Sentinel 2 - L2A "Cirrus Correction" and the "Terrain Correction"

Hi

I performed the convertion in L2A 10m (Thanks to ChristianSeverin for his Workaround :wink:) and it seems to work!

I read in the PDF manual “sen2cor-sum-2.0” that the “Cirrus Correction” and the “Terrain Correction” are both optional;

How can I perform these corrections?

I use Sen2Cor in SNAP under Windows 10 64 bit

Thanks in advance!

Roberto

Hi,
I suppose that you have to set the proper values in your config file L2A_GIPP.xml

Line 8: A folder to download the SRTM terrain data
Line 95: A 1 for cirrus correction
Line 96: A 1 for BRDF correction

Thanks for the answer @ralf,

I modified the file L2A_GIPP.xml

DEM_Directory>DEM</DEM_Directory>
Cirrus_Correction>1</Cirrus_Correction>
BRDF_Correction>1</BRDF_Correction>

I performed the correction with Cirrus and Terrain Correction (RES 10m),

The script seem to work, it downloads a DEM in my folder, unzip it and all my granules are processed but…

the outputs seem the same that I had with the normal L2A conversion (without cirrus and terrain)

any idea??

Thanks

1 Like

Hi,

the terrain correction works in my case, although the DEM resolution (~90m) - which is used automatically - seems to be not optimal for the 10m processing.

For all the other options i’ve got the same problem!!! no matter which option i choose via the L2A_GIPP.xml, the output is unchanged.

However I found an explanation in the Sentinel-2 MSI – Level 2A Products Algorithm Theoretical Basis Document why 2 options are not working:

  • the “issue of cirrus removal is left open” for Sentinel-2 data
  • the Haze removal is also disabled Sentinel-2, if i got it right…

has anyone an idea how to change the slope-criterion > 6° => perform topographic correction??

Hi,
Have you figured this out?

I got the same problem, I run with sen2cor 2.1.1
DEM works, but both brdf and cirrus correction are not working, the output is unchanged at all.

Many thanks!

unfortunately not…sorry!

Hello everyone,

topographic correction does not work in my case…
I change the L2A_GIPP.xml as follows:
DEM_Directory > DEM

Also, i tried to do it using my own DEM in GeoTiff format (.tif, geographic WGS 84) and Global.

So, i changed the dem directory:
DEM_Directory > DemGreece

But the result is unchanged at all…

Am i doing something wrong; or that’s it because the terrain is rugged; I think that i could recognize some difference if existed.

Cirrus correction does not work for sure.

@HaiyanHuang @michi HELP :sweat:

You meant no difference in the output no matter you with topographic
correction or without topographic correction?

I have no idea…

I do not have my own DEM files, what I do is to set the line 10 in
L2A_GIPP.xml to <DEM_Directory>dem/srtm</DEM_Directory>, it downloads DEM
automatically according to the L1C megadata.

BWT, I run the linux version 2.1.1

I am trying to perform the topographic (slope) and cirrus correction of L1C images from Sentinel 2A, using the SNAP with Sen2Cor.

However, it is very difficult to find a user manual for these two operations (I have also reviewed this STEP forum, the L2A-SUM]%20S2-PDGS-MPC-L2A-SUM%20[2.3.0].pdf document, the software help, etc.). I was wondering if you have any additional information about how to do this (publication, video, etc.).

@r.bianconi,I had the same problem. I wonder if you had solved this problem now?
I also modified the file L2A_GIPP.xml,
DEM_Directory>DEM
Cirrus_Correction>True
BRDF_Correction>1
I also performed the correction with Cirrus and Terrain Correction. But the results are without eliminating the topographic effects.

Hi all,

Has anyone been able to solve this issue? I’ve tried Sen2Cor with the stand alone version in windows and the SNAP extension GUI. I’ve run both with the SRTM with terrain correction enabled and disabled at three locations around the world with large differences in elevation and there is no difference in the results. Does anyone have a way of verifying that the terrain correction is actually occurring other than saving the output DEMs? I can verify that the DEMs are being downloaded and saved, but there is no difference in the values of the output rasters.
Thanks,
Robbie

Hi, After some struggling, I have managed to do the correction using the DEM. I think the main problem, which is also still happening in the current SNAP version is that the link to download the DEM is not correct. You can change it in the stand alone version of L2A_GIPP.xml to
<DEM_Reference>http://srtm.csi.cgiar.org/wp-content/uploads/files/srtm_5x5/TIFF/</DEM_Reference>.
It would be good to see an update of this issue by ESA soon.
Cheers
Gonçalo Vieira

1 Like