notebooks/venus.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Trick to import local code:\n",
    "import sys\n",
    "#sys.path = ['/home/mp/michelj/src/sensorsio/src'] + sys.path\n",
    "sys.path = ['/home/uz/vinascj/srcsensorio/sensorsio/src'] + sys.path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import rasterio as rio\n",
    "import numpy as np\n",
    "from matplotlib import pyplot\n",
    "from sensorsio import venus, utils"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Constructor of ```venus.Venus``` class reads the path to the ***Venus L2A*** product:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#dataset = venus.Venus('/datalake/VENUS-L2A/SUDOUE-1/2020/02/08/VENUS-XS_20200208-105747-000_L2A_SUDOUE-1_C_V2-2/')\n",
    "dataset = venus.Venus('/datalake/VENUS-L2A/GERAWH-1/2020/01/02/VENUS-XS_20200102-003542-000_L2A_GERAWH-1_C_V2-2/')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Printing the returned objects gives basic information on the product:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VENUS, 2020-01-02 00:00:00, GERAWH-1\n"
     ]
    }
   ],
   "source": [
    "print(dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Among the object attributes, one can find information on acquisition date as a ```datetime.datetime``` object, acquisition year, and day of year:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "date: 2020-01-02 00:00:00, year: 2020, day of year: 2\n"
     ]
    }
   ],
   "source": [
    "print(f\"date: {dataset.date}, year: {dataset.year}, day of year: {dataset.day_of_year}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One can also read the satellite id, the site code-name and the corresponding coordinate reference system:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Satellite: VENUS, Observed site: dataset.site, CRS: EPSG:32755\n"
     ]
    }
   ],
   "source": [
    "print(f\"Satellite: {dataset.satellite.value}, Observed site: dataset.site, CRS: {dataset.crs}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Last, the object also contains the geographic bounds expressed in the given CRS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Product bounds: BoundingBox(left=301645.0, bottom=5920255.0, right=345030.0, top=5965215.0)\n"
     ]
    }
   ],
   "source": [
    "print(f\"Product bounds: {dataset.bounds}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Basic reading"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading to numpy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selecting bands"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```venus.Venus.GROUP_5M``` is a convenient variable to select all 5m bands:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<Band.B1: 'B1'>,\n",
       " <Band.B2: 'B2'>,\n",
       " <Band.B3: 'B3'>,\n",
       " <Band.B4: 'B4'>,\n",
       " <Band.B5: 'B5'>,\n",
       " <Band.B6: 'B6'>,\n",
       " <Band.B7: 'B7'>,\n",
       " <Band.B8: 'B8'>,\n",
       " <Band.B9: 'B9'>,\n",
       " <Band.B10: 'B10'>,\n",
       " <Band.B11: 'B11'>,\n",
       " <Band.B12: 'B12'>]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "venus.Venus.GROUP_5M"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One can of course make her own selection of bands:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_bands = [venus.Venus.B2, venus.Venus.B8]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A first read"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the following we will use the ``venus.Venus.GROUP_5M``` to read all 5m bands to a numpy array:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "bands, masks, xcoords, ycoords, crs = dataset.read_as_numpy(venus.Venus.GROUP_5M)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first returned value is a ```numpy.ndarray``` with shape ```[bands, height, width]``` containing pixels reflectances:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12, 4496, 4338)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bands.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The second returned value is a ```numpy.ndarray``` with shape ```[masks, height, width]``` containing masks values (by default, the 4 masks are read):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4, 4496, 4338)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "masks.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3rd and 4th returned values are the row and col coordinates expressed in the given crs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([301645., 301655., 301665., ..., 344995., 345005., 345015.]),\n",
       " array([5965215., 5965205., 5965195., ..., 5920285., 5920275., 5920265.]))"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xcoords, ycoords"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Last returned value is the crs as a string:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CRS.from_epsg(32755)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "crs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By default, pixels values are scaled to $[0,1]$, and missing values are set to ```numpy.nan``` (hence the use of ```nanmin``` and ```nanmax``` in the following):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([-11.166, -11.124, -11.105, -11.079, -11.066, -11.066, -11.06 ,\n",
       "        -11.056, -11.053, -11.052, -11.048, -11.042], dtype=float32),\n",
       " array([1.951, 1.592, 1.487, 1.327, 1.264, 1.257, 1.244, 1.218, 1.213,\n",
       "        1.24 , 1.271, 1.262], dtype=float32))"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.nanmin(bands, axis=(1,2)), np.nanmax(bands, axis=(1,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Displaying the image"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```utils``` module provides a convenient function that will prepare the array for display with ```matplotlib```. The ```bands ``` argument indicates which slice of the first dimension to map to the red, green and blue channels. ```dmin``` and ```dmax``` arguments allows to select the displayed range for each band. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2b1661a666a0>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "arr_rgb, dmin, dmax = utils.rgb_render(bands, bands=[2,1,0], \n",
    "                                 dmin=np.array([0., 0., 0.]), \n",
    "                                 dmax=np.array([0.2,0.2,0.2]))\n",
    "pyplot.imshow(arr_rgb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading to xarray Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```xarray``` library provides a nice decoration of ```numpy``` arrays for even more abstraction:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "xrds = dataset.read_as_xarray(venus.Venus.GROUP_5M)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<xarray.Dataset>\n",
      "Dimensions:  (t: 1, x: 4338, y: 4496)\n",
      "Coordinates:\n",
      "  * t        (t) datetime64[ns] 2020-01-02\n",
      "  * x        (x) float64 3.016e+05 3.017e+05 3.017e+05 ... 3.45e+05 3.45e+05\n",
      "  * y        (y) float64 5.965e+06 5.965e+06 5.965e+06 ... 5.92e+06 5.92e+06\n",
      "Data variables:\n",
      "    B1       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    SAT      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    CLM      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    EDG      (t, y, x) uint8 1 1 1 1 1 1 1 1 1 1 1 1 ... 1 1 1 1 1 1 1 1 1 1 1 1\n",
      "    MG2      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 1 1 1 1 1 1 1 1 1 1 1 1\n",
      "    B2       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B3       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B4       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B5       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B6       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B7       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B8       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B9       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B10      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B11      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "    B12      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
      "Attributes:\n",
      "    site:     GERAWH-1\n",
      "    type:     FRE\n",
      "    crs:      EPSG:32755\n"
     ]
    }
   ],
   "source": [
    "print(xrds)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The created datasets contains several variables, with the requested bands and masks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Data variables:\n",
       "    B1       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    SAT      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
       "    CLM      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
       "    EDG      (t, y, x) uint8 1 1 1 1 1 1 1 1 1 1 1 1 ... 1 1 1 1 1 1 1 1 1 1 1 1\n",
       "    MG2      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 1 1 1 1 1 1 1 1 1 1 1 1\n",
       "    B2       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B3       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B4       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B5       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B6       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B7       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B8       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B9       (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B10      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B11      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0\n",
       "    B12      (t, y, x) float32 -10.0 -10.0 -10.0 -10.0 ... -10.0 -10.0 -10.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xrds.data_vars"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Those variables are indexed with 3 sets of coordinates: x coordinates, y coordinates and time coordinates:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Coordinates:\n",
       "  * t        (t) datetime64[ns] 2020-01-02\n",
       "  * x        (x) float64 3.016e+05 3.017e+05 3.017e+05 ... 3.45e+05 3.45e+05\n",
       "  * y        (y) float64 5.965e+06 5.965e+06 5.965e+06 ... 5.92e+06 5.92e+06"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xrds.coords"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The time coordinate has only a single value (since a dataset represents one acquisition date) but it would be very easy to stack several datasets together along the time dimension."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is rather easy to use ```xarray``` to derive new variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<xarray.DataArray (t: 1, y: 4496, x: 4338)>\n",
      "array([[[-0., -0., -0., ..., -0., -0., -0.],\n",
      "        [-0., -0., -0., ..., -0., -0., -0.],\n",
      "        [-0., -0., -0., ..., -0., -0., -0.],\n",
      "        ...,\n",
      "        [-0., -0., -0., ..., -0., -0., -0.],\n",
      "        [-0., -0., -0., ..., -0., -0., -0.],\n",
      "        [-0., -0., -0., ..., -0., -0., -0.]]], dtype=float32)\n",
      "Coordinates:\n",
      "  * t        (t) datetime64[ns] 2020-01-02\n",
      "  * x        (x) float64 3.016e+05 3.017e+05 3.017e+05 ... 3.45e+05 3.45e+05\n",
      "  * y        (y) float64 5.965e+06 5.965e+06 5.965e+06 ... 5.92e+06 5.92e+06\n"
     ]
    }
   ],
   "source": [
    "ndvi = (xrds.B8-xrds.B4)/(1e-6 + xrds.B8 + xrds.B4)\n",
    "print(ndvi)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Of course, one can always access the underlying ```numpy``` arrays:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[-0., -0., -0., ..., -0., -0., -0.],\n",
       "        [-0., -0., -0., ..., -0., -0., -0.],\n",
       "        [-0., -0., -0., ..., -0., -0., -0.],\n",
       "        ...,\n",
       "        [-0., -0., -0., ..., -0., -0., -0.],\n",
       "        [-0., -0., -0., ..., -0., -0., -0.],\n",
       "        [-0., -0., -0., ..., -0., -0., -0.]]], dtype=float32)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ndvi.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Advanced read"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, examples will be demonstrated with ```read_as_xarray(...)``` but one can achieve similar operations with ```read_as_numpy(...)```."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading a spatial subset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```region``` argument allows to read a spatial subset of the data by providing ```(startx, starty, endx, endy)``` integer pixel coordinates tuple. Beware that those coordinates relate to the coordinates in the array after resampling or reprojection, and not input image coordinates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<xarray.Dataset>\n",
      "Dimensions:  (t: 1, x: 200, y: 200)\n",
      "Coordinates:\n",
      "  * t        (t) datetime64[ns] 2020-01-02\n",
      "  * x        (x) float64 3.116e+05 3.117e+05 3.117e+05 ... 3.136e+05 3.136e+05\n",
      "  * y        (y) float64 5.955e+06 5.955e+06 5.955e+06 ... 5.953e+06 5.953e+06\n",
      "Data variables:\n",
      "    B1       (t, y, x) float32 0.025 0.022 0.029 0.027 ... 0.079 0.083 0.088\n",
      "    SAT      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    CLM      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    EDG      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    MG2      (t, y, x) uint8 0 0 0 0 0 0 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0 0 0\n",
      "    B2       (t, y, x) float32 0.041 0.04 0.044 0.043 ... 0.13 0.126 0.121 0.121\n",
      "    B3       (t, y, x) float32 0.046 0.041 0.045 0.048 ... 0.162 0.157 0.158\n",
      "    B4       (t, y, x) float32 0.061 0.057 0.06 0.061 ... 0.226 0.218 0.219\n",
      "    B5       (t, y, x) float32 0.069 0.065 0.07 0.072 ... 0.295 0.286 0.28 0.281\n",
      "    B6       (t, y, x) float32 0.07 0.065 0.071 0.074 ... 0.295 0.287 0.28 0.282\n",
      "    B7       (t, y, x) float32 0.08 0.074 0.081 0.084 ... 0.333 0.326 0.325\n",
      "    B8       (t, y, x) float32 0.098 0.093 0.097 0.099 ... 0.353 0.347 0.348\n",
      "    B9       (t, y, x) float32 0.142 0.138 0.131 0.132 ... 0.372 0.365 0.368\n",
      "    B10      (t, y, x) float32 0.159 0.154 0.147 0.148 ... 0.393 0.385 0.388\n",
      "    B11      (t, y, x) float32 0.192 0.188 0.181 0.181 ... 0.431 0.424 0.428\n",
      "    B12      (t, y, x) float32 0.197 0.195 0.187 0.188 ... 0.427 0.424 0.426\n",
      "Attributes:\n",
      "    site:     GERAWH-1\n",
      "    type:     FRE\n",
      "    crs:      EPSG:32755\n"
     ]
    }
   ],
   "source": [
    "print(dataset.read_as_xarray(venus.Venus.GROUP_5M, \n",
    "                             region=(1000,1000,1200,1200)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```region``` argument also accepts a ```rio.coords.BoundingBox``` object representing a geo-referenced bounding-box in the given ```crs``` (default to input image ```crs```):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "georoi = rio.coords.BoundingBox(399960.0, 4690200.0, 401960.0, 4692200.0) # left, bottom, right, top"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<xarray.Dataset>\n",
      "Dimensions:  (t: 1, x: 0, y: 0)\n",
      "Coordinates:\n",
      "  * t        (t) datetime64[ns] 2020-01-02\n",
      "  * x        (x) float64 \n",
      "  * y        (y) float64 \n",
      "Data variables:\n",
      "    B1       (t, y, x) float32 \n",
      "    SAT      (t, y, x) uint8 \n",
      "    CLM      (t, y, x) uint8 \n",
      "    EDG      (t, y, x) uint8 \n",
      "    MG2      (t, y, x) uint8 \n",
      "    B2       (t, y, x) float32 \n",
      "    B3       (t, y, x) float32 \n",
      "    B4       (t, y, x) float32 \n",
      "    B5       (t, y, x) float32 \n",
      "    B6       (t, y, x) float32 \n",
      "    B7       (t, y, x) float32 \n",
      "    B8       (t, y, x) float32 \n",
      "    B9       (t, y, x) float32 \n",
      "    B10      (t, y, x) float32 \n",
      "    B11      (t, y, x) float32 \n",
      "    B12      (t, y, x) float32 \n",
      "Attributes:\n",
      "    site:     GERAWH-1\n",
      "    type:     FRE\n",
      "    crs:      EPSG:32755\n"
     ]
    }
   ],
   "source": [
    "print(dataset.read_as_xarray(venus.Venus.GROUP_5M,\n",
    "                             region=georoi))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Changing projection and sampling grid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ```crs``` argument allows to resample the data into a different Coordinates Reference System  upon reading:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(dataset.read_as_xarray(venus.Venus.GROUP_5M, crs='epsg:2154')) # Lambert93 EPSG code"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One can also change the image sampling grid by overloading the geographic bounds:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "newbounds = rio.coords.BoundingBox(399965.0, 4690205.0, 401965.0, 4692205.0) # left, bottom, right, top"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(dataset.read_as_xarray(venus.Venus.GROUP_5M, bounds=newbounds))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Note:*** This is different from using the region keyword which only does a subsetting. The ```bounds``` argument allows to modify the image sampling grid to match the grid of another image for instance."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Pangeo 2020-12",
   "language": "python",
   "name": "pangeo_202012"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}