Abstract
Abstract. Data provided by spatial sensors combined with remote sensing techniques and analysis of the optical properties of waters allow the mapping of the suspended sediment concentration (SSC) in aquatic bodies. For this, estimation models require data with the lowest possible amount of atmospheric artifacts. In this study we compared the water remote sensing reflectance (Rrs) of the Santo Antônio Hydroelectric Power Plant reservoir in Porto Velho-RO, Brazil, after applying three different atmospheric corrections algorithms in Sentinel-2/MSI imagery products. The atmospheric corrected reflectances of the MODIS sensor were also used for reference. SSC was calculated with models based on the red and near-infrared (NIR) bands over three distinct regions of the reservoir. Reflectance data showed significant variations for Sentinel-2, bands 4 and 8a, and MODIS, bands RED and IR, when different atmospheric correction algorithms were used. SSC maps and estimates were produced to show sediment load variation as a function of hydrological regime. The analyzes showed that the SSC estimates done with Sentinel-2 / MSI satellite images using GRS (Glint Remove Sentinel) atmospheric correction presented an average difference of 27.3% and were the closest to the in situ measurements. SSC estimates from MODIS products were around 34.6% different from estimates made using the GRS atmospheric correction applied to Sentinel-2 / MSI products.
Highlights
Remote sensing data has been widely used to study the optical properties of waters and to support analyzes of the water bodies sedimentary behavior (Cavali et al 2019; Palmer et al 2014)
3.1 Comparison between Sentinel-2 reflectances estimated from GRS, MAJA and Sen2cor atmospheric corrections
Comparing the band 4 Sentinel-2 / MSI reflectances (Table 1) resulting from each processing atmospheric corrections, we observe that the MAJA and Sen2Cor products show close behavior and that both processing show distinct values in relation to the GRS processing
Summary
Remote sensing data has been widely used to study the optical properties of waters and to support analyzes of the water bodies sedimentary behavior (Cavali et al 2019; Palmer et al 2014). An advantageous mechanism to estimate suspended sediment concentrations (SSC) in water bodies is the application of bio-optical algorithms, which, from data provided by orbital sensors, derive parameters from aquatic constituents (Ogashwara, 2015). Such data must be free of artifacts such as clouds, water vapor, aerosols, shadows and reflections of sunlight and sky. The MAJA atmospheric correction (Haggole et al 2017) introduces a multitemporal correction to better detect variations from atmospheric processes while SEN2COR (MüllerW, 2015) uses a library of radiation transfer data for different parameters, generating a scene classification to produce the corrected images Both algorithms were developed for the French and European space agencies respectively, and provide generic atmospheric corrections of Sentinel-2 images for continental surfaces, without particular specification for continental waters. The SSC estimates retrieved with the different satellite products were compared with water samplings realized in the Madeira River by the international monitoring network SO HYBAM (Geodynamical, hydrological and biogeochemical control of erosion/alteration and material transport in the Amazon, Orinoco and Congo basins) maintained by the Federal University of Amazonas (UFAM) and the IRD (Institut de Recherche pour le Développement) since 2003 in Brazil
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
