We are starting to use sen2agri. We have observed that the L2A products generated by sen2agri differ from L2A productos distributed by https://theia.cnes.fr/. In agricultural stubble and bare soils we find strange that band 12 sometimes has high values compared to band 11 in L2A products generated by sen2agri, not in theia L2A products or esa L2A. Any idea of why these differences?
Thank you for your message. We have also discovered this issue. It comes from the GIPP files (the Ground Image Processing Parameters) prepared by CNES and used by MACCS, they were not totally aligned between the Sen2-Agri system and the MAJA production center at CNES. We have recently received the corrected version and we’ve just completed the final tests to validate their use. Surface reflectances are now fully aligned.
The corrected GIPP files will be delivered early July via the Sen2-Agri web site and every registered user will be informed by email.
The new version 1.8.2, released this morning, is provided with updated GIPP files used for the atmospheric correction. This new version is fixing the issue that you encountered before.
Please refer to this post to have the full overview of the content of this new version along with detailed instructions on how to update your system.
Let us know if you encounter any issue with that new version,