A watershed scale numerical model of the impact of land use change on bed material transport in suburban Maryland, United States

I have developed a numerical model for routing bedload through channel networks. The model treats the network as a series of connected reaches, with each reach being tens of channel widths in length. Processes represented within each reach include bank and bed erosion and deposition, bedload inputs from upstream, and bedload transport out of the reach. The rate of bank erosion is proportional to the percentage of the upstream watershed area under construction, and bank erosion supplies additional bedload to the reach. Inputs from upstream are assumed to be known, either as boundary conditions or from a previous computation at the adjacent upstream reach. The numerical model is intended to simulate the evolution of river channels in response to watershed scale changes in land use and climate change. The performance of the model has been evaluated, with satisfactory results, by comparing the simulated values and the measured data from studies that represent a watershed scale sediment budget, and a laboratory flume experiment. The field data used to evaluate the model comes from the Good Hope tributary, a second-order stream that is located in the Anacostia River watershed. Simulation with the model reproduced most of the major features of the field measurements, including sediment budget data and inferred patterns of erosion and deposition. In the Good Hope Tributary the streambed has in most cases responded to land use stress by longitudinal profile adjustment and grain size coarsening, especially in the 85 th percentile. A version of the model without tributaries predicted increased bed erosion when compared with the network model. Comparison of the network and the single channel model results indicates limited to approximately 200 meters impact of lateral inputs of sediment from tributaries on grain size distribution of the main channel. Finally, numerical runs of the model simulating field conditions from 1952 to 2042 indicate that the river morphology is still adjusting fluid and sediment transport to new urbanized settings. Grain size distributions predicted by the model indicate that the process of grain size adjustment will probably continue for decades after the disturbance from construction has stopped.