In this era of geographic mapping, we can all access cloud services like Google Earth Engine. We can take advantage of the rich datasets available there.
But do we actually care about the processing used to prepare them for analysis? Are you aware that the Sentinel-2 L2A images available on the Google Earth Engine platform have been atmospherically corrected using the Sen2Cor algorithm? That is good for land-related applications because the Sen2Cor algorithm was created using inputs collected from land object properties. For water-related applications, this L2A dataset may not be appropriate. As a result, the conclusions may be equivocal for water-related applications.
POLYMER, ACOLITE, and iCOR are some of the atmospheric correction technologies specifically designed for water-related applications. In a recent study, ACOLITE and iCOR showed good results in water-related atmospheric correction.
References:
Allam, M., Meng, Q., Elhag, M., Giardino, C., Ghirardi, N., Su, Y., Al-Hababi, M.A. and Menenti, M., 2024. Atmospheric Correction Algorithms Assessment for Sentinel-2A Imagery over Inland Waters of China: Case Study, Qiandao Lake. Earth Systems and Environment, 8(1), pp.105-119.
Dehkordi, A.T., Zoej, M.J.V., Mehran, A., Jafari, M. and Chegoonian, A.M., 2024. Fuzzy similarity analysis of effective training samples to improve machine learning estimations of water quality parameters using Sentinel-2 remote sensing data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
