Multi-Faceted Approach To Advancing Basin-Scale Sediment Source and Transport Models

Excess sediment is one of the leading causes for river reach impairment in the U.S. While the negative impacts of sediment on the environment and infrastructure are numerous, sediment also plays a critical role in maintaining a healthy aquatic ecosystem. To effectively mitigate water-quality impairment from excess fine sediment loading, it is imperative to understand and quantify the dominant sediment sources and transport processes that contribute fine sediment to streams. Both landscape and stream monitoring as well as modeling strategies can contribute to our understanding of the two important contributors of fine sediment, upland and in-channel sources. Landscape and stream monitoring efforts are often limited in geographic extent due to cost and field logistics. Whereas basin-scale sediment source and transport models can predict basin-scale sediment dynamics at an annual scale but often do not distinguish between upland and in-channel sediment sources. Improvement of basin-scale sediment models that can identify and estimate the contributions from upland and in-channel sources is needed to 1) better inform land managers to prioritize sediment mitigation strategies, and 2) understand and predict the effectiveness of these strategies as well as the transport and fate of sediment and sediment-related contaminants like salinity and phosphorus, and their response to climate change. As part of a new project to model suspended sediment in 3 geographically diverse regions—the Delaware River, Illinois River, and the Upper Colorado River—we have developed a multi-faceted approach to improve upon existing sediment source and transport models and move towards a modeling framework which will allow for the prediction of fine sediment at a range of geographic locations. First, we are using high-resolution DEMs to derive channel openness and slope-area indices to better describe the channel geomorphology. Second, we are examining hysteresis in the 3 basins where continuous turbidity and discharge data are available to understand sediment source dynamics, derive model variables to help predict sediment response to changing discharge, and to determine whether hysteresis characteristics can be generalized based on watershed and climatic data. Third, an AI-ML model is being built using daily climatic and hydrologic data along with watershed characteristics to improve data-driven insights on sediment sources and transport. Fourth, sediment budgets and fingerprinting analysis in smaller sub-basins will be analyzed and used to check model predictions of sediment source. Fifth, we are using available spatial data combined with an understanding of sediment transport process to derive a CONUS-scale map of dominant processes which will serve as another check on the model results and as future guide for building predictive sediment models. We are hopeful that these integrated efforts will enhance our ability to both understand and predict fine sediment sources and transport in a range of geographic locations.