Assessment of surface water storage in the Amazon floodplains by hydrological modelling and earth observation data

• A new estimate of Amazonian floodplains water storage is provided. • SWAT-FP simulated correctly the floodplain water surface elevations (R 2 > 0.62) • Interannual water storage reach 1,800 ± 854 km 3. • The main contributors are the Madeira and the Negro subbasins. • The volumes of water stored in each subsystem have a significant temporality. Inland surface waters in tropical regions are fundamental to global hydrological and biogeochemical cycles, yet the quantification of floodplain freshwater storage and its variability remains limited. This study provides a basin-wide assessment of the spatiotemporal dynamics of floodplain water storage in the Amazon Basin over the period 2000–2018. We developed an integrated framework combining in situ discharge measurements, multi-sensor remote sensing datasets, and process-based hydrological modelling to quantify daily floodplain hydrodynamics at high spatial resolution. Floodplain extent was derived from L-band passive microwave observations from Soil Moisture and Ocean Salinity (SMOS) satellite, while water surface elevation was constrained using altimetry data (Jason-2/3 and Sentinel-3). The hydrological model (SWAF-FP), specifically improved to represent floodplain–river interactions, was calibrated at eight gauging stations and evaluated using approximately 4,000 altimetry-based floodplain water level records. Model performance demonstrated high skill, with a mean R 2 of 0.83 for discharge and R 2 > 0.62 for floodplain water surface elevation. The results indicate a mean annual surface water storage of 1800 ± 854 km 3 across the Amazon Basin, with pronounced seasonal cycles and interannual stability at the basin scale. The Negro and Madeira sub-basins were identified as the dominant contributors to total floodplain storage, accounting for approximately 480 km 3 (33.6 %) and 312 km 3 (12.5 %), respectively. This study provides the first consistent long-term quantification of Amazonian floodplain water storage dynamics at daily resolution. The results emphasize the critical role of floodplains in regulating basin-scale hydrological fluxes and offer a robust observational–modeling framework for improving large-scale hydrological and Earth system models in tropical regions.