Articles | Volume 15, issue 4
https://doi.org/10.5194/amt-15-895-2022
https://doi.org/10.5194/amt-15-895-2022
Research article
 | 
21 Feb 2022
Research article |  | 21 Feb 2022

Deep-learning-based post-process correction of the aerosol parameters in the high-resolution Sentinel-3 Level-2 Synergy product

Antti Lipponen, Jaakko Reinvall, Arttu Väisänen, Henri Taskinen, Timo Lähivaara, Larisa Sogacheva, Pekka Kolmonen, Kari Lehtinen, Antti Arola, and Ville Kolehmainen

Data sets

POPCORN Sentinel-3 Synergy aerosol data (https://a3s.fi/swift/v1/AUTH_ca5072b7b22e463b85a2739fd6cd5732/POPCORNdata/readme.html) Antti Lipponen, Jaakko Reinvall, Arttu Väisänen, Henri Taskinen, Timo Lähivaara, Larisa Sogacheva, Pekka Kolmonen, Kari Lehtinen, Antti Arola, and Ville Kolehmainen https://doi.org/10.23728/FMIB2SHARE.C81ADE576E1C49E4AEF9CA1CA8A7621A

Model code and software

S3 POPCORN with accuracy and spatial anomaly corrections v1.0.0 (https://a3s.fi/swift/v1/AUTH_ca5072b7b22e463b85a2739fd6cd5732/POPCORNdata/readme.html) Antti Lipponen, Jaakko Reinvall, Arttu Väisänen, Henri Taskinen, Timo Lähivaara, Larisa Sogacheva, Pekka Kolmonen, Kari Lehtinen, Antti Arola, and Ville Kolehmainen https://doi.org/10.5281/zenodo.6042568

Video supplement

Animation of Sentinel-3 aerosol optical depth over Europe in 2019 Antti Lipponen https://doi.org/10.5281/zenodo.5287243

Download
Short summary
We have developed a machine-learning-based model that can be used to correct the Sentinel-3 satellite-based aerosol parameter data of the Synergy data product. The strength of the model is that the original satellite data processing does not have to be carried out again but the correction can be carried out with the data already available. We show that the correction significantly improves the accuracy of the satellite aerosol parameters.