Evaluation of S3-OLCI water reflectance from updated processors in the highly turbid waters of Río de la Plata (Argentina)

DOGLIOTTI, A.I.; GOSSN, J.I

Encuentro; 6th Sentinel-3 Validation Team Virtual Meeting; 2020

European Space Agency (ESA)

In order to assess and ensure high-quality satellite-derived geophysical data products, validation of the primary product, i.e. water reflectance, using in situ measurements isessential. In view of installing an automated hyperspectral radiometer as part of WATERHYPERNET network, a fixed station has been proposed as a test site for validation of satellite missions, like Sentinel-3, in the highly turbid waters of the Río de la Plata estuary, an ideal scenario to test atmospheric correction algorithms performance. In this work the recently updated Bright Pixel Correction (BPC) with and without new vicarious calibration gains, the current BPAC/BAC (L2-std), the latest version of Case 2 Regional Coast Colour processor (C2RCC.v2) and the successful Base Line Residual (BLR) approach are tested and its performance is compared to in situ measurements. Field measurements have been collected between 2017 and 2019 from the Fishermen Pier in Buenos Aires, one of the possible locations for the hyperspectral radiometer. Measurements of water reflectance using a Trios/RAMSES radiometer and turbidity (portable HACH) have been collected simultaneously with Sentinel-3 A & B overpasses. A considerable improvement have been observed when using BPC using the new vicarious calibration coefficients compared to using the old ones and also to current L2-std, specially in the blue bands, retrieving less negative values in the shortest wavelengths, as shown by the flags. However, a general overestimation at all bands can be observed. In turn, C2RCCv2 show high underestimation in bands <500 nm while in the NIR, together with BLR ,showed the best performance.