Sen2cor failing with “TCI image export failed”

sen2corsentinelsentinel-2toa-calibration

I'm trying to calibrate some Sentinel L1C products to get level 2 data, but I always end up with the following message. Any idea what could cause that?

I have tested it with sen2cor 2.8.0 and sen2cor 2.9.0 with the same result. The product I'm working with is S2B_MSIL1C_20180421T100029_N0206_R122_T33UVR_20180421T120642.SAFE if that could be of any interest.

Sentinel-2 Level 2A Processor (Sen2Cor). Version: 2.8.0, created: 2019.02.20, supporting Level-1C product version 14.2 - 14.5 started ...
Product version: 14.5
Operation mode: TOOLBOX
Processing baseline: 99.99
Progress[%]:  0.00 : Generating datastrip metadata
L1C datastrip found, L2A datastrip successfully generated
Selected resolution: 10 m
Progress[%]: 0.05 : PID-10, L2A_ProcessTile: 20 m resolution must be processed first, elapsed time[s]: 1.261, total: 0:00:07.319807
Progress[%]: 0.12 : PID-10, L2A_ProcessTile: processing with resolution 20 m, elapsed time[s]: 1.571, total: 0:00:08.890806
Progress[%]: 0.12 : PID-10, L2A_ProcessTile: start of pre processing, elapsed time[s]: 0.000, total: 0:00:08.891253
Progress[%]: 0.12 : PID-10, L2A_Tables: start import, elapsed time[s]: 0.057, total: 0:00:08.948456
Progress[%]: 0.18 : PID-10, L2A_Tables: band B01 imported, elapsed time[s]: 1.522, total: 0:00:10.470635
...
Progress[%]: 50.05 : PID-10, L2A_Tables: band AOT exported, elapsed time[s]: 3.815, total: 0:20:07.139011
Progress[%]: 51.00 : PID-10, L2A_Tables: band WVP exported, elapsed time[s]: 22.923, total: 0:20:30.062450
TCI image export failed
Progress[%]: 51.34 : PID-10, L2A_Tables: TCI image export failed, elapsed time[s]: 8.100, total: 0:20:38.162143
Progress[%]: 51.34 : PID-10, L2A_Tables: stop export, elapsed time[s]: 0.073, total: 0:20:38.235323
Module L2A_T33UVR_A005868_20180421T100427 failed
Progress[%]: 100.00 : Application terminated with at least one error.

Best Answer

Okay, the problem was that I ran out of RAM. I would appreciate better error messages from sen2cor though, but yes, it was so simple.

Related Solutions

[GIS] Image Export fails in Google Earth Engine because “export region contains no valid (un-masked) pixels

1st: The exact code you posted (https://code.earthengine.google.com/ce1a151ce06497b20cf1793715cb0120) did export the image correctly. So the error cannot be reproduced. May be, you changed the 'ROI' to a place where the filtered collection had no images.

2nd: There are no images because you filtered by cloud percentage, and so, it found images that suited the condition only in the left part of the ROI. If you comment like:

//.filter(ee.Filter.lte('CLOUDY_PIXEL_PERCENTAGE', 5))

you'll see that there are more images and you get the whole ROI (of course you get a part of the image plenty of clouds).

3rd: if automated exporting is your goal, maybe you should migrate to Python, but if you don't mind clicking the run button over and over, this would be one approach to download images only if the filtered collection has images (don't blame me on the getInfo =):

//Define date range
var startDate = ee.Date.fromYMD(2017,7,1);
var endDate = ee.Date.fromYMD(2017,7,31);

//Load Sentinel-2 image collections
var coll_s2 = ee.ImageCollection("COPERNICUS/S2");

//Filter Sentinel-2 collection for ROI and cloud-coverage.
//Keep only images with less than 5% clouds
var coll_s2_filtered = coll_s2.filterDate(startDate, endDate)
                              .filter(ee.Filter.lte('CLOUDY_PIXEL_PERCENTAGE', 5))
                              .sort('system:time_start',false);

// Define the FeatureCollection!
var fc = ee.FeatureCollection(LIST_OF_POINT_FEATURES)

// Iterate over the FeatureCollection to create a List
// with images where there is data avialable
var listn = ee.List(fc.iterate(function(elem, ini){
  var rectangle = ee.Feature(elem.buffer(20000).bounds());
  var roi = rectangle.geometry();
  var col = coll_s2_filtered.filterBounds(roi);
  return ee.Algorithms.If(col.size(), 
            ee.List(ini).add(coll_s2_filtered.select(['B4','B3','B2']).mosaic().clip(roi)), 
            ee.List(ini))
}, ee.List([])))

listn = ee.List(listn)

print(listn)

// As Export is a Client-side function, you have to iterate the list
// in a client-side way

for (var n = 0; n<listn.size().getInfo(); n++) {
  var i = ee.Image(listn.get(n))
  Export.image.toDrive({
    image: i,
    description: 'image_'+n.toString(),
    scale: 10,
    folder: "tests",
    region: i.geometry()
    });
}

But if you want to change the filter regarding on whether it finds or not images, that would be different.

[GIS] Export Sentinal 2 TIFF image from SNAP with all the bands intact

It is basically a simple File > Export > 'GeoTiff / BigTiff' operation, however, it will throw an error as Sentinel-2 bands have various resolutions (10, 20, 60m).

Resample it first, then export.

Start Raster > Geometric Operations > Resampling. On the Resampling Parameters tab, you may choose any of B2, B3, B4 for the reference band, to resample all the bands to 10m resolution.
Create bands subset by Raster > Subset.
On the Product Explorer pane, click on the subset data to highlight then go to File > Export > 'GeoTiff / BigTiff'.

All bands such as the Band 1 (resolution: 60m) will be resampled to 10m in the first step. So this may not give you what you need, if the goal is to stack all the bands as-is.

Best Answer

Related Solutions

[GIS] Image Export fails in Google Earth Engine because “export region contains no valid (un-masked) pixels

[GIS] Export Sentinal 2 TIFF image from SNAP with all the bands intact

Related Question