ASSESSMENT OF PLANET IMAGERY USE FOR WETLAND MAPPING IN THE UPPER PARAGUAY BASIN

Resumo

This study investigates the application of Planet/NICFI mosaics for the monthly mapping of surface water and other wetlands in the Upper Paraguay River Basin (BAP) from January 2021 to April 2024. Mapping was made using image classification techniques, based on machine learning and cloud computing processing. The study demonstrated the potential of these images to provide detailed information on the water surface in the BAP, especially in the plateau region, for the identification of new water bodies. Furthermore, the surface water mapping was satisfactory to monitor the monthly variation of the flood pulse in the floodplain. However, the study identified spectral limitations of the Planet sensor, with lower sensitivity for mapping wetlands beyond the water surface. The results indicate that the Planet/NICFI monthly mosaics are a promising tool for monitoring the water surface in the BAP, especially to add detail at the local scale when compared to existing mappings with other orbital images. The temporal analysis of the data also revealed that 2023 was the year with the highest volume of water in the period studied, while the first months of 2024 indicate a strong drought trend, showing the potential of the adopted processing flow as a tool for early detection of drought and flood trends.