Missing quality indicator sen2cor v2.10.01

sylar11 · June 10, 2022, 1:04pm

Hi all,
I’m using the sen2cor 2.10.01 and I’m facing this problem: the quality indicators are less then usual and result as Not Available

I tested this with these two files:

S2A_MSIL1C_20220319T100731_N0400_R022_T32TPS_20220319T111153.zip
S2A_MSIL1C_20220322T101711_N0400_R065_T33UUP_20220322T123834.zip

in the first case I only find in the S2A_MSIL2A_20220319T100731_N9999_R022_T32TPS_20220610T074029.SAFE/GRANULE/L2A_T32TPS_A035197_20220319T101418/MTD_TL.xml file the following indicators:

<Image_Content_QI>
<CLOUDY_PIXEL_PERCENTAGE>0.125205291289677</CLOUDY_PIXEL_PERCENTAGE>
<DEGRADED_MSI_DATA_PERCENTAGE>0.0035</DEGRADED_MSI_DATA_PERCENTAGE>
<SNOW_PIXEL_PERCENTAGE>40.5256054226761</SNOW_PIXEL_PERCENTAGE>
</Image_Content_QI>

In the second case, I find the usual ones
<Image_Content_QI>
<CLOUDY_PIXEL_PERCENTAGE>0.000204</CLOUDY_PIXEL_PERCENTAGE> <CLOUDY_PIXEL_OVER_LAND_PERCENTAGE>0.000000</CLOUDY_PIXEL_OVER_LAND_PERCENTAGE>
<DEGRADED_MSI_DATA_PERCENTAGE>0.000000</DEGRADED_MSI_DATA_PERCENTAGE>
<NODATA_PIXEL_PERCENTAGE>93.493390</NODATA_PIXEL_PERCENTAGE> <SATURATED_DEFECTIVE_PIXEL_PERCENTAGE>0.000000</SATURATED_DEFECTIVE_PIXEL_PERCENTAGE>
<DARK_FEATURES_PERCENTAGE>2.550255</DARK_FEATURES_PERCENTAGE>
<CLOUD_SHADOW_PERCENTAGE>0.000000</CLOUD_SHADOW_PERCENTAGE>
<VEGETATION_PERCENTAGE>43.945229</VEGETATION_PERCENTAGE>
<NOT_VEGETATED_PERCENTAGE>53.176230</NOT_VEGETATED_PERCENTAGE>
<WATER_PERCENTAGE>0.323951</WATER_PERCENTAGE>
<UNCLASSIFIED_PERCENTAGE>0.004130</UNCLASSIFIED_PERCENTAGE> <MEDIUM_PROBA_CLOUDS_PERCENTAGE>0.000153</MEDIUM_PROBA_CLOUDS_PERCENTAGE> <HIGH_PROBA_CLOUDS_PERCENTAGE>0.000000</HIGH_PROBA_CLOUDS_PERCENTAGE>
<THIN_CIRRUS_PERCENTAGE>0.000051</THIN_CIRRUS_PERCENTAGE>
<SNOW_ICE_PERCENTAGE>0.000000</SNOW_ICE_PERCENTAGE>
<RADIATIVE_TRANSFER_ACCURACY>0.0</RADIATIVE_TRANSFER_ACCURACY> <WATER_VAPOUR_RETRIEVAL_ACCURACY>0.0</WATER_VAPOUR_RETRIEVAL_ACCURACY>
<AOT_RETRIEVAL_ACCURACY>0.0</AOT_RETRIEVAL_ACCURACY>
<AOT_RETRIEVAL_METHOD>SEN2COR_DDV</AOT_RETRIEVAL_METHOD>
<GRANULE_MEAN_AOT>0.065000</GRANULE_MEAN_AOT>
<GRANULE_MEAN_WV>0.272039</GRANULE_MEAN_WV>
<OZONE_SOURCE>CONFIG</OZONE_SOURCE>
<OZONE_VALUE>331.000000</OZONE_VALUE>
</Image_Content_QI>

@Jan sorry to disturb you again but could you please tell me if I’m doing something wrong?
I’m calling the sen2cor in the same way in both cases

Jan · June 13, 2022, 11:23am

Hello @sylar11

For some reason, the L1C of 32TPS won’t process for me in standalone. But I’ve had a look at the L2A Tile in the catalogue (
S2A_MSIL2A_20220319T100731_N0400_R022_T32TPS_20220319T131812), and it has

<Image_Content_QI>
  <CLOUDY_PIXEL_PERCENTAGE>0.498516</CLOUDY_PIXEL_PERCENTAGE>
  <CLOUDY_PIXEL_OVER_LAND_PERCENTAGE>0.499935</CLOUDY_PIXEL_OVER_LAND_PERCENTAGE>
  <DEGRADED_MSI_DATA_PERCENTAGE>0.003500</DEGRADED_MSI_DATA_PERCENTAGE>
  <NODATA_PIXEL_PERCENTAGE>0.000053</NODATA_PIXEL_PERCENTAGE>
  <SATURATED_DEFECTIVE_PIXEL_PERCENTAGE>0.000000</SATURATED_DEFECTIVE_PIXEL_PERCENTAGE>
  <DARK_FEATURES_PERCENTAGE>6.787857</DARK_FEATURES_PERCENTAGE>
  <CLOUD_SHADOW_PERCENTAGE>0.063580</CLOUD_SHADOW_PERCENTAGE>
  <VEGETATION_PERCENTAGE>20.261148</VEGETATION_PERCENTAGE>
  <NOT_VEGETATED_PERCENTAGE>33.675244</NOT_VEGETATED_PERCENTAGE>
  <WATER_PERCENTAGE>0.232411</WATER_PERCENTAGE>
  <UNCLASSIFIED_PERCENTAGE>0.012442</UNCLASSIFIED_PERCENTAGE>
  <MEDIUM_PROBA_CLOUDS_PERCENTAGE>0.209688</MEDIUM_PROBA_CLOUDS_PERCENTAGE>
  <HIGH_PROBA_CLOUDS_PERCENTAGE>0.270311</HIGH_PROBA_CLOUDS_PERCENTAGE>
  <THIN_CIRRUS_PERCENTAGE>0.018517</THIN_CIRRUS_PERCENTAGE>
  <SNOW_ICE_PERCENTAGE>38.468802</SNOW_ICE_PERCENTAGE>
  <RADIATIVE_TRANSFER_ACCURACY>0.0</RADIATIVE_TRANSFER_ACCURACY>
  <WATER_VAPOUR_RETRIEVAL_ACCURACY>0.0</WATER_VAPOUR_RETRIEVAL_ACCURACY>
  <AOT_RETRIEVAL_ACCURACY>0.0</AOT_RETRIEVAL_ACCURACY>
  <AOT_RETRIEVAL_METHOD>SEN2COR_DDV</AOT_RETRIEVAL_METHOD>
  <GRANULE_MEAN_AOT>0.157872</GRANULE_MEAN_AOT>
  <GRANULE_MEAN_WV>0.321216</GRANULE_MEAN_WV>
  <OZONE_SOURCE>AUX_ECMWFT</OZONE_SOURCE>
  <OZONE_VALUE>392.913835</OZONE_VALUE>
</Image_Content_QI>

Thus, Sen2Cor appears to handle the L1C OK in the Ground Segment processing. I note that the Ground Segment processing - or your run of the L1C of 32UUP - makes no mention of the SNOW_PIXEL_PERCENTAGE found in your run of 32TPS. Are you 100% sure you used the same configuration?

Cheers

Jan

sylar11 · June 13, 2022, 12:35pm

@Jan Thanks for your reply.
I’m sure I’m using the same configuration because the two files are in a file list ingested by a code calling the sen2cor in a for-loop.

I’m surprised because using the v2.55 I never found this problem.

sylar11 · June 16, 2022, 11:17am

@Jan I think I found the problem: it seems that if sen2cor found, in the destination folder of the L2A process, a .SAFE folder having the same name this creates some problems and the final output is not correct.
If the destination folder is empty and/or no .SAFE folder with the same name is present the process is correctly completed

Jan · June 17, 2022, 10:46am

Hi @sylar11
That’s good to know. Is it a bug in Sen2Cor, or the unpacking of the SAFE file that is in error, do you know? If you think that it’s a Sen2Cor issue, you can raise it to eosupport@copernicus.esa.int, and they’ll pass it along to the expertise.

Cheers

Jan

OPT-MPC S2 Technical Manager

sylar11 · June 20, 2022, 7:17am

Hi @Jan,

In my opinion, the problem could be in Sen2cor because, probably, it is looking for certain aux data after unpacking the .zip file. As a consequence, if there is more than one folder that satisfies the search criteria, it fails in calculating the quality indices.

In my code, I put an exception before unpacking deleting all folders that have the same initial part of the name:
MMM_MSIL2A_YYYYMMDDHHMMSS_N9999_ROOO_Txxxxx_*.SAFE

I don’t know if it is too much, but in this way, I’m sure that the process will not be corrupted by existing folders and it will end successfully.

What do you think?

Regards,

Bartolomeo