVCoVAoOoEynhSKf3KEEBohxD4hRJEQ4v81LDNr6LOXKoSoFkKkCCF+1tB/TqFQKP6pUcaTQqFYBNwCRgDzhRB2wJ+BV4CtwGBgGbAY+N/eGqRCoVD0FZTxpFAoBGCGoV2TwNCO41vABillhJSyRkp5G9gCTBBCWPTeUBUKhaL3Ue1ZFIp/coQQGmA/sBD4JaAF3pJSjuzVgSkUCkUfRTUGVij+yZFS6oHnjf8XQvwEQyd6hUKhULSACtspFIqHKQKcW1ohhNgghHB4zONRKBSKPoUynhQKxcOcA8YIISY2XiiEmAH8DajulVEpFApFH0EZTwqFoglSylvAHiBMCBEohLBuMJz2AL+QUmp7d4QKhULRu6iEcYVC0YyGirp/A17CIGFwD/ijlPJPvTowhUKh6AMo40mhUCgUCoWiE6iwnUKhUCgUCkUnUMaTQqFQKBQKRSdQxpNCoVAoFApFJ1DGk0KhUCgUCkUnUMaTQqFQKBQKRSdQxpNCoVAoFApFJ1DGk0KhUCgUCkUnUMaTQqFQKBQKRSdQxpNCoVAoFApFJ/j/1n1EvvnZJ0IAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 468x468 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fontsize=12\n",
    "hour = 18\n",
    "\n",
    "skewt_plot = metpy.plots.SkewT()\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_1000_latlon_mean[hour,:]-273.15, lw=3, label='phys. Tenv', c='firebrick')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_400_latlon_mean[hour,:]-273.15, lw=3, label='cont. Tenv', c='coral', ls='--')\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_1000_latlon_mean[hour,:]-273.15, lw=3, label='phys. Td', c='dodgerblue')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_400_latlon_mean[hour,:]-273.15, lw=3, label='cont. Td', c='skyblue', ls='--')\n",
    "\n",
    "pprofile_400 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_400_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_400_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "pprofile_1000 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_1000_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_1000_latlon_mean[hour,:][0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_1000, label='phys. profile', c='0.1', lw=3, ls='-')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_400, label='cont. profile', c='0.5', lw=3, ls='--')\n",
    "\n",
    "skewt_plot.ax.set_ylim(1013.25,100)\n",
    "skewt_plot.ax.set_xlim(-45,35)\n",
    "\n",
    "fig = mp.gcf()\n",
    "fig.set_size_inches(6.5,6.5)\n",
    "fig.tight_layout()\n",
    "\n",
    "skewt_plot.ax.tick_params(labelsize=fontsize)\n",
    "skewt_plot.ax.set_ylabel('hPa', fontsize=fontsize)\n",
    "skewt_plot.ax.set_xlabel('$\\degree$C', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.plot_dry_adiabats(color='0.1',linewidth=1,linestyle='-')\n",
    "skewt_plot.plot_moist_adiabats(color='0.1',linestyle=':')\n",
    "skewt_plot.plot_mixing_lines(color='0.1',linestyle='--')\n",
    "\n",
    "handles, labels = skewt_plot.ax.get_legend_handles_labels()\n",
    "skewt_plot.ax.legend(handles, labels, fontsize=fontsize, scatterpoints=1, \\\n",
    "                     bbox_to_anchor=(1.03,1.0), \\\n",
    "                     bbox_transform=skewt_plot.ax.transAxes, loc='upper left', \\\n",
    "                     borderaxespad=0, labelspacing=0.25, fancybox=False, \\\n",
    "                     edgecolor='0', framealpha=0, borderpad=0.25, \\\n",
    "                     columnspacing=0.5, handletextpad=0.5, markerscale=1)\n",
    "\n",
    "fig.savefig('figures/both_cape_skewT_hour'+str(hour-4)+'_amazon_region.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 10 day mean skewT diagrams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T15:01:21.791315Z",
     "start_time": "2018-08-16T15:01:21.777483Z"
    }
   },
   "outputs": [],
   "source": [
    "time_datetimes = pandas.to_datetime([i.decode('utf-8') for i in time], format = '%Y-%m-%d_%H:%M:%S')\n",
    "time_datetimes_andes = time_datetimes+datetime.timedelta(hours=-5)\n",
    "time_datetimes_amazon = time_datetimes+datetime.timedelta(hours=-4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T15:01:23.568952Z",
     "start_time": "2018-08-16T15:01:23.560176Z"
    }
   },
   "outputs": [],
   "source": [
    "hour_indices = [t.hour in [15] and t.day in range(2,11) for t in time_datetimes_amazon]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T15:01:32.070236Z",
     "start_time": "2018-08-16T15:01:31.330007Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(<Quantity(1952.832754099332, 'joule / kilogram')>, <Quantity(-9.152163586034494, 'joule / kilogram')>)\n",
      "(<Quantity(1609.3138224378679, 'joule / kilogram')>, <Quantity(-9.535432253950521, 'joule / kilogram')>)\n"
     ]
    }
   ],
   "source": [
    "ptot_lo = 100\n",
    "ptot_hi = 1013.25\n",
    "\n",
    "tk_400_latlon_afternoon = numpy.nanmean(tk_400_latlon_mean[hour_indices,:], axis=0)\n",
    "td_400_latlon_afternoon = numpy.nanmean(td_400_latlon_mean[hour_indices,:], axis=0)\n",
    "\n",
    "tk_1000_latlon_afternoon = numpy.nanmean(tk_1000_latlon_mean[hour_indices,:], axis=0)\n",
    "td_1000_latlon_afternoon = numpy.nanmean(td_1000_latlon_mean[hour_indices,:], axis=0)\n",
    "\n",
    "ptot_indices = (ptot_latlon_subset_timemean/100.<=ptot_hi)&(ptot_latlon_subset_timemean/100.>=ptot_lo)\n",
    "\n",
    "cape_400 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_400_latlon_afternoon[ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_400_latlon_afternoon[ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "cape_1000 = metpy.calc.surface_based_cape_cin(ptot_latlon_subset_timemean[ptot_indices]/100.*metpy.units.units.hPa, \\\n",
    "                                  (tk_1000_latlon_afternoon[ptot_indices]-273.15)*metpy.units.units.degC, \\\n",
    "                                  (td_1000_latlon_afternoon[ptot_indices]-273.15)*metpy.units.units.degC)\n",
    "\n",
    "print(cape_400)\n",
    "print(cape_1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T15:17:31.841652Z",
     "start_time": "2018-08-16T15:17:27.271746Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 320.4x306 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fontsize=12\n",
    "hour = 19\n",
    "\n",
    "skewt_plot = metpy.plots.SkewT()\n",
    "\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_400_latlon_afternoon-273.15, lw=2, label='cont. T$_{env}$', ls='-', c=color_list[1])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_1000_latlon_afternoon-273.15, lw=2, label='phys. T$_{env}$', ls='--', c=color_list[0])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_400_latlon_afternoon-273.15, lw=2, label='cont. T$_d$', ls='-', c=color_list[3])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_1000_latlon_afternoon-273.15, lw=2, label='phys. T$_d$', ls='--', c=color_list[2])\n",
    "\n",
    "pprofile_400 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_400_latlon_afternoon[0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_400_latlon_afternoon[0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "pprofile_1000 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_1000_latlon_afternoon[0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_1000_latlon_afternoon[0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_400, label='cont. profile', c='0.45', lw=2, ls='-')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_1000, label='phys. profile', c='0.1', lw=2, ls='--')\n",
    "\n",
    "\n",
    "\n",
    "# fill in control and phys CAPE\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_400_latlon_afternoon-273.15,\\\n",
    "                      numpy.array(pprofile_400),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.333,\\\n",
    "                      label='cont. CAPE')\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_1000_latlon_afternoon-273.15,\\\n",
    "                      numpy.array(pprofile_1000),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.666,\\\n",
    "                      label='phys. CAPE')\n",
    "\n",
    "skewt_plot.ax.set_ylim(1013.25,600)\n",
    "skewt_plot.ax.set_xlim(10,30)\n",
    "\n",
    "fig = mp.gcf()\n",
    "fig.set_size_inches(4.45,4.25)\n",
    "fig.tight_layout()\n",
    "\n",
    "skewt_plot.ax.tick_params(labelsize=fontsize)\n",
    "skewt_plot.ax.set_ylabel('hPa', fontsize=fontsize)\n",
    "skewt_plot.ax.set_xlabel('$\\degree$C', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.plot_dry_adiabats(color='0.5',linewidth=1,linestyle='-')\n",
    "skewt_plot.plot_moist_adiabats(color='0.5',linestyle=':')\n",
    "skewt_plot.plot_mixing_lines(color='0.5',linestyle='--',lw=1)\n",
    "\n",
    "handles, labels = skewt_plot.ax.get_legend_handles_labels()\n",
    "skewt_plot.ax.legend(handles, labels, fontsize=fontsize, scatterpoints=1, \\\n",
    "                     bbox_to_anchor=(0.625,0.99), \\\n",
    "                     bbox_transform=skewt_plot.ax.transAxes, loc='upper left', \\\n",
    "                     borderaxespad=0, labelspacing=0.2, fancybox=False, \\\n",
    "                     edgecolor='0', framealpha=0, borderpad=0.25, \\\n",
    "                     columnspacing=0.5, handletextpad=0.5, markerscale=1, handlelength=1.5)\n",
    "\n",
    "skewt_plot.ax.text(s='skew-T diagram at 15:00 on days 2–10',\\\n",
    "                  x=0,y=1.02,\\\n",
    "                  transform=skewt_plot.ax.transAxes,\\\n",
    "                  ha='left',va='bottom', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.ax.annotate(\"\",xy=(11.8,700), \\\n",
    "                       xycoords='data',\\\n",
    "                       xytext=(13.8, 700), \\\n",
    "                       textcoords='data',\\\n",
    "                       arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=9))\n",
    "\n",
    "skewt_plot.ax.annotate(\"\",xy=(11.1,700), \\\n",
    "                       xycoords='data',\\\n",
    "                       xytext=(9.1, 700), \\\n",
    "                       textcoords='data',\\\n",
    "                       arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=9))\n",
    "\n",
    "CAPE_CIN_text = 'CAPE$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CAPE$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[1]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[1]))+' J kg$^{\\,-1}$'\n",
    "skewt_plot.ax.text(s=CAPE_CIN_text,x=0.02,y=0.02,ha='left',va='bottom',\\\n",
    "                   transform=skewt_plot.ax.transAxes,fontsize=fontsize,linespacing=0.9)\n",
    "\n",
    "fig.savefig('figures/manuscript_both_cape_skewT_hour'+str(hour-4)+'_zoom_ALL_DAYS_amazon_region.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:59:59.457105Z",
     "start_time": "2018-08-16T14:59:55.013884Z"
    }
   },
   "source": [
    "fontsize=12\n",
    "\n",
    "skewt_plot = metpy.plots.SkewT()\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_1000_latlon_afternoon-273.15, lw=3, label='phys. T$_{env}$', c=color_list[0])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., tk_400_latlon_afternoon-273.15, lw=3, label='cont. T$_{env}$', ls='--', c=color_list[1])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_1000_latlon_afternoon-273.15, lw=3, label='phys. T$_d$', c=color_list[2])\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., td_400_latlon_afternoon-273.15, lw=3, label='cont. T$_d$', ls='--', c=color_list[3])\n",
    "\n",
    "pprofile_400 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_400_latlon_afternoon[0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_400_latlon_afternoon[0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "pprofile_1000 = metpy.calc.parcel_profile(ptot_latlon_subset_timemean/100.*metpy.units.units.hPa, \\\n",
    "                                         (tk_1000_latlon_afternoon[0]-273.15)*metpy.units.units.degC, \\\n",
    "                                         (td_1000_latlon_afternoon[0]-273.15)*metpy.units.units.degC).to('degC')\n",
    "\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_1000, label='phys. profile', c='0.1', lw=3, ls='-')\n",
    "skewt_plot.plot(ptot_latlon_subset_timemean/100., pprofile_400, label='cont. profile', c='0.5', lw=3, ls='--')\n",
    "\n",
    "\n",
    "# fill in control and phys CAPE\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_1000_latlon_afternoon-273.15,\\\n",
    "                      numpy.array(pprofile_1000),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.666,\\\n",
    "                      label='phys. CAPE')\n",
    "\n",
    "skewt_plot.shade_cape(ptot_latlon_subset_timemean/100.,\\\n",
    "                      tk_400_latlon_afternoon-273.15,\\\n",
    "                      numpy.array(pprofile_400),\\\n",
    "                      edgecolor='None',facecolor=matplotlib.cm.tab20c(8/20),alpha=0.333,\\\n",
    "                      label='cont. CAPE')\n",
    "\n",
    "skewt_plot.ax.set_ylim(1013.25,500.)\n",
    "skewt_plot.ax.set_xlim(10,30)\n",
    "\n",
    "fig = mp.gcf()\n",
    "fig.set_size_inches(4.45,4.25)\n",
    "fig.tight_layout()\n",
    "\n",
    "skewt_plot.ax.tick_params(labelsize=fontsize)\n",
    "skewt_plot.ax.set_ylabel('hPa', fontsize=fontsize)\n",
    "skewt_plot.ax.set_xlabel('$\\degree$C', fontsize=fontsize)\n",
    "\n",
    "skewt_plot.plot_dry_adiabats(color='0.5',linewidth=1,linestyle='-')\n",
    "skewt_plot.plot_moist_adiabats(color='0.5',linestyle=':')\n",
    "skewt_plot.plot_mixing_lines(color='0.5',linestyle='--',lw=1)\n",
    "\n",
    "handles, labels = skewt_plot.ax.get_legend_handles_labels()\n",
    "skewt_plot.ax.legend(handles, labels, fontsize=fontsize, scatterpoints=1, \\\n",
    "                     bbox_to_anchor=(0.6,1.0), \\\n",
    "                     bbox_transform=skewt_plot.ax.transAxes, loc='upper left', \\\n",
    "                     borderaxespad=0, labelspacing=0.25, fancybox=False, \\\n",
    "                     edgecolor='0', framealpha=0, borderpad=0.25, \\\n",
    "                     columnspacing=0.5, handletextpad=0.5, markerscale=1)\n",
    "\n",
    "skewt_plot.ax.text(s='skew-T diagram at 15:00 on days 2–10',\\\n",
    "                  x=0,y=1.02,\\\n",
    "                  transform=skewt_plot.ax.transAxes,\\\n",
    "                  ha='left',va='bottom', fontsize=fontsize)\n",
    "\n",
    "# skewt_plot.ax.annotate(\"\",xy=(11.6,700), \\\n",
    "#                        xycoords='data',\\\n",
    "#                        xytext=(13.6, 700), \\\n",
    "#                        textcoords='data',\\\n",
    "#                        arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=10))\n",
    "\n",
    "# skewt_plot.ax.annotate(\"\",xy=(10.9,700), \\\n",
    "#                        xycoords='data',\\\n",
    "#                        xytext=(8.9, 700), \\\n",
    "#                        textcoords='data',\\\n",
    "#                        arrowprops=dict(facecolor='0.1',ec='None',width=2,headwidth=6,headlength=10))\n",
    "\n",
    "CAPE_CIN_text = 'CAPE$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CAPE$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[0]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,cont}$ = '+'{:.0f}'.format(numpy.array(cape_400[1]))+' J kg$^{\\,-1}$\\n'\\\n",
    "                'CIN$_{\\,phys}$ = '+'{:.0f}'.format(numpy.array(cape_1000[1]))+' J kg$^{\\,-1}$'\n",
    "skewt_plot.ax.text(s=CAPE_CIN_text,x=0.02,y=0.02,ha='left',va='bottom',\\\n",
    "                   transform=skewt_plot.ax.transAxes,fontsize=fontsize,linespacing=1)\n",
    "\n",
    "fig.savefig('manuscript_both_cape_skewT_hour'+str(hour)+'_zoom_ALL_DAYS_amazon_region.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot cape variables in 4x4 plot for first day and 2-10 days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.553544Z",
     "start_time": "2018-08-16T14:22:29.567Z"
    }
   },
   "outputs": [],
   "source": [
    "fig = mp.figure(figsize=(5.333,5.666))\n",
    "\n",
    "fontsize=12\n",
    "lon_offset=25.\n",
    "\n",
    "map_proj = cartopy.crs.PlateCarree()\n",
    "\n",
    "ax1 = fig.add_subplot(221, projection=map_proj)\n",
    "ax2 = fig.add_subplot(222, projection=map_proj)\n",
    "ax3 = fig.add_subplot(223, projection=map_proj)\n",
    "ax4 = fig.add_subplot(224, projection=map_proj)\n",
    "\n",
    "cp1 = ax1.pcolormesh(xlong,xlat,\\\n",
    "                     cape2d_mcape_diffs[:24,:,:].mean(axis=0),\\\n",
    "                     cmap='PuOr_r', vmin=-150,vmax=150)\n",
    "\n",
    "cp2 = ax2.pcolormesh(xlong,xlat,\\\n",
    "                     cape2d_mcin_diffs[:24,:,:].mean(axis=0),\\\n",
    "                     cmap='PuOr', vmin=-5,vmax=5)\n",
    "\n",
    "cp3 = ax3.pcolormesh(xlong,xlat,\\\n",
    "                     cape2d_lcl_diffs[:24,:,:].mean(axis=0),\\\n",
    "                     cmap=cmocean.cm.amp, vmin=0,vmax=150)\n",
    "\n",
    "cp4 = ax4.pcolormesh(xlong,xlat,\\\n",
    "                     cape2d_lfc_diffs[:24,:,:].mean(axis=0),\\\n",
    "                     cmap=cmocean.cm.amp, vmin=0,vmax=150)\n",
    "\n",
    "ax1.coastlines()\n",
    "ax2.coastlines()\n",
    "ax3.coastlines()\n",
    "ax4.coastlines()\n",
    "\n",
    "ax1.set_extent([273,329,-30,15], crs=map_proj)\n",
    "ax2.set_extent([273,329,-30,15], crs=map_proj)\n",
    "ax3.set_extent([273,329,-30,15], crs=map_proj)\n",
    "ax4.set_extent([273,329,-30,15], crs=map_proj)\n",
    "\n",
    "ax1.set_xticks([-360+280,-360+300,-360+320])\n",
    "ax1.set_yticks([-20,-10,0,10])\n",
    "ax1.tick_params(labelsize=fontsize)\n",
    "ax1.set_xticklabels(['$280$'+DEGREE_SIGN,'$300$'+DEGREE_SIGN,'$320$'+DEGREE_SIGN])\n",
    "ax1.set_yticklabels(['$-20$'+DEGREE_SIGN,'$-10$'+DEGREE_SIGN,'$0$'+DEGREE_SIGN,'$10$'+DEGREE_SIGN])\n",
    "\n",
    "ax2.set_xticks([-360+280,-360+300,-360+320])\n",
    "ax2.set_yticks([-20,-10,0,10])\n",
    "ax2.tick_params(labelsize=fontsize)\n",
    "ax2.set_xticklabels(['$280$'+DEGREE_SIGN,'$300$'+DEGREE_SIGN,'$320$'+DEGREE_SIGN])\n",
    "ax2.set_yticklabels(['$-20$'+DEGREE_SIGN,'$-10$'+DEGREE_SIGN,'$0$'+DEGREE_SIGN,'$10$'+DEGREE_SIGN])\n",
    "\n",
    "ax3.set_xticks([-360+280,-360+300,-360+320])\n",
    "ax3.set_yticks([-20,-10,0,10])\n",
    "ax3.tick_params(labelsize=fontsize)\n",
    "ax3.set_xticklabels(['$280$'+DEGREE_SIGN,'$300$'+DEGREE_SIGN,'$320$'+DEGREE_SIGN])\n",
    "ax3.set_yticklabels(['$-20$'+DEGREE_SIGN,'$-10$'+DEGREE_SIGN,'$0$'+DEGREE_SIGN,'$10$'+DEGREE_SIGN])\n",
    "\n",
    "ax4.set_xticks([-360+280,-360+300,-360+320])\n",
    "ax4.set_yticks([-20,-10,0,10])\n",
    "ax4.tick_params(labelsize=fontsize)\n",
    "ax4.set_xticklabels(['$280$'+DEGREE_SIGN,'$300$'+DEGREE_SIGN,'$320$'+DEGREE_SIGN])\n",
    "ax4.set_yticklabels(['$-20$'+DEGREE_SIGN,'$-10$'+DEGREE_SIGN,'$0$'+DEGREE_SIGN,'$10$'+DEGREE_SIGN])\n",
    "\n",
    "title = ax1.text(s='(a) $\\Delta$ CAPE', x=0, y=1.02, ha='left', va='bottom', fontsize=fontsize, transform=ax1.transAxes)\n",
    "title = ax2.text(s='(b) $\\Delta$ CIN', x=0, y=1.02, ha='left', va='bottom', fontsize=fontsize, transform=ax2.transAxes)\n",
    "title = ax3.text(s='(c) $\\Delta$ LCL', x=0, y=1.02, ha='left', va='bottom', fontsize=fontsize, transform=ax3.transAxes)\n",
    "title = ax4.text(s='(d) $\\Delta$ LFC', x=0, y=1.02, ha='left', va='bottom', fontsize=fontsize, transform=ax4.transAxes)\n",
    "\n",
    "fig.tight_layout()\n",
    "\n",
    "cbar_thickness=0.025\n",
    "cbar_yshift=0#-0.08\n",
    "\n",
    "bbox_position = ax1.get_position()\n",
    "cbar_axes = fig.add_axes([bbox_position.x0, bbox_position.y0+cbar_yshift, bbox_position.width, cbar_thickness])\n",
    "cbar = mp.colorbar(cp1, orientation='horizontal', cax=cbar_axes)\n",
    "#cbar.set_label(label='mm day$^{-1}$', fontsize=fontsize)\n",
    "cbar_axes.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-150,0,150])\n",
    "\n",
    "bbox_position = ax2.get_position()\n",
    "cbar_axes = fig.add_axes([bbox_position.x0, bbox_position.y0+cbar_yshift, bbox_position.width, cbar_thickness])\n",
    "cbar = mp.colorbar(cp2, orientation='horizontal', cax=cbar_axes)\n",
    "#cbar.set_label(label='mm day$^{-1}$', fontsize=fontsize)\n",
    "cbar_axes.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-5,-2.5,0,2.5,5])\n",
    "\n",
    "bbox_position = ax3.get_position()\n",
    "cbar_axes = fig.add_axes([bbox_position.x0, bbox_position.y0+cbar_yshift, bbox_position.width, cbar_thickness])\n",
    "cbar = mp.colorbar(cp3, orientation='horizontal', cax=cbar_axes)\n",
    "#cbar.set_label(label='W m$^{-2}$', fontsize=fontsize)\n",
    "cbar_axes.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([0,50,100,150])\n",
    "\n",
    "bbox_position = ax4.get_position()\n",
    "cbar_axes = fig.add_axes([bbox_position.x0, bbox_position.y0+cbar_yshift, bbox_position.width, cbar_thickness])\n",
    "cbar = mp.colorbar(cp4, orientation='horizontal', cax=cbar_axes)\n",
    "#cbar.set_label(label=degree_sign+'C', fontsize=fontsize)\n",
    "cbar_axes.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([0,50,100,150])\n",
    "\n",
    "ax1.text(s='J kg$^{\\,-1}$', x=0.95, y=0.95, \\\n",
    "        ha='right', va='top', \\\n",
    "        fontsize=fontsize, \\\n",
    "        transform=ax1.transAxes)\n",
    "ax2.text(s='J kg$^{\\,-1}$', x=0.95, y=0.95, \\\n",
    "        ha='right', va='top', \\\n",
    "        fontsize=fontsize, \\\n",
    "        transform=ax2.transAxes)\n",
    "ax3.text(s='m', x=0.95, y=0.95, \\\n",
    "        ha='right', va='top', \\\n",
    "        fontsize=fontsize, \\\n",
    "        transform=ax3.transAxes)\n",
    "ax4.text(s='m', x=0.95, y=0.93, \\\n",
    "        ha='right', va='top', \\\n",
    "        fontsize=fontsize, \\\n",
    "        transform=ax4.transAxes)\n",
    "\n",
    "#fig.savefig('cape_cin_lcl_lfc_days2to10.png', bbox_inches='tight', dpi=600, )\n",
    "fig.savefig('cape_cin_lcl_lfc_day1.png', bbox_inches='tight', dpi=600, )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.554885Z",
     "start_time": "2018-08-16T14:22:29.572Z"
    }
   },
   "outputs": [],
   "source": [
    "net_atm_diffs_1000_sw_latlon_subset = numpy.nanmean(net_atm_diffs_1000_sw[:,latlon_indices], axis=1)\n",
    "net_atm_diffs_400_sw_latlon_subset = numpy.nanmean(net_atm_diffs_400_sw[:,latlon_indices], axis=1)\n",
    "\n",
    "net_atm_diffs_1000_lw_latlon_subset = numpy.nanmean(net_atm_diffs_1000_lw[:,latlon_indices], axis=1)\n",
    "net_atm_diffs_400_lw_latlon_subset = numpy.nanmean(net_atm_diffs_400_lw[:,latlon_indices], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.556247Z",
     "start_time": "2018-08-16T14:22:29.575Z"
    }
   },
   "outputs": [],
   "source": [
    "net_atm_diffs_1000_swc_latlon_subset = numpy.nanmean(net_atm_diffs_1000_swc[:,latlon_indices], axis=1)\n",
    "net_atm_diffs_400_swc_latlon_subset = numpy.nanmean(net_atm_diffs_400_swc[:,latlon_indices], axis=1)\n",
    "\n",
    "net_atm_diffs_1000_lwc_latlon_subset = numpy.nanmean(net_atm_diffs_1000_lwc[:,latlon_indices], axis=1)\n",
    "net_atm_diffs_400_lwc_latlon_subset = numpy.nanmean(net_atm_diffs_400_lwc[:,latlon_indices], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.568188Z",
     "start_time": "2018-08-16T14:22:29.580Z"
    }
   },
   "outputs": [],
   "source": [
    "xvals = numpy.arange(net_atm_diffs_400_sw_latlon_subset.size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.569582Z",
     "start_time": "2018-08-16T14:22:29.584Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axl = mp.subplots(3,2)\n",
    "fig.set_size_inches(8.5,5)\n",
    "\n",
    "axl[0,0].plot(xvals-4,net_atm_diffs_400_sw_latlon_subset/(60*60),c='firebrick')\n",
    "axl[1,0].plot(xvals-4,net_atm_diffs_400_lw_latlon_subset/(60*60),c='firebrick')\n",
    "#axl[1,0].plot(xvals-4,net_atm_diffs_1000_lw_latlon_subset/(60*60),c='coral')\n",
    "\n",
    "axl[2,0].plot(xvals-4,(net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "\n",
    "axl[0,1].plot(xvals-4,(net_atm_diffs_1000_sw_latlon_subset-net_atm_diffs_400_sw_latlon_subset)/(60*60),c='firebrick')\n",
    "axl[1,1].plot(xvals-4,(net_atm_diffs_1000_lw_latlon_subset-net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "axl[2,1].plot(xvals-4,\\\n",
    "              ((net_atm_diffs_1000_sw_latlon_subset+net_atm_diffs_1000_lw_latlon_subset)-\\\n",
    "               (net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset))/(60*60),c='firebrick')\n",
    "\n",
    "axl[0,0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "#axl[1,0].axhline(0, lw=1, c='0.1', ls='--')\n",
    "axl[2,0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[0,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[1,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[2,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "\n",
    "axl[0,0].set_ylim(0,400)\n",
    "axl[1,0].set_ylim(-300,-150)\n",
    "axl[2,0].set_ylim(-400,200)\n",
    "axl[0,1].set_ylim(-20,10)\n",
    "axl[1,1].set_ylim(-10,10)\n",
    "axl[2,1].set_ylim(-5,5)\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axl.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axl.flatten()]\n",
    "[ax.set_ylabel('W m$^{\\,-2}$', fontsize=fontsize) for ax in axl.flatten()[0::2]]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in axl.flatten()[-2:]]\n",
    "[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axl.flatten()]\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axl.flatten()]\n",
    "[ax.xaxis.grid() for ax in axl.flatten()]\n",
    "# [ax.set_axisbelow(True) for ax in axl.flatten()]\n",
    "\n",
    "plot_labels = [\\\n",
    "'net atmos. shortwave (control)',\\\n",
    "'phys.– cont. net atmos. shortwave',\\\n",
    "'net atmos. longwave (control)',\\\n",
    "'phys.– cont. net atmos. longwave',\\\n",
    "'net atmos. rad. (control)',\\\n",
    "'phys.– cont. net atmos. rad.']\n",
    "\n",
    "[axl.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.03,\\\n",
    "        transform=axl.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axl.flatten().size)]\n",
    "\n",
    "fig.tight_layout(h_pad=2)\n",
    "\n",
    "fig.savefig('radiation_changes_with_control.png',\\\n",
    "           dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.570947Z",
     "start_time": "2018-08-16T14:22:29.588Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axl = mp.subplots(3,2)\n",
    "fig.set_size_inches(8.5,5)\n",
    "\n",
    "axl[0,0].plot(xvals-4,net_atm_diffs_400_sw_latlon_subset/(60*60),c='firebrick')\n",
    "axl[1,0].plot(xvals-4,net_atm_diffs_400_lw_latlon_subset/(60*60),c='firebrick')\n",
    "#axl[1,0].plot(xvals-4,net_atm_diffs_1000_lw_latlon_subset/(60*60),c='coral')\n",
    "\n",
    "axl[2,0].plot(xvals-4,(net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "\n",
    "axl[0,1].plot(xvals-4,(net_atm_diffs_1000_sw_latlon_subset-net_atm_diffs_400_sw_latlon_subset)/(60*60),c='firebrick')\n",
    "axl[1,1].plot(xvals-4,(net_atm_diffs_1000_lw_latlon_subset-net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "axl[2,1].plot(xvals-4,((net_atm_diffs_1000_sw_latlon_subset+net_atm_diffs_1000_lw_latlon_subset)-(net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset))/(60*60),c='firebrick')\n",
    "\n",
    "axl[0,0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "#axl[1,0].axhline(0, lw=1, c='0.1', ls='--')\n",
    "axl[2,0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[0,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[1,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axl[2,1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "\n",
    "axl[0,0].set_ylim(0,400)\n",
    "axl[1,0].set_ylim(-300,-150)\n",
    "axl[2,0].set_ylim(-400,200)\n",
    "axl[0,1].set_ylim(-20,10)\n",
    "axl[1,1].set_ylim(-10,10)\n",
    "axl[2,1].set_ylim(-5,5)\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axl.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axl.flatten()]\n",
    "[ax.set_ylabel('W m$^{\\,-2}$', fontsize=fontsize) for ax in axl.flatten()[0::2]]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in axl.flatten()[-2:]]\n",
    "[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axl.flatten()]\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axl.flatten()]\n",
    "[ax.xaxis.grid() for ax in axl.flatten()]\n",
    "# [ax.set_axisbelow(True) for ax in axl.flatten()]\n",
    "\n",
    "plot_labels = [\\\n",
    "'net atmos. shortwave (control)',\\\n",
    "'phys.– cont. net atmos. shortwave',\\\n",
    "'net atmos. longwave (control)',\\\n",
    "'phys.– cont. net atmos. longwave',\\\n",
    "'phys.– cont. net atmos. rad.',\\\n",
    "'net atmos. rad. (control)' ]\n",
    "\n",
    "[axl.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.03,\\\n",
    "        transform=axl.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axl.flatten().size)]\n",
    "\n",
    "fig.tight_layout(h_pad=2)\n",
    "\n",
    "fig.savefig('radiation_changes_with_control.png',\\\n",
    "           dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# calculate cloud fraction changes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.572293Z",
     "start_time": "2018-08-16T14:22:29.592Z"
    }
   },
   "outputs": [],
   "source": [
    "cldfra_diffs = cldfra_1000-cldfra_400\n",
    "cldfra_diffs_latlon_subset = numpy.mean(cldfra_diffs.values[:,:,latlon_indices], axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.573647Z",
     "start_time": "2018-08-16T14:22:29.597Z"
    }
   },
   "outputs": [],
   "source": [
    "cape2d_diffs_latlon_mean = numpy.nanmean(cape2d_diffs[:,:,latlon_indices], axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.575002Z",
     "start_time": "2018-08-16T14:22:29.601Z"
    }
   },
   "outputs": [],
   "source": [
    "cape3d_400_latlon_mean = numpy.nanmean(cape3d_400[:,:,:,latlon_indices], axis=3)\n",
    "cape3d_1000_latlon_mean = numpy.nanmean(cape3d_1000[:,:,:,latlon_indices], axis=3)\n",
    "\n",
    "cape3d_diffs_latlon_mean = cape3d_1000_latlon_mean - cape3d_400_latlon_mean"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot cloud changes and radiative changes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.576423Z",
     "start_time": "2018-08-16T14:22:29.607Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axes = mp.subplots(5,1)\n",
    "fig.set_size_inches(4.25,8)\n",
    "\n",
    "axes[0].plot(xvals-4,(net_atm_diffs_1000_sw_latlon_subset-net_atm_diffs_400_sw_latlon_subset)/(60*60),c='firebrick')\n",
    "axes[1].plot(xvals-4,(net_atm_diffs_1000_lw_latlon_subset-net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "axes[2].plot(xvals-4,\\\n",
    "            ((net_atm_diffs_1000_sw_latlon_subset+net_atm_diffs_1000_lw_latlon_subset)-\\\n",
    "             (net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset))/(60*60),c='firebrick')\n",
    "\n",
    "levels = numpy.arange(-.1,.11,0.02)*100.\n",
    "contour_plot = axes[3].contourf(numpy.arange(241)-4,\\\n",
    "                                ptot_latlon_subset_timemean/100.,\\\n",
    "                                cldfra_diffs_latlon_subset.T*100.,\\\n",
    "                                levels=levels, extend='both',\\\n",
    "                                cmap=cmocean.cm.balance_r)\n",
    "\n",
    "levels = numpy.arange(-.25,.251,0.05)\n",
    "contour_plot2 = axes[4].pcolormesh(numpy.arange(241)-4,\\\n",
    "                                ptot_latlon_subset_timemean/100.,\\\n",
    "                                cape3d_cape_diffs_latlon_mean.T,\\\n",
    "                                vmin=-200,vmax=200,\\\n",
    "                                #levels=levels, extend='both',\\\n",
    "                                cmap='PuOr')\n",
    "axes[3].set_ylim(1013.25,700)\n",
    "axes[4].set_ylim(1013.25,700)\n",
    "\n",
    "axes[0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axes[1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axes[2].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "\n",
    "axes[1].set_ylim(-10,10)\n",
    "axes[2].set_ylim(-5,5)\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axes.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axes.flatten()]\n",
    "[ax.set_ylabel('W m$^{\\,-2}$', fontsize=fontsize) for ax in axes.flatten()[0:3]]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in [axes[-1]]]\n",
    "[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axes.flatten()]\n",
    "[ax.xaxis.grid() for ax in axes.flatten()]\n",
    "\n",
    "axes[3].set_ylabel('hPa', fontsize=fontsize)\n",
    "axes[3].set_yticks([700,800,900,1000])\n",
    "\n",
    "axes[4].set_ylabel('hPa', fontsize=fontsize)\n",
    "axes[4].set_yticks([700,800,900,1000])\n",
    "\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axes.flatten()]\n",
    "[ax.set_xticklabels([]) for ax in axes[0:4]]\n",
    "# [ax.set_axisbelow(True) for ax in axes.flatten()]\n",
    "\n",
    "plot_labels = [\\\n",
    "'(a) $\\Delta$ net atmos. shortwave',\\\n",
    "'(b) $\\Delta$ net atmos. longwave',\\\n",
    "'(c) $\\Delta$ net atmos. rad.',\\\n",
    "'(d) $\\Delta$ cloud fraction', \\\n",
    "'(e) $\\Delta$ CAPE']\n",
    "\n",
    "[axes.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.03,\\\n",
    "        transform=axes.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axes.flatten().size)]\n",
    "\n",
    "fig.tight_layout(h_pad=1.5)\n",
    "\n",
    "ax4pos = axes[3].get_position()\n",
    "cbar_ax = fig.add_axes([ax4pos.x1+0.025,ax4pos.y0,0.025,ax4pos.height])\n",
    "cbar = fig.colorbar(contour_plot, cax=cbar_ax)\n",
    "cbar_ax.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-10,-5,0,5,10])\n",
    "cbar.set_label('%', fontsize=fontsize)\n",
    "\n",
    "ax5pos = axes[4].get_position()\n",
    "cbar_ax = fig.add_axes([ax5pos.x1+0.025,ax5pos.y0,0.025,ax5pos.height])\n",
    "cbar = fig.colorbar(contour_plot2, cax=cbar_ax)\n",
    "cbar_ax.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-200,-100,0,100,200])\n",
    "cbar.set_label('kJ kg$^{\\,-1}$', fontsize=fontsize)\n",
    "\n",
    "fig.savefig('radiation_cloud_cape_changes.png',\\\n",
    "           dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-27T00:46:57.411845Z",
     "start_time": "2018-06-27T00:46:57.391487Z"
    }
   },
   "source": [
    "# plot cape and cin background"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.577933Z",
     "start_time": "2018-08-16T14:22:29.612Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axes = mp.subplots(2,2)\n",
    "fig.set_size_inches(8.5,4)\n",
    "\n",
    "cp1 = axes[0,0].pcolormesh(numpy.arange(241)-4,\\\n",
    "                         ptot_latlon_subset_timemean/100.,\\\n",
    "                         cape3d_400_latlon_mean[0,:].T,\\\n",
    "                         vmin=0,vmax=1500,\\\n",
    "                         #levels=levels, extend='both',\\\n",
    "                         cmap=cmocean.cm.amp)\n",
    "axes[0,0].set_ylim(1013.25,700)\n",
    "\n",
    "cp2 = axes[0,1].pcolormesh(numpy.arange(241)-4,\\\n",
    "                         ptot_latlon_subset_timemean/100.,\\\n",
    "                         cape3d_diffs_latlon_mean[0,:].T,\\\n",
    "                         vmin=-200,vmax=200,\\\n",
    "                         cmap=cmocean.cm.balance)\n",
    "axes[0,1].set_ylim(1013.25,700)\n",
    "\n",
    "\n",
    "cp3 = axes[1,0].pcolormesh(numpy.arange(241)-4,\\\n",
    "                         ptot_latlon_subset_timemean/100.,\\\n",
    "                         cape3d_400_latlon_mean[1,:].T,\\\n",
    "                         vmin=0,vmax=20,\\\n",
    "                         #levels=levels, extend='both',\\\n",
    "                         cmap=cmocean.cm.deep)\n",
    "axes[1,0].set_ylim(1013.25,700)\n",
    "\n",
    "\n",
    "\n",
    "cp4 = axes[1,1].pcolormesh(numpy.arange(241)-4,\\\n",
    "                         ptot_latlon_subset_timemean/100.,\\\n",
    "                         cape3d_diffs_latlon_mean[1,:].T,\\\n",
    "                         vmin=-10,vmax=10,\\\n",
    "                         cmap=cmocean.cm.balance_r)\n",
    "axes[1,1].set_ylim(1013.25,700)\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axes.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axes.flatten()]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in [axes[1,0],axes[1,1]]]\n",
    "#[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axes.flatten()]\n",
    "[ax.xaxis.grid() for ax in axes.flatten()]\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axes.flatten()]\n",
    "[ax.set_xticklabels([]) for ax in axes.flatten()[0:2]]\n",
    "[ax.set_ylabel('hPa', fontsize=fontsize) for ax in axes.flatten()[0::2]]\n",
    "[ax.set_yticks([700,800,900,1000]) for ax in axes.flatten()]\n",
    "\n",
    "fig.tight_layout(h_pad=3,w_pad=5)\n",
    "\n",
    "plot_labels = [\\\n",
    "'(a) background CAPE (J kg$^{\\,-1}$)',\\\n",
    "'(b) phys.– cont. CAPE (J kg$^{\\,-1}$)',\\\n",
    "'(c) background CIN (J kg$^{\\,-1}$)',\\\n",
    "'(d) phys.– cont. CIN (J kg$^{\\,-1}$)']\n",
    "\n",
    "[axes.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.02,\\\n",
    "        transform=axes.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axes.flatten().size)]\n",
    "\n",
    "cbar_list = [cp1,cp2,cp3,cp4]\n",
    "pos_list = [ax.get_position() for ax in axes.flatten()]\n",
    "cbar_ax_list = [fig.add_axes([axpos.x1+0.02,axpos.y0,0.02,axpos.height]) for axpos in pos_list]\n",
    "#cbar_ax_list += [fig.add_axes([axpos.x0,axpos.y0-.1,axpos.width,0.025]) for axpos in pos_list[2:4]]\n",
    "\n",
    "cbar_list = [fig.colorbar(cbar_list[i], cax=cbar_ax_list[i], orientation='vertical') for i in range(4)]\n",
    "[cax.tick_params(labelsize=fontsize) for cax in cbar_ax_list]\n",
    "\n",
    "fig.savefig('background_and_change_CAPE_CIN.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.579290Z",
     "start_time": "2018-08-16T14:22:29.617Z"
    }
   },
   "outputs": [],
   "source": [
    "pos_list[0::2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.580863Z",
     "start_time": "2018-08-16T14:22:29.633Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axes = mp.subplots(3,1)\n",
    "fig.set_size_inches(4.25,5)\n",
    "\n",
    "levels = numpy.arange(-.1,.11,0.02)*100.\n",
    "contour_plot = axes[0].contourf(numpy.arange(241)-4,\\\n",
    "                                ptot_latlon_subset_timemean/100.,\\\n",
    "                                cldfra_diffs_latlon_subset.T*100.,\\\n",
    "                                levels=levels, extend='both',\\\n",
    "                                cmap=cmocean.cm.balance_r)\n",
    "\n",
    "axes[1].plot(numpy.arange(241)-4,\\\n",
    "             cape2d_diffs_latlon_mean[0,:], c='firebrick')\n",
    "\n",
    "axes[2].plot(numpy.arange(241)-4,\\\n",
    "             cape2d_diffs_latlon_mean[1,:], c='firebrick')\n",
    "# levels = numpy.arange(-.25,.251,0.05)\n",
    "# contour_plot2 = axes[1].pcolormesh(numpy.arange(241)-4,\\\n",
    "#                                 ptot_latlon_subset_timemean/100.,\\\n",
    "#                                 cape3d_cape_diffs_latlon_mean.T,\\\n",
    "#                                 vmin=-200,vmax=200,\\\n",
    "#                                 #levels=levels, extend='both',\\\n",
    "#                                 cmap='PuOr')\n",
    "axes[0].set_ylim(1013.25,700)\n",
    "\n",
    "axes[0].set_ylabel('W m$^{\\,-2}$', fontsize=fontsize)\n",
    "axes[1].set_ylabel('J kg$^{\\,-1}$', fontsize=fontsize)\n",
    "axes[2].set_ylabel('J kg$^{\\,-1}$', fontsize=fontsize)\n",
    "\n",
    "axes[1].axhline(0, lw='1', c='0.1', ls='--')\n",
    "axes[2].axhline(0, lw='1', c='0.1', ls='--')\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axes.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axes.flatten()]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in [axes[-1]]]\n",
    "[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axes.flatten()]\n",
    "[ax.xaxis.grid() for ax in axes.flatten()]\n",
    "\n",
    "axes[0].set_ylabel('hPa', fontsize=fontsize)\n",
    "axes[0].set_yticks([700,800,900,1000])\n",
    "\n",
    "# axes[1].set_ylabel('hPa', fontsize=fontsize)\n",
    "# axes[1].set_yticks([700,800,900,1000])\n",
    "\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axes.flatten()]\n",
    "[ax.set_xticklabels([]) for ax in axes[0:2]]\n",
    "# [ax.set_axisbelow(True) for ax in axes.flatten()]\n",
    "\n",
    "plot_labels = [\\\n",
    "'(a) $\\Delta$ cloud fraction', \\\n",
    "'(b) $\\Delta$ CAPE', \\\n",
    "'(c) $\\Delta$ CIN']\n",
    "\n",
    "[axes.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.03,\\\n",
    "        transform=axes.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axes.flatten().size)]\n",
    "\n",
    "fig.tight_layout(h_pad=2)\n",
    "\n",
    "ax1pos = axes[0].get_position()\n",
    "cbar_ax = fig.add_axes([ax1pos.x1+0.025,ax1pos.y0,0.025,ax1pos.height])\n",
    "cbar = fig.colorbar(contour_plot, cax=cbar_ax)\n",
    "cbar_ax.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-10,-5,0,5,10])\n",
    "cbar.set_label('%', fontsize=fontsize)\n",
    "\n",
    "# ax2pos = axes[1].get_position()\n",
    "# cbar_ax = fig.add_axes([ax2pos.x1+0.025,ax2pos.y0,0.025,ax2pos.height])\n",
    "# cbar = fig.colorbar(contour_plot2, cax=cbar_ax)\n",
    "# cbar_ax.tick_params(labelsize=fontsize)\n",
    "# cbar.set_ticks([-200,-100,0,100,200])\n",
    "# cbar.set_label('kJ kg$^{\\,-1}$', fontsize=fontsize)\n",
    "\n",
    "fig.savefig('cloud_cape_cin_timeseries_changes.png',\\\n",
    "           dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.593957Z",
     "start_time": "2018-08-16T14:22:29.638Z"
    }
   },
   "outputs": [],
   "source": [
    "fontsize = 12\n",
    "\n",
    "fig,axes = mp.subplots(4,1)\n",
    "fig.set_size_inches(4.25,6)\n",
    "\n",
    "axes[0].plot(xvals-4,(net_atm_diffs_1000_sw_latlon_subset-net_atm_diffs_400_sw_latlon_subset)/(60*60),c='firebrick',label='full')\n",
    "axes[1].plot(xvals-4,(net_atm_diffs_1000_lw_latlon_subset-net_atm_diffs_400_lw_latlon_subset)/(60*60),c='firebrick')\n",
    "axes[2].plot(xvals-4,\\\n",
    "            ((net_atm_diffs_1000_sw_latlon_subset+net_atm_diffs_1000_lw_latlon_subset)-\\\n",
    "             (net_atm_diffs_400_sw_latlon_subset+net_atm_diffs_400_lw_latlon_subset))/(60*60),c='firebrick')\n",
    "\n",
    "\n",
    "axes[0].plot(xvals-4,(net_atm_diffs_1000_swc_latlon_subset-net_atm_diffs_400_swc_latlon_subset)/(60*60),c='coral',label='clear sky')\n",
    "axes[1].plot(xvals-4,(net_atm_diffs_1000_lwc_latlon_subset-net_atm_diffs_400_lwc_latlon_subset)/(60*60),c='coral')\n",
    "axes[2].plot(xvals-4,\\\n",
    "            ((net_atm_diffs_1000_swc_latlon_subset+net_atm_diffs_1000_lwc_latlon_subset)-\\\n",
    "             (net_atm_diffs_400_swc_latlon_subset+net_atm_diffs_400_lwc_latlon_subset))/(60*60),c='coral')\n",
    "\n",
    "contour_plot = axes[3].contourf(numpy.arange(241)-4,\\\n",
    "                                ptot_latlon_subset_timemean/100.,\\\n",
    "                                cldfra_diffs_latlon_subset.T,\\\n",
    "                                levels=levels, extend='both',\\\n",
    "                                cmap=cmocean.cm.balance_r)\n",
    "\n",
    "axes[3].set_ylim(1013.25,700)\n",
    "\n",
    "axes[0].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axes[1].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "axes[2].axhline(0, lw=1, c='0.1', ls='--',zorder=0)\n",
    "\n",
    "axes[1].set_ylim(-10,10)\n",
    "axes[2].set_ylim(-5,5)\n",
    "\n",
    "[ax.tick_params(labelsize=fontsize) for ax in axes.flatten()]\n",
    "[ax.set_xlim(0,240) for ax in axes.flatten()]\n",
    "[ax.set_ylabel('W m$^{\\,-2}$', fontsize=fontsize) for ax in axes.flatten()[0:3]]\n",
    "[ax.set_xlabel('hour', fontsize=fontsize) for ax in [axes[-1]]]\n",
    "[ax.get_yaxis().set_label_coords(-0.15,0.5) for ax in axes.flatten()]\n",
    "[ax.xaxis.grid() for ax in axes.flatten()]\n",
    "\n",
    "axes[3].set_ylabel('hPa', fontsize=fontsize)\n",
    "axes[3].set_yticks([700,800,900,1000])\n",
    "\n",
    "[ax.set_xticks(numpy.arange(0,241,24)) for ax in axes.flatten()]\n",
    "[ax.set_xticklabels([]) for ax in axes[0:3]]\n",
    "# [ax.set_axisbelow(True) for ax in axes.flatten()]\n",
    "\n",
    "plot_labels = [\\\n",
    "'phys.– cont. net atmos. shortwave',\\\n",
    "'phys.– cont. net atmos. longwave',\\\n",
    "'phys.– cont. net atmos. rad.',\\\n",
    "'phys.– cont. cloud fraction change' ]\n",
    "\n",
    "[axes.flatten()[i].text(s=plot_labels[i], ha='left',\\\n",
    "        va='bottom',\\\n",
    "        x=0,y=1.03,\\\n",
    "        transform=axes.flatten()[i].transAxes,\\\n",
    "        fontsize=fontsize)\\\n",
    "for i in range(axes.flatten().size)]\n",
    "\n",
    "fig.tight_layout(h_pad=2)\n",
    "\n",
    "ax4pos = axes[3].get_position()\n",
    "cbar_ax = fig.add_axes([ax4pos.x1+0.025,ax4pos.y0,0.025,ax4pos.height])\n",
    "cbar = fig.colorbar(contour_plot, cax=cbar_ax)\n",
    "cbar_ax.tick_params(labelsize=fontsize)\n",
    "cbar.set_ticks([-.1,-0.05,0,0.05,0.1])\n",
    "cbar.set_label('%', fontsize=fontsize)\n",
    "\n",
    "handles,labels=axes[0].get_legend_handles_labels()\n",
    "axes[0].legend(handles, labels, fontsize=fontsize, scatterpoints=1, ncol=1, \\\n",
    "               bbox_to_anchor=(1.03,1.0), bbox_transform=axes[0].transAxes, \\\n",
    "               loc='upper left', borderaxespad=0,\\\n",
    "               fancybox=False, edgecolor='0', framealpha=0, borderpad=0., \\\n",
    "               handletextpad=0.5)\n",
    "\n",
    "fig.savefig('radiation_AND_CLOUD_changes_with_control_full_and_clearsky.png',\\\n",
    "           dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# get time data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.595290Z",
     "start_time": "2018-08-16T14:22:29.671Z"
    }
   },
   "outputs": [],
   "source": [
    "time_datetimes = pandas.to_datetime([i.decode('utf-8') for i in time], format = '%Y-%m-%d_%H:%M:%S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.596622Z",
     "start_time": "2018-08-16T14:22:29.676Z"
    }
   },
   "outputs": [],
   "source": [
    "skip_time = 24\n",
    "n_intervals = 8\n",
    "hour_strings = ['06','09','12','15','18','21','00','03']\n",
    "hour_strings_local = ['02', '05', '08', '11', '14', '17','20', '23']\n",
    "\n",
    "times_00 = [t.hour in [23,0,1] for t in time_datetimes[skip_time:]]\n",
    "times_03 = [t.hour in [2,3,4] for t in time_datetimes[skip_time:]]\n",
    "times_06 = [t.hour in [5,6,7] for t in time_datetimes[skip_time:]]\n",
    "times_09 = [t.hour in [8,9,10] for t in time_datetimes[skip_time:]]\n",
    "times_12 = [t.hour in [11,12,13] for t in time_datetimes[skip_time:]]\n",
    "times_15 = [t.hour in [14,15,16] for t in time_datetimes[skip_time:]]\n",
    "times_18 = [t.hour in [17,18,19] for t in time_datetimes[skip_time:]]\n",
    "times_21 = [t.hour in [20,21,22] for t in time_datetimes[skip_time:]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# lat indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.598089Z",
     "start_time": "2018-08-16T14:22:29.683Z"
    }
   },
   "outputs": [],
   "source": [
    "lat_hi = 10.\n",
    "lat_lo = -10.\n",
    "\n",
    "# subset between lat_hi and lat_lo\n",
    "lat_indices = (xlat<=lat_hi)&(xlat>=lat_lo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.599543Z",
     "start_time": "2018-08-16T14:22:29.687Z"
    }
   },
   "outputs": [],
   "source": [
    "lon_lo = 290-360\n",
    "lon_hi = 320-360\n",
    "lon_indices = (xlong[0,:]<lon_lo)|(xlong[0,:]>lon_hi)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# cldfra"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-14T06:19:13.611597Z",
     "start_time": "2018-06-14T06:08:31.797511Z"
    }
   },
   "source": [
    "cldfra_diffs = cldfra_1000.values-cldfra_400.values\n",
    "\n",
    "cldfra_diffs_landmask = numpy.copy(cldfra_diffs)\n",
    "cldfra_diffs_landmask[:,:,landmask==1]=numpy.nan\n",
    "\n",
    "cldfra_diffs_lat_subset = cldfra_diffs[:,:,lat_indices[:,0],:]\n",
    "\n",
    "cldfra_400_landmask = numpy.copy(cldfra_400)\n",
    "cldfra_400_landmask[:,:,landmask==1]=numpy.nan"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-06-14T06:20:20.509608Z",
     "start_time": "2018-06-14T06:19:14.812532Z"
    }
   },
   "source": [
    "cldfra_400_lat_subset = cldfra_400[:,:,lat_indices[:,0],:]\n",
    "cldfra_diffs_lat_mean = numpy.nanmean(cldfra_diffs_lat_subset, axis=2)\n",
    "cldfra_400_lat_mean = numpy.nanmean(cldfra_400_lat_subset, axis=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# geopotential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.600887Z",
     "start_time": "2018-08-16T14:22:29.845Z"
    }
   },
   "outputs": [],
   "source": [
    "geopot_400_landmask = numpy.copy(geopot_400)\n",
    "geopot_400_landmask[:,:,landmask==1]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.602245Z",
     "start_time": "2018-08-16T14:22:29.849Z"
    }
   },
   "outputs": [],
   "source": [
    "geopot_400_lat_subset = geopot_400[:,:,lat_indices[:,0],:]\n",
    "geopot_400_lat_mean = numpy.nanmean(geopot_400_lat_subset, axis=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# net_bottom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.603591Z",
     "start_time": "2018-08-16T14:22:29.852Z"
    }
   },
   "outputs": [],
   "source": [
    "net_bottom_diffs = net_bottom_diffs_1000-net_bottom_diffs_400"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.604943Z",
     "start_time": "2018-08-16T14:22:29.858Z"
    }
   },
   "outputs": [],
   "source": [
    "net_bottom_diffs_landmask = numpy.copy(net_bottom_diffs)\n",
    "net_bottom_diffs_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "net_bottom_400_landmask = numpy.copy(net_bottom_400)\n",
    "net_bottom_400_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "net_bottom_1000_landmask = numpy.copy(net_bottom_1000)\n",
    "net_bottom_1000_landmask[:,landmask==0]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.606287Z",
     "start_time": "2018-08-16T14:22:29.861Z"
    }
   },
   "outputs": [],
   "source": [
    "net_bottom_diffs_lat_subset = net_bottom_diffs_landmask[:,lat_indices[:,0],:]\n",
    "net_bottom_400_lat_subset = net_bottom_400_landmask[:,lat_indices[:,0],:]\n",
    "net_bottom_1000_lat_subset = net_bottom_1000_landmask[:,lat_indices[:,0],:]\n",
    "\n",
    "net_bottom_400_lat_subset[:,:,lon_indices]=numpy.nan\n",
    "net_bottom_1000_lat_subset[:,:,lon_indices]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.618191Z",
     "start_time": "2018-08-16T14:22:29.865Z"
    }
   },
   "outputs": [],
   "source": [
    "net_bottom_diffs_lat_mean = numpy.nanmean(net_bottom_diffs_lat_subset, axis=1)\n",
    "net_bottom_400_lat_mean = numpy.nanmean(net_bottom_400_lat_subset, axis=1)\n",
    "net_bottom_1000_lat_mean = numpy.nanmean(net_bottom_1000_lat_subset, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# fira"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.619653Z",
     "start_time": "2018-08-16T14:22:29.870Z"
    }
   },
   "outputs": [],
   "source": [
    "fira_diffs = fira_1000.values-fira_400.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.620976Z",
     "start_time": "2018-08-16T14:22:29.874Z"
    }
   },
   "outputs": [],
   "source": [
    "fira_diffs_landmask = numpy.copy(fira_diffs)\n",
    "fira_diffs_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "fira_400_landmask = numpy.copy(fira_400)\n",
    "fira_400_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "fira_1000_landmask = numpy.copy(fira_1000)\n",
    "fira_1000_landmask[:,landmask==0]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.622332Z",
     "start_time": "2018-08-16T14:22:29.878Z"
    }
   },
   "outputs": [],
   "source": [
    "fira_diffs_lat_subset = fira_diffs_landmask[:,lat_indices[:,0],:]\n",
    "fira_400_lat_subset = fira_400_landmask[:,lat_indices[:,0],:]\n",
    "fira_1000_lat_subset = fira_1000_landmask[:,lat_indices[:,0],:]\n",
    "\n",
    "fira_400_lat_subset[:,:,lon_indices]=numpy.nan\n",
    "fira_1000_lat_subset[:,:,lon_indices]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.623684Z",
     "start_time": "2018-08-16T14:22:29.882Z"
    }
   },
   "outputs": [],
   "source": [
    "fira_diffs_lat_mean = numpy.nanmean(fira_diffs_lat_subset, axis=1)\n",
    "fira_400_lat_mean = numpy.nanmean(fira_400_lat_subset, axis=1)\n",
    "fira_1000_lat_mean = numpy.nanmean(fira_1000_lat_subset, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# fsa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.625027Z",
     "start_time": "2018-08-16T14:22:29.886Z"
    }
   },
   "outputs": [],
   "source": [
    "fsa_diffs = fsa_1000.values-fsa_400.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.626373Z",
     "start_time": "2018-08-16T14:22:29.889Z"
    }
   },
   "outputs": [],
   "source": [
    "fsa_diffs_landmask = numpy.copy(fsa_diffs)\n",
    "fsa_diffs_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "fsa_400_landmask = numpy.copy(fsa_400)\n",
    "fsa_400_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "fsa_1000_landmask = numpy.copy(fsa_1000)\n",
    "fsa_1000_landmask[:,landmask==0]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.627781Z",
     "start_time": "2018-08-16T14:22:29.893Z"
    }
   },
   "outputs": [],
   "source": [
    "fsa_diffs_lat_subset = fsa_diffs_landmask[:,lat_indices[:,0],:]\n",
    "fsa_400_lat_subset = fsa_400_landmask[:,lat_indices[:,0],:]\n",
    "fsa_1000_lat_subset = fsa_1000_landmask[:,lat_indices[:,0],:]\n",
    "\n",
    "fsa_400_lat_subset[:,:,lon_indices]=numpy.nan\n",
    "fsa_1000_lat_subset[:,:,lon_indices]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.629306Z",
     "start_time": "2018-08-16T14:22:29.898Z"
    }
   },
   "outputs": [],
   "source": [
    "fsa_diffs_lat_mean = numpy.nanmean(fsa_diffs_lat_subset, axis=1)\n",
    "fsa_400_lat_mean = numpy.nanmean(fsa_400_lat_subset, axis=1)\n",
    "fsa_1000_lat_mean = numpy.nanmean(fsa_1000_lat_subset, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# glw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.630660Z",
     "start_time": "2018-08-16T14:22:29.901Z"
    }
   },
   "outputs": [],
   "source": [
    "glw_diffs = glw_1000.values-glw_400.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.642458Z",
     "start_time": "2018-08-16T14:22:29.904Z"
    }
   },
   "outputs": [],
   "source": [
    "glw_diffs_landmask = numpy.copy(glw_diffs)\n",
    "glw_diffs_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "glw_400_landmask = numpy.copy(glw_400)\n",
    "glw_400_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "glw_1000_landmask = numpy.copy(glw_1000)\n",
    "glw_1000_landmask[:,landmask==0]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.643887Z",
     "start_time": "2018-08-16T14:22:29.908Z"
    }
   },
   "outputs": [],
   "source": [
    "glw_diffs_lat_subset = glw_diffs_landmask[:,lat_indices[:,0],:]\n",
    "glw_400_lat_subset = glw_400_landmask[:,lat_indices[:,0],:]\n",
    "glw_1000_lat_subset = glw_1000_landmask[:,lat_indices[:,0],:]\n",
    "\n",
    "glw_400_lat_subset[:,:,lon_indices]=numpy.nan\n",
    "glw_1000_lat_subset[:,:,lon_indices]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.645202Z",
     "start_time": "2018-08-16T14:22:29.912Z"
    }
   },
   "outputs": [],
   "source": [
    "glw_diffs_lat_mean = numpy.nanmean(glw_diffs_lat_subset, axis=1)\n",
    "glw_400_lat_mean = numpy.nanmean(glw_400_lat_subset, axis=1)\n",
    "glw_1000_lat_mean = numpy.nanmean(glw_1000_lat_subset, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# swdown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.646543Z",
     "start_time": "2018-08-16T14:22:29.918Z"
    }
   },
   "outputs": [],
   "source": [
    "swdown_diffs = swdown_1000.values-swdown_400.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.647924Z",
     "start_time": "2018-08-16T14:22:29.921Z"
    }
   },
   "outputs": [],
   "source": [
    "swdown_diffs_landmask = numpy.copy(swdown_diffs)\n",
    "swdown_diffs_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "swdown_400_landmask = numpy.copy(swdown_400)\n",
    "swdown_400_landmask[:,landmask==0]=numpy.nan\n",
    "\n",
    "swdown_1000_landmask = numpy.copy(swdown_1000)\n",
    "swdown_1000_landmask[:,landmask==0]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.649276Z",
     "start_time": "2018-08-16T14:22:29.926Z"
    }
   },
   "outputs": [],
   "source": [
    "swdown_diffs_lat_subset = swdown_diffs_landmask[:,lat_indices[:,0],:]\n",
    "swdown_400_lat_subset = swdown_400_landmask[:,lat_indices[:,0],:]\n",
    "swdown_1000_lat_subset = swdown_1000_landmask[:,lat_indices[:,0],:]\n",
    "\n",
    "swdown_400_lat_subset[:,:,lon_indices]=numpy.nan\n",
    "swdown_1000_lat_subset[:,:,lon_indices]=numpy.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.650777Z",
     "start_time": "2018-08-16T14:22:29.929Z"
    }
   },
   "outputs": [],
   "source": [
    "swdown_diffs_lat_mean = numpy.nanmean(swdown_diffs_lat_subset, axis=1)\n",
    "swdown_400_lat_mean = numpy.nanmean(swdown_400_lat_subset, axis=1)\n",
    "swdown_1000_lat_mean = numpy.nanmean(swdown_1000_lat_subset, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 hourly means of net_bottom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.652239Z",
     "start_time": "2018-08-16T14:22:29.933Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "net_bottom_400_t00mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "net_bottom_400_t03mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "net_bottom_400_t06mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "net_bottom_400_t09mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "net_bottom_400_t12mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "net_bottom_400_t15mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "net_bottom_400_t18mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "net_bottom_400_t21mean = numpy.mean(net_bottom_400_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "net_bottom_1000_t00mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "net_bottom_1000_t03mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "net_bottom_1000_t06mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "net_bottom_1000_t09mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "net_bottom_1000_t12mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "net_bottom_1000_t15mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "net_bottom_1000_t18mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "net_bottom_1000_t21mean = numpy.nanmean(net_bottom_1000_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "net_bottom_400_array = numpy.array([net_bottom_400_t06mean, net_bottom_400_t09mean, net_bottom_400_t12mean, net_bottom_400_t15mean, net_bottom_400_t18mean, net_bottom_400_t21mean, net_bottom_400_t00mean, net_bottom_400_t03mean])\n",
    "net_bottom_1000_array = numpy.array([net_bottom_1000_t06mean, net_bottom_1000_t09mean, net_bottom_1000_t12mean, net_bottom_1000_t15mean, net_bottom_1000_t18mean, net_bottom_1000_t21mean, net_bottom_1000_t00mean, net_bottom_1000_t03mean])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.653933Z",
     "start_time": "2018-08-16T14:22:29.938Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "net_bottom_diffs_t00mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_00,:], axis=0)\n",
    "net_bottom_diffs_t03mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_03,:], axis=0)\n",
    "net_bottom_diffs_t06mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_06,:], axis=0)\n",
    "net_bottom_diffs_t09mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_09,:], axis=0)\n",
    "net_bottom_diffs_t12mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_12,:], axis=0)\n",
    "net_bottom_diffs_t15mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_15,:], axis=0)\n",
    "net_bottom_diffs_t18mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_18,:], axis=0)\n",
    "net_bottom_diffs_t21mean = numpy.nanmean(net_bottom_diffs_lat_mean[skip_time:,:][times_21,:], axis=0)\n",
    "\n",
    "net_bottom_diffs_array = numpy.array([net_bottom_diffs_t06mean, net_bottom_diffs_t09mean, net_bottom_diffs_t12mean, net_bottom_diffs_t15mean, net_bottom_diffs_t18mean, net_bottom_diffs_t21mean, net_bottom_diffs_t00mean, net_bottom_diffs_t03mean])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 hourly means of fira"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.655306Z",
     "start_time": "2018-08-16T14:22:29.941Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "fira_400_t00mean = numpy.mean(fira_400_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "fira_400_t03mean = numpy.mean(fira_400_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "fira_400_t06mean = numpy.mean(fira_400_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "fira_400_t09mean = numpy.mean(fira_400_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "fira_400_t12mean = numpy.mean(fira_400_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "fira_400_t15mean = numpy.mean(fira_400_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "fira_400_t18mean = numpy.mean(fira_400_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "fira_400_t21mean = numpy.mean(fira_400_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "fira_1000_t00mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "fira_1000_t03mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "fira_1000_t06mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "fira_1000_t09mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "fira_1000_t12mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "fira_1000_t15mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "fira_1000_t18mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "fira_1000_t21mean = numpy.mean(fira_1000_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "fira_400_array = numpy.array([fira_400_t06mean, fira_400_t09mean, fira_400_t12mean, fira_400_t15mean, fira_400_t18mean, fira_400_t21mean, fira_400_t00mean, fira_400_t03mean])\n",
    "fira_1000_array = numpy.array([fira_1000_t06mean, fira_1000_t09mean, fira_1000_t12mean, fira_1000_t15mean, fira_1000_t18mean, fira_1000_t21mean, fira_1000_t00mean, fira_1000_t03mean])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.656677Z",
     "start_time": "2018-08-16T14:22:29.946Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "fira_diffs_t00mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_00,:], axis=0)\n",
    "fira_diffs_t03mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_03,:], axis=0)\n",
    "fira_diffs_t06mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_06,:], axis=0)\n",
    "fira_diffs_t09mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_09,:], axis=0)\n",
    "fira_diffs_t12mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_12,:], axis=0)\n",
    "fira_diffs_t15mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_15,:], axis=0)\n",
    "fira_diffs_t18mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_18,:], axis=0)\n",
    "fira_diffs_t21mean = numpy.mean(fira_diffs_lat_mean[skip_time:,:][times_21,:], axis=0)\n",
    "\n",
    "fira_diffs_array = numpy.array([fira_diffs_t06mean, fira_diffs_t09mean, fira_diffs_t12mean, fira_diffs_t15mean, fira_diffs_t18mean, fira_diffs_t21mean, fira_diffs_t00mean, fira_diffs_t03mean])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 hourly means of fsa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.668198Z",
     "start_time": "2018-08-16T14:22:29.950Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "fsa_400_t00mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "fsa_400_t03mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "fsa_400_t06mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "fsa_400_t09mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "fsa_400_t12mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "fsa_400_t15mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "fsa_400_t18mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "fsa_400_t21mean = numpy.mean(fsa_400_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "fsa_1000_t00mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "fsa_1000_t03mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "fsa_1000_t06mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "fsa_1000_t09mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "fsa_1000_t12mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "fsa_1000_t15mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "fsa_1000_t18mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "fsa_1000_t21mean = numpy.mean(fsa_1000_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "fsa_400_array = numpy.array([fsa_400_t06mean, fsa_400_t09mean, fsa_400_t12mean, fsa_400_t15mean, fsa_400_t18mean, fsa_400_t21mean, fsa_400_t00mean, fsa_400_t03mean])\n",
    "fsa_1000_array = numpy.array([fsa_1000_t06mean, fsa_1000_t09mean, fsa_1000_t12mean, fsa_1000_t15mean, fsa_1000_t18mean, fsa_1000_t21mean, fsa_1000_t00mean, fsa_1000_t03mean])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.669552Z",
     "start_time": "2018-08-16T14:22:29.953Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "fsa_diffs_t00mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_00,:], axis=0)\n",
    "fsa_diffs_t03mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_03,:], axis=0)\n",
    "fsa_diffs_t06mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_06,:], axis=0)\n",
    "fsa_diffs_t09mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_09,:], axis=0)\n",
    "fsa_diffs_t12mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_12,:], axis=0)\n",
    "fsa_diffs_t15mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_15,:], axis=0)\n",
    "fsa_diffs_t18mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_18,:], axis=0)\n",
    "fsa_diffs_t21mean = numpy.mean(fsa_diffs_lat_mean[skip_time:,:][times_21,:], axis=0)\n",
    "\n",
    "fsa_diffs_array = numpy.array([fsa_diffs_t06mean, fsa_diffs_t09mean, fsa_diffs_t12mean, fsa_diffs_t15mean, fsa_diffs_t18mean, fsa_diffs_t21mean, fsa_diffs_t00mean, fsa_diffs_t03mean])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 hourly means of glw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.671248Z",
     "start_time": "2018-08-16T14:22:29.959Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "glw_400_t00mean = numpy.mean(glw_400_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "glw_400_t03mean = numpy.mean(glw_400_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "glw_400_t06mean = numpy.mean(glw_400_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "glw_400_t09mean = numpy.mean(glw_400_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "glw_400_t12mean = numpy.mean(glw_400_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "glw_400_t15mean = numpy.mean(glw_400_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "glw_400_t18mean = numpy.mean(glw_400_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "glw_400_t21mean = numpy.mean(glw_400_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "glw_1000_t00mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "glw_1000_t03mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "glw_1000_t06mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "glw_1000_t09mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "glw_1000_t12mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "glw_1000_t15mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "glw_1000_t18mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "glw_1000_t21mean = numpy.mean(glw_1000_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "glw_400_array = numpy.array([glw_400_t06mean, glw_400_t09mean, glw_400_t12mean, glw_400_t15mean, glw_400_t18mean, glw_400_t21mean, glw_400_t00mean, glw_400_t03mean])\n",
    "glw_1000_array = numpy.array([glw_1000_t06mean, glw_1000_t09mean, glw_1000_t12mean, glw_1000_t15mean, glw_1000_t18mean, glw_1000_t21mean, glw_1000_t00mean, glw_1000_t03mean])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.672632Z",
     "start_time": "2018-08-16T14:22:29.964Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "glw_diffs_t00mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_00,:], axis=0)\n",
    "glw_diffs_t03mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_03,:], axis=0)\n",
    "glw_diffs_t06mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_06,:], axis=0)\n",
    "glw_diffs_t09mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_09,:], axis=0)\n",
    "glw_diffs_t12mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_12,:], axis=0)\n",
    "glw_diffs_t15mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_15,:], axis=0)\n",
    "glw_diffs_t18mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_18,:], axis=0)\n",
    "glw_diffs_t21mean = numpy.mean(glw_diffs_lat_mean[skip_time:,:][times_21,:], axis=0)\n",
    "\n",
    "glw_diffs_array = numpy.array([glw_diffs_t06mean, glw_diffs_t09mean, glw_diffs_t12mean, glw_diffs_t15mean, glw_diffs_t18mean, glw_diffs_t21mean, glw_diffs_t00mean, glw_diffs_t03mean])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 hourly means of swdown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.673996Z",
     "start_time": "2018-08-16T14:22:29.968Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "swdown_400_t00mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "swdown_400_t03mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "swdown_400_t06mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "swdown_400_t09mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "swdown_400_t12mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "swdown_400_t15mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "swdown_400_t18mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "swdown_400_t21mean = numpy.mean(swdown_400_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "swdown_1000_t00mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_00,:], axis=0)\n",
    "swdown_1000_t03mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_03,:], axis=0)\n",
    "swdown_1000_t06mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_06,:], axis=0)\n",
    "swdown_1000_t09mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_09,:], axis=0)\n",
    "swdown_1000_t12mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_12,:], axis=0)\n",
    "swdown_1000_t15mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_15,:], axis=0)\n",
    "swdown_1000_t18mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_18,:], axis=0)\n",
    "swdown_1000_t21mean = numpy.mean(swdown_1000_lat_mean[skip_time:][times_21,:], axis=0)\n",
    "\n",
    "swdown_400_array = numpy.array([swdown_400_t06mean, swdown_400_t09mean, swdown_400_t12mean, swdown_400_t15mean, swdown_400_t18mean, swdown_400_t21mean, swdown_400_t00mean, swdown_400_t03mean])\n",
    "swdown_1000_array = numpy.array([swdown_1000_t06mean, swdown_1000_t09mean, swdown_1000_t12mean, swdown_1000_t15mean, swdown_1000_t18mean, swdown_1000_t21mean, swdown_1000_t00mean, swdown_1000_t03mean])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.675368Z",
     "start_time": "2018-08-16T14:22:29.973Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate 3 hourly means for 400\n",
    "swdown_diffs_t00mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_00,:], axis=0)\n",
    "swdown_diffs_t03mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_03,:], axis=0)\n",
    "swdown_diffs_t06mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_06,:], axis=0)\n",
    "swdown_diffs_t09mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_09,:], axis=0)\n",
    "swdown_diffs_t12mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_12,:], axis=0)\n",
    "swdown_diffs_t15mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_15,:], axis=0)\n",
    "swdown_diffs_t18mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_18,:], axis=0)\n",
    "swdown_diffs_t21mean = numpy.mean(swdown_diffs_lat_mean[skip_time:,:][times_21,:], axis=0)\n",
    "\n",
    "swdown_diffs_array = numpy.array([swdown_diffs_t06mean, swdown_diffs_t09mean, swdown_diffs_t12mean, swdown_diffs_t15mean, swdown_diffs_t18mean, swdown_diffs_t21mean, swdown_diffs_t00mean, swdown_diffs_t03mean])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## create grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.676715Z",
     "start_time": "2018-08-16T14:22:29.977Z"
    }
   },
   "outputs": [],
   "source": [
    "xx,zz=numpy.meshgrid(xlong[0,:],geopot_400_lat_mean[0,:,:].mean(axis=1)/10000.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot fira minus fsa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.678078Z",
     "start_time": "2018-08-16T14:22:29.981Z"
    }
   },
   "outputs": [],
   "source": [
    "#fig,ax_array = mp.subplots(4,2)\n",
    "#fig.set_size_inches(7.5,6)\n",
    "fontsize=12\n",
    "\n",
    "fig = mp.figure(figsize=(7.5,6))\n",
    "\n",
    "ax_list = []\n",
    "for t in range(8):\n",
    "    ax=fig.add_subplot(4,2,t+1)\n",
    "    radiation_smoothed = pandas.Series(fira_diffs_array[t]-fsa_diffs_array[t]).rolling(window=3, center=True, min_periods=1).mean()\n",
    "    ax.plot(xx[0,:]+360., radiation_smoothed, c='firebrick')\n",
    "    ax.text(s='('+string.ascii_lowercase[t]+') hour '+hour_strings_local[t], x=0, y=1.02,\\\n",
    "            transform=ax.transAxes,fontsize=fontsize,ha='left',va='bottom')\n",
    "    ax.tick_params(labelsize=fontsize)\n",
    "    ax_list.append(ax)\n",
    "    ax.axhline(0, lw=1, c='0.1', ls='--', zorder=0)\n",
    "fig.tight_layout()\n",
    "\n",
    "fig.savefig('net_radiation_atmosphere_3hour.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot swdown minus glw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# plot net bottom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-08-16T14:27:50.679436Z",
     "start_time": "2018-08-16T14:22:29.988Z"
    }
   },
   "outputs": [],
   "source": [
    "#fig,ax_array = mp.subplots(4,2)\n",
    "#fig.set_size_inches(7.5,6)\n",
    "fontsize=12\n",
    "\n",
    "fig = mp.figure(figsize=(7.5,6))\n",
    "\n",
    "ax_list = []\n",
    "for t in range(8):\n",
    "    ax=fig.add_subplot(4,2,t+1)\n",
    "    radiation_smoothed = pandas.Series(net_bottom_diffs_array[t]).rolling(window=3, center=True, min_periods=1).mean()\n",
    "    ax.plot(xx[0,:]+360., radiation_smoothed/3600., c='firebrick')\n",
    "    ax.text(s='('+string.ascii_lowercase[t]+') hour '+hour_strings_local[t], x=0, y=1.02,\\\n",
    "            transform=ax.transAxes,fontsize=fontsize,ha='left',va='bottom')\n",
    "    ax.tick_params(labelsize=fontsize)\n",
    "    ax_list.append(ax)\n",
    "    ax.axhline(0, lw=1, c='0.1', ls='--', zorder=0)\n",
    "    if t in [0,2,4,6]:\n",
    "        ax.set_ylabel('W m$^{\\,-2}$', fontsize=fontsize)\n",
    "fig.tight_layout()\n",
    "\n",
    "fig.savefig('net_radiation_bottom_3hour.png', dpi=600, bbox_inches='tight', transparent=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}