Â
Abstract
River inundation has significant impacts on water, energy and carbon cycles of the Amazon Basin. Modeling river flow and inundation of this basin with the continental-scale model faces a number of challenges including uncertainties in floodplain topography, channel geometry and roughness, and flow representation. Efforts were made to handle these uncertainties when surface water dynamics of the Amazonia were simulated by using the MOSART–Inundation model, a surface-water transport model extended with a macro-scale inundation scheme. Spatially diverse biases embedded in model inputs of floodplain topography, channel cross-sectional geometry and channel roughness were alleviated. The model results showed that refining floodplain and channel morphology, channel roughness, as well as accounting for backwater effects helped to evidently improve the simulated surface water dynamics of the Amazon Basin. The understanding obtained in this study could be helpful to improving the modeling of surface hydrology in watersheds with extensive inundation, especially at regional or larger scales.