Global-scale simulation of river-floodplain water and sediment exchange within a riverine modeling framework

Thursday, 18 December 2014: 5:15 PM
Sagy Cohen, University of Alabama, Tuscaloosa, AL, United States, Albert J Kettner, University of Colorado, Boulder, CO, United States, James P Syvitski, University of Colorado at Boulder, CSDMS/INSTAAR, Boulder, CO, United States and Dai Yamazaki, JAMSTEC Japan Agency for Marine-Earth Science and Technology, Kanagawa, Japan
The exchange of water, sediment and nutrients between a river and its floodplain can significantly alter the fluvial flux towards the ocean. The flux of water and sediment from the active river channel to the floodplain during overbank flows can considerably reduce peak water discharge downstream and store large volumes of sediment. On the other hand floodplains can serve as short-term reservoirs of water, increasing base flow discharge following a flood event. For sediment, floodplains can serve as longer-term reservoirs, affecting the sediment flux as a function of river migration and flooding frequency. The effect of these relatively local processes can propagate downstream, potentially effecting riverine fluxes, especially in larger river systems. Here we present a floodplain module within our spatially and temporally explicit global riverine model (WBMsed v.2). The module predicts daily water and sediment exchange between the active river channel and its floodplain. Given the large spatial scale (and resulting resolution) of our predictions (6 arc-min) the floodplain module is based on simple estimations of bankfull discharge for every river location globally. During flood events, when daily-simulated water discharge is to “exceed” overbank flow, excess water and sediment will be routed to the floodplain. After the flood, water will be re-injected to the river when simulated water discharge is lower than bankfull discharge. These returning waters can transport sediment from the floodplain but here we assume that all the sediment transported during a flood event will be deposited on the floodplain. This will allow us to estimate floodplain sedimentation rates. Our results show that the WBMsed water discharge simulations significantly improved when introducing the floodplain reservoir module. Here we present our modeling approach, results and future developments.