Google Earth Engine – Solve ‘Collection Query Aborted After Accumulating Over 5000 Elements’

google-earth-engineradarsamplingsentinel-1sentinel-2

I am trying to sample image and to get scatter plot of its' two bands- NDVI and db from senttinel 1.
For some reason, probably because of the size of the shapefile, I get the error:

Error generating chart: Collection query aborted after accumulating
over 5000 elements.

Collection query aborted after accumulating over 5000 elements

That's happens after I try to sample my image using this:

// Generate a sample of points within the region
var sample = pairedImage.sampleRegions(geometry, null,3);

I have playes with the last number in the funnction , I started with 50 but then went down to 30,10,5 ,1 and 3, but non of them worked and in all I have gotten the same error. is it possible that it happens because of the size of the image and there is nothing I can do?

Here is a link to my code:

https://code.earthengine.google.com/f967f358a45cc1c40ccc6d0373066270

My end goal: to be able to sample this image and create the chart

Best Answer

The collection you pass to ui.Chart.feature.byFeature() cannot contain more than 5000 features. In your case, your sample collection contains way more than 5000.

pairedImage.sampleRegions(geometry, null,3);

This samples your image at a scale of 3 meters. Since your image comes from Sentinel 2, it’s pointless to sampling at a scale smaller than 10 meters. Sampling your geometry at 10 meters gives you more than 600,000 features. That leaves you with two options: Sample at a larger scale or sample a smaller area. The below script samples at 100 meters:

https://code.earthengine.google.com/ce8aa2625c3bdf29e583364a19994dc4

Related Solutions

[GIS] Earth Engine: issues with NAIP image collection

The NAIP image collection contains over 1.4 million images, and printing the image collection requests metadata for each of the images. This is probably not what a user actually wants to do, so the query limit of 5000 elements (images in this case) seems to be a reasonable limit.

If you do want to see the metadata for the NAIP images, do it on a subset of the collection. For example, limit it to 10 images:

var NAIP = ee.ImageCollection('USDA/NAIP/DOQQ');
print("NAIP sample", NAIP.limit(10));

Backing up to your original goal, a map-based visualization of Earth Engine data can be done using the Map.addLayer() command. For example, the NAIP images for 2015 can be visualized with the following code:

var NAIP2015 = NAIP.filterDate('2015', '2016');
Map.addLayer(NAIP2015, {bands: 'R,G,B'}, 'NAIP 2016');

Note-1: NAIP only covers selected US states each year, so there may be no data where you are looking.

Note-2: The spectral bands of NAIP images has changed over the years and may not even be consistent within a single year, so you may need to filter the collection to make it homogeneous before displaying it. For example, this code filters out the 2006 images that do not contain the blue 'B' band:

var NAIP2006 = NAIP.filterDate('2006', '2007')
  .filter(ee.Filter.listContains('system:band_names', 'B'));
Map.addLayer(NAIP2006, {bands: 'R,G,B'}, 'NAIP 2006');

Google Earth Engine – Error Collection Query Aborted After 5000 Elements

First to address your question, the Error occurs when you are trying to print more than 5000 elements in your console which it says. Since you have 5099 images in specified date range, the error arises. Since you don't want to do a global analysis as you are clipping images, you can actually use

.filterBounds(geometry)

on your image collection along with your other filters so that you only get images within your region which should be much lower. As for viewing mean NDVI, i can see your layer without any issue. however if you are not seeing the NDVI layer at all then that is probably because the images in that date range in your area has cloud cover more than 20 percent which is filtered out by

.filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20))

in your code. You could try increasing the value 20 to something higher but you might still get empty pixels because of the cloud masking. So it might be a good idea to see the images first without removing clounds and then play with dates so that you get a good number of non-cloudy pixels.

Best Answer

Related Solutions

[GIS] Earth Engine: issues with NAIP image collection

Google Earth Engine – Error Collection Query Aborted After 5000 Elements

Related Question