5:15 PM - 7:15 PM
[AHW29-P06] From One to Many: Integrating Bifurcations and Multi-Directional River Channels in Large-Scale Flood Models
Keywords:Flood, River, Risk, SWOT
Over 70% of recorded flood events in the past two decades, as documented in the Global Flood Database and WorldFloods dataset, have occurred in regions where river networks diverge, such as bifurcations and deltas. However, large-scale flood models have historically relied on single-threaded river networks due to the absence of a dataset incorporating bifurcations. As a result, these models fail to accurately capture the physical processes in these often densely populated areas, leading to suboptimal flood risk estimates.
Using the new Global River Topology (GRIT) dataset, a global bifurcation and multi-directional river network (Wortmann et al. 2024), we extend the river channel bathymetry estimation routine of Neal et al. (2021), including SWOT observations, to model flood events with LISFLOOD-FP. We compare the multi-thread model results to observations and previous versions of LISFLOOD-FP that use a single-threaded river network. We use test cases in the Irrawaddy, Mekong and inner Niger rivers at 1 arc second (~30m).
Incorporating GRIT significantly improves the accuracy of flood modelling, particularly in capturing flood timing and the connectivity between non-mainstem channels and floodplains. This leads to a more realistic representation of flood processes compared to single-threaded river networks. Our findings advance the understanding of global flood risk and highlight the dynamic, evolving nature of geomorphologically active river systems. Moving forward, we will continue to model various bifurcation types and multi-directional river systems to deepen our insights into the complexities of global river networks.
Using the new Global River Topology (GRIT) dataset, a global bifurcation and multi-directional river network (Wortmann et al. 2024), we extend the river channel bathymetry estimation routine of Neal et al. (2021), including SWOT observations, to model flood events with LISFLOOD-FP. We compare the multi-thread model results to observations and previous versions of LISFLOOD-FP that use a single-threaded river network. We use test cases in the Irrawaddy, Mekong and inner Niger rivers at 1 arc second (~30m).
Incorporating GRIT significantly improves the accuracy of flood modelling, particularly in capturing flood timing and the connectivity between non-mainstem channels and floodplains. This leads to a more realistic representation of flood processes compared to single-threaded river networks. Our findings advance the understanding of global flood risk and highlight the dynamic, evolving nature of geomorphologically active river systems. Moving forward, we will continue to model various bifurcation types and multi-directional river systems to deepen our insights into the complexities of global river networks.