From Free-Flowing to Fragmented: Using Calibrated Models to Assess Impacts of Multiple Dams on Watershed Evolution

Over the past few centuries, the natural flow of river water and sediment has been significantly disrupted by human activities, including land‐use change, dam and reservoir construction, and variable precipitation. Sediment accumulation in reservoirs leads to declining storage capacity, reduced water quality, and navigational challenges. While hydraulic models can characterize these issues over annual to decadal timescales, they are less effective for predicting sedimentation trajectories over decades to centuries. At these longer timescales, feedbacks between reservoir sedimentation and upstream erosion and deposition influence delta growth and sediment delivery, complicating the development of long-term sediment management strategies. To address this gap, we developed a workflow for building and calibrating open source, coupled landscape evolution models (LEMs) and delta sedimentation models (DSMs) for real-world watersheds. Here, we present preliminary results from the Chattahoochee River in the southeastern United States. The river is segmented by six major dams, each creating a reservoir and corresponding sub-catchment. We constructed a set of six LEMs using Landlab (one for each sub-catchment) and run them sequentially from upstream to downstream, using the sediment outflux from each model as input for the next. The LEMs are calibrated using cosmogenic ¹⁰Be mean catchment erosion rates, modern land-use data, and sediment trapping calculations. We then evaluated how well each model reproduced watershed sediment fluxes inferred from late 21st century suspended-sediment measurements. The DSMs are constructed using PyDeltaRCM and are driven by output sediment flux and provenance data from the LEMs. Using the pre-reservoir topography as a boundary condition, we validate the models by replicating the post-reservoir delta growth. We then use variable land use and hydraulic forcings in the LEMs to assess different future sedimentation patterns in the deltas. Our workflow can be easily applied to any reservoir with bathymetric data and can help stakeholders understand how upstream human impacts may influence a range of possible sedimentation patterns over the coming decades.