Simulating sediment supply from the Congo watershed over the last 155 ka

The Congo River is the world's second largest river in terms of drainage area and water discharge. Monitored for decades, a large dataset is available, onshore for both the hydrology and sediment load, and offshore by many paleo-environmental proxies compiled at the Late-Quaternary time-scale. These numerous data allow for accurate calibration of numerical modeling. In this study, we aim to numerically quantify the evolution of sediment supply leaving the tropical Congo watershed during the last 155 ka and to decipher the forcing parameters that control this sediment supply over glacial/interglacial stages. For this, a modified version of the model HydroTrend, that besides morphologic, climatic, hydrologic, lithologic, land cover and anthropogenic factors now also considers sediment deposition on the floodplain, is used. In addition, a method to quantify the impact of natural vegetation changes is developed. Simulations match well the present-day observed data. They indicate that a significant portion of suspended sediments is trapped on the floodplain. Long-term simulations indicate that environmental changes between glacial and interglacial stages account for a 30% maximum variation of sediment supply. Climatic changes - precipitation and temperature, account for a maximum decrease in sediment supply of 20% during cold periods while conversely, induced land cover changes (loss of forest during colder and dryer stages) lead to enhanced sediment supply up to 30%. Over a longer period, the average sediment supply remained almost constant during glacial and interglacial periods, while peaks may have occurred during a warming period, just before forests had time to recover the catchment, i.e. during post-glacial periods. These moderate changes in sediment export, despite major changes in climate and vegetation cover, can be explained by the efficiency of sediment trapping of large tropical catchments that buffer fluvial fluxes towards the ocean