I submitted a manuscript that uses Sentinel SAFE (Standard Archive Format) Level-2A (bottom of atmosphere) data. Sentinel-2A data includes scene classification and an atmospheric correction - https://sentinel.esa.int/web/sentinel/user-guides/sentinel-2-msi/processing-levels/level-2.

My understanding of Sentinel-2A data is that it is essentially ready for analysis, so after downloading it I used it in analysis without any further corrections.

A reviewer stated, "I didn't see that the authors had carried out any spectral correction of topographic effects or bidirectional reflectance distribution function in the sentinel data. These effects may significantly distort the signal the satellite sensor is registering and therefore they should be removed or, alternatively, there should be a convincing explanation why these effects are not relevant in this particular data set."

Are the corrections mentioned needed for Sentinel-2A data?


Topographic effects are accounted for in Sentinel-2 L2A data, although with some approximation.

Bidirectional reflectance distribution function variations are not accounted for, because in order to perform such a correction, you need to know very well the 3D subpixel geometry of each pixel. Or you need measurements from different viewing angles at different times of the day. However, Sentinel-2 field of view is rather narrow, so directional effects due to viewing angle variation are quite small. Those due to sun zenith angle are still high, but the sun eznith angle varies slowly with time.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.