1

I am struggling to find a best practice for exporting data from Google Earth Engine at a reduced resolution for further analysis outside of GEE, while maintaining similar results analysis in the code editor. My principle questions are when and how to define projection to maintain positional and area accuracy.

You have to project in order to reduce resolution to a specified scale, but do you also include projection/scale information on export? Do you reproject the raw data first, before reducing resolution, then project again to define a larger pixel size? Do you work with a snap raster before exporting? I have tried various ways of doing this and am so far unable to maintain similar summary statistics at a reduced resolution. I know reprojecting and adjusting resolution changes pixel values, but I think I should at least be able to get close. I would be open to javascript of python solutions, as I'm thinking of migrating workflow to the python API.

My workflow currently:

var reduced.image = RawImage.reduceResolution({
                 reducer: ee.Reducer.mean()
                 maxPixels: maxPixels //number of pixels to aggregate
                 }).reproject(crs, null, scale)//crs taken from desired NLCD projection//scale pixel size of intended image

//exporting
Export.image.toDrive({
  image: reduced.image,
  description: 'reduced.image',
  folder: 'reduced.image_folder',
  scale: scale,
  region: region,
  maxPixels:1e13,
  //crs: crs  //do I include crs and transform again here? 
  //crsTransform: crsTransform,
  //pyramidingPolicy:{".default":"mode"},
});

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.