The routing of a flood can influence whether a flood wave is attenuated (i.e., slowed and/or reduced in magnitude) as it moves downstream, which is critical for mitigating flood hazards. Floodplains can influence the routing of floods by reducing flow velocity and providing temporary storage of floodwaters. Existing literature shows how various reach characteristics (e.g., slope, roughness, etc.) and flood wave characteristics influence flood attenuation. However, there has been little focus on how the variability of topographic features, or geomorphic heterogeneity (GH), of floodplains influences flood attenuation by altering unit-scale hydraulic processes. We performed unsteady flow simulations using HEC-RAS 2D models for two study sites in Vermont, United States that had contrasting reach characteristics. The sites also had contrasting floodplain topographic characteristics—different composition (shapes, sizes, etc.) and spatial configurations (distances from channel, orientations, etc.) of meso-scale topographic units (landforms on the order of a few meters to tens of meters in length). We isolated the effect of floodplain topography at each site by comparing flood routing between the natural floodplain and a smoothed floodplain in which the meso-scale topography was removed, leaving no distinct topographic units. Differences in topographic unit variability between the sites were evaluated using unit area, aggregation, distance from channel, elongation, orientation and overall rugosity of the floodplain. We analyzed changes in discharge attenuation, celerity, and stream power at the reach scale and changes in flow velocity and inundation duration at the topographic unit scale. We examined differences between three flood magnitudes at each site and between sites. At the reach scale, we found that the Lemon Fair River with its lower gradient and wider floodplain naturally attenuated >50% of upstream discharge, whereas the Lewis Creek, with a steeper gradient and narrower floodplain, attenuated ~4%. For both sites, we found that floodplain GH had a minimal (though measurable) impact on discharge attenuation but a much more pronounced effect on reducing stream power. Topographic units at Lemon Fair River had more variation in distance from the main channel, concave units were less elongated and had a wider range of orientations relative to those Lewis Creek, which were largely oriented downstream. Removing GH from the floodplain had consistent effects on inundation duration and flow velocity for convex units but different effects for concave and flat units between the two sites. These variations suggest that the potential effects of floodplain GH on flood routing are partly controlled by the specific composition and configuration of floodplain units as well as partly controlled by key reach characteristics (e.g., slope). Findings of this study showed that floodplain GH at our sites has a much stronger potential to mitigate fluvial erosion hazards via stream power reduction than potential to reduce inundation hazards via discharge attenuation. Relationships between topographic unit shapes and flow hydraulics can be applied to alter inundation durations (e.g., to affect biogeochemical processes) or to reduce floodplain flow velocities (e.g., to encourage sediment deposition). This study also demonstrates how hydrodynamic models that represent floodplains as flat and featureless may be missing important changes in routing being induced by floodplain GH.
Stewart Kabis (Wed,) studied this question.