EXPLOITATION OF HYPERSPECTRAL PRISMA DATA FOR WATER QUALITY MAPPING

Nowadays, waterbodies have been seriously threatened by climate change: in this context, new-generation satellite hyperspectral sensors are contributing to the water resources studies. This work shows the main goals achieved by the exploitation of PRISMA data for aquatic ecosystems mapping in the framework of the ASI-PRISCAV, PANDA-WATER and PRIMEWATER projects. The suitability of PRISMA radiance and reflectance products was demonstrated (Giardino et al. 2020; Braga et al. 2022) through comparison with reference data: Sentinel-2, Sentinel-3 and in-situ measurements obtained through ad-hoc campaigns, where reflectance and water quality data are collected simultaneously. To demonstrate the PRISMA water quality mapping two approaches were tested: the bio-optical model inversion with the BOMBER code and adaptive semi-empirical algorithms. Bottom substrate (BS) maps were generated by running BOMBER, Total Suspended Matter (TSM) maps with the semi-empirical algorithms and Chlorophyll-a (Chl-a) maps with both approaches. The validation of the water quality products was performed by comparison with in-situ data. The maps obtained with BOMBER showed an accuracy of 80% for the characterization of the BS and RMSE=2.0 mgm-3 for Chl-a retrieval. The maps obtained with the semi-empirical algorithms showed the following results: RMSE=1.9 gm-3 for TSM retrieval and RMSE=3.1 mgm-3 for Chl-a retrieval. Despite the need of further analysis, PRISMA is providing valuable information and - in synergy with multispectral missions - is offering support to water management activities.