docs/2_DownloadandENVI.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![UCL](images/ucl_logo.png)\n",
    "\n",
    "# Download and visualise data \n",
    "\n",
    "## Download"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, you will learn how to download full image data from a custom directory for this course. These are pre processed anual composites for use in ENVI. We will also explore the Shenzhen region using a web-based app developed using Google Earth Engine (GEE). GEE has access to a large number of Earth Observation and other datasets, and makes it easy to access these and do a some processing (at least for a limited spatial extent if you want high resolution data).\n",
    "\n",
    "You are provided with a web App built using GEE:\n",
    "\n",
    "This is available to you as a specific [App for the Shenzhen area](https://plewis.users.earthengine.app/view/shenzhen). The use of the App should be quite intuitive, but we will outline how to use it below.\n",
    "\n",
    "The app gives you access to 5 bands of Landsat surface reflectance data (see figure below), along with two vegetation indices, NDVI and NDWI. The datasets are annual composites, with each pixel the median of valid surface reflectance data for that year. \n",
    "\n",
    "This notebook has a few interactive elements you may wish to use in JupyterHub. To view these look [HERE (via nbviewer)](https://nbviewer.org/github/UCL-EO/GEOG0027_Coursework/blob/main/docs/2_DownloadandENVI.ipynb).\n",
    "\n",
    "(NOTE: If you don't want to use nbviewer - you can follow this  <a href=\"https://plewis.users.earthengine.app/view/shenzhen\" target=\"_blank\">link to the app instead</a>)\n",
    "\n",
    "First load the App:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"490\"\n",
       "            src=\"https://plewis.users.earthengine.app/view/shenzhen\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x7f8d8a6f5f28>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from IPython.display import IFrame\n",
    "IFrame('https://plewis.users.earthengine.app/view/shenzhen','100%',490)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualise time series\n",
    "\n",
    "You may have noticed a time series graph on the left-hand panel of the app. This appears when you click on any location in the image panel, and provides a visualisation of the time series of reflectance (and vegetation indices) for that location. After selecting a particular point on the map, wait for a few minutes and the time series will appear. You can pop the graph out to another browser tab (the grey square with arrow in the top left of the plot). \n",
    "\n",
    "\n",
    "It will look like this:\n",
    "\n",
    "![appgraph](images/appgraph.png)\n",
    "\n",
    "This is useful for giving you an idea of the land cover history of this location, so you might want to produce graphs for several locations with different cover types, to use in your report.\n",
    "\n",
    "You can directly save the plot (`Download png` button) or save the dataset (`Download csv` button). Or alternatively you can screenshot the plot, though be careful to keep it looking tidy.\n",
    "\n",
    "If you download the `csv` file, you can then produce your own plots of the data (for your report).\n",
    "\n",
    "The example below are given in Python, but you should use whatever graphing package you are familiar with. Please make sure that (unlike below, which are for illustration only) you put full titles, labels and legends on graphs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>system:time_start</th>\n",
       "      <th>B1_blue</th>\n",
       "      <th>B2_green</th>\n",
       "      <th>B3_red</th>\n",
       "      <th>B4_nir</th>\n",
       "      <th>B5_swir</th>\n",
       "      <th>B6_NDVI</th>\n",
       "      <th>B7_NDWI</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1986-06-06</td>\n",
       "      <td>0.102</td>\n",
       "      <td>0.175</td>\n",
       "      <td>0.238</td>\n",
       "      <td>0.347</td>\n",
       "      <td>0.388</td>\n",
       "      <td>0.187</td>\n",
       "      <td>-0.055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1987-06-06</td>\n",
       "      <td>0.103</td>\n",
       "      <td>0.177</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.299</td>\n",
       "      <td>0.412</td>\n",
       "      <td>0.136</td>\n",
       "      <td>-0.159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1988-06-06</td>\n",
       "      <td>0.101</td>\n",
       "      <td>0.157</td>\n",
       "      <td>0.206</td>\n",
       "      <td>0.296</td>\n",
       "      <td>0.315</td>\n",
       "      <td>0.181</td>\n",
       "      <td>-0.030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1989-06-06</td>\n",
       "      <td>0.104</td>\n",
       "      <td>0.158</td>\n",
       "      <td>0.192</td>\n",
       "      <td>0.262</td>\n",
       "      <td>0.295</td>\n",
       "      <td>0.154</td>\n",
       "      <td>-0.058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1990-06-06</td>\n",
       "      <td>0.154</td>\n",
       "      <td>0.221</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.330</td>\n",
       "      <td>0.381</td>\n",
       "      <td>0.097</td>\n",
       "      <td>-0.072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1991-06-06</td>\n",
       "      <td>0.133</td>\n",
       "      <td>0.187</td>\n",
       "      <td>0.211</td>\n",
       "      <td>0.260</td>\n",
       "      <td>0.300</td>\n",
       "      <td>0.105</td>\n",
       "      <td>-0.072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1992-06-06</td>\n",
       "      <td>0.145</td>\n",
       "      <td>0.185</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.239</td>\n",
       "      <td>0.295</td>\n",
       "      <td>0.078</td>\n",
       "      <td>-0.104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1993-06-06</td>\n",
       "      <td>0.144</td>\n",
       "      <td>0.188</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.249</td>\n",
       "      <td>0.293</td>\n",
       "      <td>0.098</td>\n",
       "      <td>-0.081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1994-06-06</td>\n",
       "      <td>0.111</td>\n",
       "      <td>0.139</td>\n",
       "      <td>0.149</td>\n",
       "      <td>0.181</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.098</td>\n",
       "      <td>-0.114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1995-06-06</td>\n",
       "      <td>0.114</td>\n",
       "      <td>0.136</td>\n",
       "      <td>0.137</td>\n",
       "      <td>0.170</td>\n",
       "      <td>0.213</td>\n",
       "      <td>0.109</td>\n",
       "      <td>-0.111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1996-06-06</td>\n",
       "      <td>0.107</td>\n",
       "      <td>0.132</td>\n",
       "      <td>0.137</td>\n",
       "      <td>0.159</td>\n",
       "      <td>0.192</td>\n",
       "      <td>0.074</td>\n",
       "      <td>-0.095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1997-06-06</td>\n",
       "      <td>0.098</td>\n",
       "      <td>0.121</td>\n",
       "      <td>0.132</td>\n",
       "      <td>0.162</td>\n",
       "      <td>0.190</td>\n",
       "      <td>0.104</td>\n",
       "      <td>-0.079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1998-06-06</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.160</td>\n",
       "      <td>0.183</td>\n",
       "      <td>0.215</td>\n",
       "      <td>0.064</td>\n",
       "      <td>-0.081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1999-06-06</td>\n",
       "      <td>0.123</td>\n",
       "      <td>0.152</td>\n",
       "      <td>0.167</td>\n",
       "      <td>0.183</td>\n",
       "      <td>0.200</td>\n",
       "      <td>0.045</td>\n",
       "      <td>-0.045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2000-06-06</td>\n",
       "      <td>0.134</td>\n",
       "      <td>0.153</td>\n",
       "      <td>0.157</td>\n",
       "      <td>0.167</td>\n",
       "      <td>0.203</td>\n",
       "      <td>0.031</td>\n",
       "      <td>-0.096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2001-06-06</td>\n",
       "      <td>0.110</td>\n",
       "      <td>0.131</td>\n",
       "      <td>0.140</td>\n",
       "      <td>0.152</td>\n",
       "      <td>0.166</td>\n",
       "      <td>0.038</td>\n",
       "      <td>-0.045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2002-06-06</td>\n",
       "      <td>0.115</td>\n",
       "      <td>0.132</td>\n",
       "      <td>0.134</td>\n",
       "      <td>0.146</td>\n",
       "      <td>0.138</td>\n",
       "      <td>0.041</td>\n",
       "      <td>0.027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2003-06-06</td>\n",
       "      <td>0.113</td>\n",
       "      <td>0.128</td>\n",
       "      <td>0.138</td>\n",
       "      <td>0.153</td>\n",
       "      <td>0.152</td>\n",
       "      <td>0.054</td>\n",
       "      <td>0.004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2004-06-06</td>\n",
       "      <td>0.133</td>\n",
       "      <td>0.169</td>\n",
       "      <td>0.186</td>\n",
       "      <td>0.218</td>\n",
       "      <td>0.257</td>\n",
       "      <td>0.079</td>\n",
       "      <td>-0.082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2005-06-06</td>\n",
       "      <td>0.140</td>\n",
       "      <td>0.181</td>\n",
       "      <td>0.199</td>\n",
       "      <td>0.233</td>\n",
       "      <td>0.257</td>\n",
       "      <td>0.079</td>\n",
       "      <td>-0.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2006-06-06</td>\n",
       "      <td>0.141</td>\n",
       "      <td>0.171</td>\n",
       "      <td>0.178</td>\n",
       "      <td>0.221</td>\n",
       "      <td>0.226</td>\n",
       "      <td>0.106</td>\n",
       "      <td>-0.012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2007-06-06</td>\n",
       "      <td>0.141</td>\n",
       "      <td>0.186</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.242</td>\n",
       "      <td>0.257</td>\n",
       "      <td>0.084</td>\n",
       "      <td>-0.029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2008-06-06</td>\n",
       "      <td>0.154</td>\n",
       "      <td>0.183</td>\n",
       "      <td>0.186</td>\n",
       "      <td>0.235</td>\n",
       "      <td>0.232</td>\n",
       "      <td>0.117</td>\n",
       "      <td>0.007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2009-06-06</td>\n",
       "      <td>0.145</td>\n",
       "      <td>0.170</td>\n",
       "      <td>0.180</td>\n",
       "      <td>0.211</td>\n",
       "      <td>0.217</td>\n",
       "      <td>0.078</td>\n",
       "      <td>-0.014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2010-06-06</td>\n",
       "      <td>0.151</td>\n",
       "      <td>0.180</td>\n",
       "      <td>0.184</td>\n",
       "      <td>0.213</td>\n",
       "      <td>0.228</td>\n",
       "      <td>0.073</td>\n",
       "      <td>-0.035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2011-06-06</td>\n",
       "      <td>0.144</td>\n",
       "      <td>0.169</td>\n",
       "      <td>0.175</td>\n",
       "      <td>0.202</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.072</td>\n",
       "      <td>-0.059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2012-06-06</td>\n",
       "      <td>0.144</td>\n",
       "      <td>0.169</td>\n",
       "      <td>0.175</td>\n",
       "      <td>0.202</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.072</td>\n",
       "      <td>-0.059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2013-06-06</td>\n",
       "      <td>0.127</td>\n",
       "      <td>0.156</td>\n",
       "      <td>0.165</td>\n",
       "      <td>0.216</td>\n",
       "      <td>0.215</td>\n",
       "      <td>0.135</td>\n",
       "      <td>0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2014-06-06</td>\n",
       "      <td>0.114</td>\n",
       "      <td>0.145</td>\n",
       "      <td>0.155</td>\n",
       "      <td>0.202</td>\n",
       "      <td>0.197</td>\n",
       "      <td>0.131</td>\n",
       "      <td>0.013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>2015-06-06</td>\n",
       "      <td>0.128</td>\n",
       "      <td>0.158</td>\n",
       "      <td>0.169</td>\n",
       "      <td>0.226</td>\n",
       "      <td>0.240</td>\n",
       "      <td>0.144</td>\n",
       "      <td>-0.029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>2016-06-06</td>\n",
       "      <td>0.117</td>\n",
       "      <td>0.146</td>\n",
       "      <td>0.157</td>\n",
       "      <td>0.221</td>\n",
       "      <td>0.240</td>\n",
       "      <td>0.168</td>\n",
       "      <td>-0.040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>2017-06-06</td>\n",
       "      <td>0.123</td>\n",
       "      <td>0.153</td>\n",
       "      <td>0.161</td>\n",
       "      <td>0.209</td>\n",
       "      <td>0.209</td>\n",
       "      <td>0.129</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>2018-06-06</td>\n",
       "      <td>0.140</td>\n",
       "      <td>0.176</td>\n",
       "      <td>0.188</td>\n",
       "      <td>0.242</td>\n",
       "      <td>0.253</td>\n",
       "      <td>0.125</td>\n",
       "      <td>-0.023</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   system:time_start  B1_blue  B2_green  B3_red  B4_nir  B5_swir  B6_NDVI  \\\n",
       "0         1986-06-06    0.102     0.175   0.238   0.347    0.388    0.187   \n",
       "1         1987-06-06    0.103     0.177   0.227   0.299    0.412    0.136   \n",
       "2         1988-06-06    0.101     0.157   0.206   0.296    0.315    0.181   \n",
       "3         1989-06-06    0.104     0.158   0.192   0.262    0.295    0.154   \n",
       "4         1990-06-06    0.154     0.221   0.272   0.330    0.381    0.097   \n",
       "5         1991-06-06    0.133     0.187   0.211   0.260    0.300    0.105   \n",
       "6         1992-06-06    0.145     0.185   0.205   0.239    0.295    0.078   \n",
       "7         1993-06-06    0.144     0.188   0.205   0.249    0.293    0.098   \n",
       "8         1994-06-06    0.111     0.139   0.149   0.181    0.227    0.098   \n",
       "9         1995-06-06    0.114     0.136   0.137   0.170    0.213    0.109   \n",
       "10        1996-06-06    0.107     0.132   0.137   0.159    0.192    0.074   \n",
       "11        1997-06-06    0.098     0.121   0.132   0.162    0.190    0.104   \n",
       "12        1998-06-06    0.125     0.150   0.160   0.183    0.215    0.064   \n",
       "13        1999-06-06    0.123     0.152   0.167   0.183    0.200    0.045   \n",
       "14        2000-06-06    0.134     0.153   0.157   0.167    0.203    0.031   \n",
       "15        2001-06-06    0.110     0.131   0.140   0.152    0.166    0.038   \n",
       "16        2002-06-06    0.115     0.132   0.134   0.146    0.138    0.041   \n",
       "17        2003-06-06    0.113     0.128   0.138   0.153    0.152    0.054   \n",
       "18        2004-06-06    0.133     0.169   0.186   0.218    0.257    0.079   \n",
       "19        2005-06-06    0.140     0.181   0.199   0.233    0.257    0.079   \n",
       "20        2006-06-06    0.141     0.171   0.178   0.221    0.226    0.106   \n",
       "21        2007-06-06    0.141     0.186   0.205   0.242    0.257    0.084   \n",
       "22        2008-06-06    0.154     0.183   0.186   0.235    0.232    0.117   \n",
       "23        2009-06-06    0.145     0.170   0.180   0.211    0.217    0.078   \n",
       "24        2010-06-06    0.151     0.180   0.184   0.213    0.228    0.073   \n",
       "25        2011-06-06    0.144     0.169   0.175   0.202    0.227    0.072   \n",
       "26        2012-06-06    0.144     0.169   0.175   0.202    0.227    0.072   \n",
       "27        2013-06-06    0.127     0.156   0.165   0.216    0.215    0.135   \n",
       "28        2014-06-06    0.114     0.145   0.155   0.202    0.197    0.131   \n",
       "29        2015-06-06    0.128     0.158   0.169   0.226    0.240    0.144   \n",
       "30        2016-06-06    0.117     0.146   0.157   0.221    0.240    0.168   \n",
       "31        2017-06-06    0.123     0.153   0.161   0.209    0.209    0.129   \n",
       "32        2018-06-06    0.140     0.176   0.188   0.242    0.253    0.125   \n",
       "\n",
       "    B7_NDWI  \n",
       "0    -0.055  \n",
       "1    -0.159  \n",
       "2    -0.030  \n",
       "3    -0.058  \n",
       "4    -0.072  \n",
       "5    -0.072  \n",
       "6    -0.104  \n",
       "7    -0.081  \n",
       "8    -0.114  \n",
       "9    -0.111  \n",
       "10   -0.095  \n",
       "11   -0.079  \n",
       "12   -0.081  \n",
       "13   -0.045  \n",
       "14   -0.096  \n",
       "15   -0.045  \n",
       "16    0.027  \n",
       "17    0.004  \n",
       "18   -0.082  \n",
       "19   -0.050  \n",
       "20   -0.012  \n",
       "21   -0.029  \n",
       "22    0.007  \n",
       "23   -0.014  \n",
       "24   -0.035  \n",
       "25   -0.059  \n",
       "26   -0.059  \n",
       "27    0.003  \n",
       "28    0.013  \n",
       "29   -0.029  \n",
       "30   -0.040  \n",
       "31    0.001  \n",
       "32   -0.023  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pylab as plt\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "#### edit this line of code\n",
    "d = pd.read_csv('files/ee-chart.csv')\n",
    "d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_=d.plot(figsize=(10,4),title='Example data plot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Full Composite Images\n",
    "\n",
    "The composite images for the Shenzhen region have been pre-processed for you and can be downloaded using the terminal app. To use with `ENVI` you'll probably want to use the geography UNIX machines, but alternatively you may wish to work on your own computer or UCL Desktop. You will want to create a new directory to store your new image data for the coursework. Remember the shell commands from previous lessons, particularly the \"make directory\" `mkdir` command. \n",
    "\n",
    "Once you've set the directory, set the year you want by modifying and running the below code (this year will not work, no data is available for this year as LandSat hadn't launched yet!):\n",
    "\n",
    "`year=\"1969\"`\n",
    "\n",
    "Then run \n",
    "\n",
    "`cp -r /archive/rsu_raid_0/plewis/public_html/GEOG0027/$year .`\n",
    "\n",
    "\n",
    "The image data can now be opened using ENVI. The file to select using the data manager is just titled with the year only. If other years are required for the analysis you wish to perform, then repeat the two previous commands, with the year adjusted.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualise spatial data\n",
    "\n",
    "You should be able to conveniently explore the spatial datasets in the GEE App. This is particularly useful for data interpretation (e.g. for training or validation) as you can have a map or high resolution image as a background.\n",
    "\n",
    "You can also load multiple years of data, and visualise the dynamics.\n",
    "\n",
    "Other than that, you will probably want to load the datasets into `envi` for visualisation. If you have correctly stored the data from the `wget` command, you should have in your data directory a series of at least one sub-directory named by year, such as:\n",
    "\n",
    "    1986\t1989\t1992\t1995\t1998\t2001\t2004\t2007\t2010\t2013\t2016\t2019\n",
    "    1987\t1990\t1993\t1996\t1999\t2002\t2005\t2008\t2011\t2014\t2017\n",
    "    1988\t1991\t1994\t1997\t2000\t2003\t2006\t2009\t2012\t2015\t2018\n",
    "    \n",
    "Inside each of these, you should see files such as the following:\n",
    "\n",
    "    22014.constant.tif\t   2014.swir.tif\n",
    "    2014.NDVI.tfw\t\t\t2014.green.tfw\t\t\t\n",
    "    2014.NDVI.tif\t\t\t2014.green.tif\t\t\t\n",
    "    2014.NDWI.tfw\t\t\t2014.nir.tfw\t\t\t\n",
    "    2014.NDWI.tif\t\t\t2014.nir.tif\t\t\t\n",
    "    2014.blue.tfw\t\t\t2014.red.tfw\t\t\t\n",
    "    2014.blue.tif\t\t\t2014.red.tif\t\t\t\n",
    "    2014.constant.tfw\t\t2014.swir.tfw\n",
    "    2014_Shenzhen            2014_Shenzhen.hdr\n",
    "\n",
    "![loaded](images/loaded.jpeg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Masking Data\n",
    "    \n",
    "Using  file `2014_Shenzhen` as an example we can use the mask in order to calculate pixel statistics.\n",
    "\n",
    "![2014_load](images/2014_load.png)\n",
    "\n",
    "![2014_view](images/2014_view.png)\n",
    "\n",
    "One further file that may be of interest is the file `2014.constant.tif` which you can use as a data mask if you like. The mask has a value of zero for Shenzhen and `no data` outside of that.\n",
    "\n",
    "![mask_image](images/mask.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To use the above mask as part of a classification it needs ot be first converted.\n",
    "Use `/Raster Management/Build Raster Mask`\n",
    "Select the mask .tif from above. Now use the following options:\n",
    "\n",
    "![mask_options](images/ENVI_mask_options.png)\n",
    "\n",
    "You need to change \"Mask Finite Values\" (select the image again)\n",
    "Then change selected areas \"off\"\n",
    "\n",
    "Save the results to memory or file (you may wish to use it on another image).\n",
    "\n",
    "The new image of [0,1] can now be slected in a classification method from the `Toolbox`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You should load and explore the datasets you have downloaded. Provide examples of e.g. interesting spectral profiles, transects, histograms or scatter plots for your report -- things that you would normally produce as part of a data exploration exercise.\n",
    "\n",
    "the histograms are of interest (think about what this means for classification)\n",
    "\n",
    "![histo2014](images/histo2014.png)\n",
    "\n",
    "and the scatterplots in many ways even more interesting:\n",
    "\n",
    "![2014scatter](images/2014scatter.png)\n",
    "\n",
    "\n",
    "You will find it particularly interesting to examine scatter plots of NDVI and NDWI for this dataset as you should see that a pretty good classification can be achieved with just these two bands of data. An advantage of this would be that the processing should be faster (fewer bands), but also that it is straightforward to visualise and appreciate the clusters that will be used to describe the land cover classes.\n",
    "\n",
    "These are not the only 'bands' to explore though! Collect plenty of images of scatterplotas, histograms and images for your write up to illustrate particular things you find. **Don't** just put images in the report with no reference to what you are using them to show!\n",
    "\n",
    "You should spend some time exploring these scatterplots for the different years data: are the patterns consistent? (show some evidence). Can you simply visualise (and explain) the clusters and their relationship to land cover types? In exploring the datasets you should save interesting plots to use in your write-up -- anything that helps you explain what is going on is of value!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Supervised and Unsupervised classification\n",
    "\n",
    "Looking at the write specifications for this part of the project:\n",
    "\n",
    "#### Part 2a:  Data Exploration and Classification Theory\n",
    "\n",
    "[30% of marks in Part 2a]\n",
    "\n",
    "Select data for a single year from [here](http://www2.geog.ucl.ac.uk/~plewis/GEOG0027/).\n",
    "\n",
    "For the data selected for the single year:\n",
    "\n",
    "* Choose one supervised and one unsupervised classification approach (we recommend [Maximum Likelihood](https://www.harrisgeospatial.com/docs/MaximumLikelihood.html) and [ISOData](https://www.harrisgeospatial.com/docs/ISODATAClassification.html));\n",
    "* Theory: describe how the approaches work (noting similarities and differences) and relate this to the information content of your data. Cite literature as appropriate. [10%]\n",
    "* Perform a supervised classification **and** an unsupervised classification using `envi`, relating the training information (e.g. class seperability) to the material presented above;\n",
    "* Present the results of the classification and assess the accuracy of these classifications; [15%]\n",
    "* Discuss the issues raised and how this might translate to unsupervised classification of the whole time series. [5%]\n",
    "\n",
    "\n",
    "We can see that we need to perform both supervised and unsupervised classifications. The supervised classification we have performed in previous lessons. Unsupervised classification algorithms are found in the `ENVI` toolboxes too. These can be run similarly, although they require no ROI. You will want to set and ajust the number of classes in order to investigate the classifier accuracy. Particularly if you wish to use this method when using GEE for the whole timeseries. An example classification:\n",
    "\n",
    "![lut3](images/lut3.png)\n",
    "\n",
    "which is an appropriate interpretation of the clusters.\n",
    "\n",
    "Of course, we might decide to use more clusters (7 here) to get more subtle interpretations, but we will find the classes rather harder to interpret. Possibly the best answer lies somewhere in between. You will need to explore that and come to some (justified) conclusion.\n",
    "\n",
    "![cluster1](images/cluster1.png)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Accuracy Assessment\n",
    "\n",
    "It is crucial to show an assessment of classifcation accuracy in order to validate the work using ENVI. This assessment will then be used to validate the decisions made during the time series and  modelling sections. \n",
    "\n",
    "## Using Ground Truth ROIs\n",
    "\n",
    "First a supervised classification can be assed easily using the `Toolbox - Classification - Post Classification - Confusion Matrix From Ground Truth ROIs`. This will by default use the ROIs used to train the classification. Why is this assessment limited? Can you do a similar assessment using an independent set of ROIs? (Hint: make a double set of regions and reasign when setting the matched classes).\n",
    "\n",
    "The output is a [confusion matrix](http://www.exelisvis.com/docs/CalculatingConfusionMatrices.html). This records all the correct pixels (and incorrect) classifcation pixels and shows the overall classification accuracy?\n",
    "\n",
    "![](images/confusion_mat.png)\n",
    "\n",
    "## Using Ground Truth Image\n",
    "\n",
    "After making a \"good\" supervised classification, this can be used to assess an unsupervised classification. This is similar to above, but using `Toolbox - Classification - Post Classification - Confusion Matrix From Ground Truth Image`. The classes need to matched manually as seen below.\n",
    "\n",
    "![](images/GT_image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pixel counting\n",
    "\n",
    "Finally, you will need to provide a count of the total area of each class, with each pixel being 30 m x 30 m. To do this, you need to count how many pixels are in each class.\n",
    "\n",
    "You can do this in `envi` following the menu items (right hand panel) `Classification -> Post Classification -> Class Statistics`.\n",
    "\n",
    "This will produce the data you need to track land cover for each date. Make a note of the pixel counts for each year.\n",
    "\n",
    "![stats](images/stats.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.12"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": false,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": false,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}