[GIS] Export satellite images for an entire country from Google Earth Engine

google-earth-engine

How can I download all the satellite images covering an entire country(Kenya) for a period of years from Google Earth Engine? I drew a large geometry polygon over the entire country and then tried to get the images by using filterBounds() but doesnt seem to be working. What am I doing wrong?

var roi = /* color: #0b4a8b */ee.FeatureCollection(
    [ee.Feature(
        ee.Geometry.Polygon(
            [[[41.484375, -1.6477220517969224],
              [39.5068359375, 3.447624666646865],
              [35.2880859375, 4.981505049328245],
              [33.99169921875, 4.23685605976896],
              [34.98046875, 2.1088986592431382],
              [33.99169921875, -0.9447814006873896],
              [39.2431640625, -4.740675384778361]]]),
        {
          "system:index": "0"
        })]);



var image = ee.Image(ee.ImageCollection('MODIS/MOD09A1')
.filterBounds(roi)
.filterDate('2010-01-01', '2013-12-31')
.sort('CLOUD_COVER')
.first()
);
Map.addLayer(image, {bands: ['sur_refl_b01', 'sur_refl_b03', 
'sur_refl_b04'], max: 0.5, gamma: 2}, "Modis Image");

Best Answer

Your problem is that MODIS doesn't have CLOUD_COVER metadata, so filtering for it returns no image. The MODIS images you're accessing are global, so that sort of metadata doesn't make any sense.

Also, your scaling is off (you need to specify a min and the images are scaled by 0.0001.

var image = ee.Image(ee.ImageCollection('MODIS/MOD09A1')
  .filterBounds(roi)
  .filterDate('2010-01-01', '2013-12-31')
  .first());
Map.addLayer(image, {
  bands: ['sur_refl_b01', 'sur_refl_b04', 'sur_refl_b03'], min:0, max: 4000},
  "Modis Image");

Related Solutions

[GIS] How to export a set of images from Google Earth Engine

Well, the documentation (https://developers.google.com/earth-engine/client_server#client-and-server-functions) says that Export is a client-side function, and map() executes on the Server-side, and I think that is the problem. Although you'll find in the same documentation that client-side for loops ain't recommended, in this case that you want to go over a client-side function, I'd use it, like:

// Import points from fusion table into feature collection
var fc = ee.FeatureCollection('ft:1i3_jdpBy0lorCbZ7eNVnsGL3NaZWyDGRVCUrmKGl');
print(fc);

var pairImage = function(feature) {
  var point = ee.Geometry.Point([feature.get('lat'), 
                                 feature.get('lon')]);

  // USGS Landsat 8 TOA Reflectance (Orthorectified) Set
  var l8 = ee.ImageCollection('LANDSAT/LC8_L1T_TOA');

  // Only get images that intersect lat/lon point
  var spatialFiltered = l8.filterBounds(point);

  // Only get images for year of 2016
  var temporalFiltered = spatialFiltered.filterDate('2016-01-01', '2016-12-31');

  // This will sort from least to most cloudy.
  var sorted = temporalFiltered.sort('CLOUD_COVER');

  // Get the first (least cloudy) image.
  var scene = ee.Image(sorted.first()).select(['B4', 'B3', 'B2']);

  var visualization = scene.visualize({
    bands: ['B4', 'B3', 'B2'],
    max: 0.3
  });
  return visualization;
}

// Get all images of points
var images = fc.map(pairImage);

// list of images (client side)
var imgs = images.getInfo()["features"]

// loop on client side
for (var i = 0; i<imgs.length;i++) {
  var im = ee.Image(imgs[i]["id"]);
  Map.addLayer(im)
  Export.image.toDrive({
    image: im,
    description: 'true-color-composite',
    scale: 30
  });
}

If you have shapes, you could make the fusion table with those, but I assume you have a list of places with lat & long values.

[GIS] Export KML from google earth engine

You could do it directly in GEE:

var shape = out.reduceToVectors({
  geometry: geometry,
  crs: ee.Projection('EPSG:4326'),
  scale: 10,
  maxPixels: 1e13
})

Export.table.toDrive({
  collection: shape, 
  description: "yourDescription", 
  folder: "yourFolder", 
  fileNamePrefix:"yourName", 
  fileFormat:"KML"
})

But I do not recommend to do it directly like that because there are a lot of small areas and it'll take too long to process (may be it won't do it). You could process it a bit more in GEE, like filtering by count property or apply a focal_mode or something, or you could download the raster and do the process locally with a raster to vector algorithm.

Best Answer

Related Solutions

[GIS] How to export a set of images from Google Earth Engine

[GIS] Export KML from google earth engine

Related Question