Spatially-Explicit Predictions of Floodplain Sedimentation Using Mixed Empirical/Processed-Based Models of Increasing Complexity
Abstract
Understanding how sediment and sediment-borne contaminants are stored and released from floodplains is important for predicting watershed sediment delivery, managing watersheds, and improving water quality. Mercury inventories from hundreds of floodplain cores representing 77 years of overbank deposition along the South River, Virginia, provide spatially-distributed estimates of sedimentation rates that are ideal for testing models of floodplain sedimentation. The South River is a sinuous, single-thread, mixed bedrock-alluvial river with average deposition rates of 0.04 cm/yr (median 0.008 cm/yr; range 5E-05 - 0.6 cm/yr). We use these data to test mixed empirical/processed-based models of increasing complexity of overbank deposition that account for flood frequency, distance from the channel, vegetation type, overbank flood pathways, and/or shear stress. We developed both 1-D and 2-D HEC-RAS models, which were constructed using LiDAR data, surveyed cross-sections, and were calibrated to stream gaging station observations. Both 1-D and 2-D HEC-RAS models were used to test the improved predictability of using the 2-D model that better captures local topographic variability and flood pathways. For quantifying vegetation, we used raw point clouds (.LAS data) to compute a vegetation index for each point of the floodplain, based on surface roughness, and used a 1-m land cover dataset. Regression equations based on the 1-D model, flood frequency, distance from channel, and vegetation account for approximately 50% of the variance in observed floodplain sedimentation rates. Sedimentation per flood averages about 0.1 cm/yr adjacent to the channel, but decreases to 0.005 cm/yr 200 m away. We find that sedimentation rates in forests are twice that of pasture. We will further discuss the improvement in our predictions by incorporating additional model complexity. These results can be used to improve sediment routing predictions for assessing management plans for restoring alluvial floodplains in general and tributaries to the Chesapeake Bay, specifically.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMEP12B..06L
- Keywords:
-
- 1815 Erosion;
- HYDROLOGYDE: 1820 Floodplain dynamics;
- HYDROLOGYDE: 1825 Geomorphology: fluvial;
- HYDROLOGYDE: 1862 Sediment transport;
- HYDROLOGY