Remote sensing of cyanobacterias with Sentinel-2 in the Salto Grande

Abstract. The Salto Grande Reservoir of the Uruguay River experiences recurrent increases of cyanobacteria bloom, which negatively impacts water quality, with adverse consequences for public health and tourism. This study uses images from the Sentinel-2 satellite to investigate different methods for detecting and monitoring cyanobacteria. Two indices are calculated and two models are applied for the detection of cyanobacteria: the Normalized Difference Vegetation Index (NDVI), the Normalized Difference Water Index (NDWI) and Chlorophyll-a models M1 and M2, which have not previously been evaluated with images from Sentinel-2. NDVI is commonly used to identify areas with vegetation cover. The NDWI, using green and near-infrared (NIR) bands, is used as a filter for the application of Chlorophyl-A models since it is very useful for the identification of aquatic areas. Both M1 and M2 estimate the concentration of chlorophyll in the water, with model M2 demonstrating greater efficiency in the detection of cyanobacteria compared to model M1. Sentinel-2 images allow the correct observation of variations in the concentration of cyanobacteria in the riverbed, facilitating the monitoring of seasonal changes and cyanobacteria blooms. The results of our investigation underlines the effectiveness of NDWI when used with the Sentinel-2 images for identifying surface water areas and the M2-Chlorophyll-a model for detecting cyanobacteria in them. Combining these indices with on-site measurements is likely to offer a robust approach to monitoring and managing of the Salto Grande Reservoir.