By Briana Edmunds
Faculty Mentor: Professor Pamela Grothe
Excess sediment runoff, as a result of anthropogenic activity, is one of the major contributors to the pollution of the Chesapeake Bay, Rappahannock River, and Hazel Run. To reduce the sediment entering different watersheds, different best management practices (BMPs) have been implemented. Agencies like the Chesapeake Bay Program and United States Geological Survey use models to predict how effective different BMPs are. Traditional models used by the USGS like ESTIMATOR use streamflow-based regression. However, regression relations fail to account for the variableness of sediment transfer during storm events (Jastram et al., 2009). Leigh et al. (2019) concluded that turbidity-based models which include temporal autocorrelation and heteroscedasticity were the most accurate and precise in predicting sediment and nutrient concentrations from high-frequency water-quality data, which we will be following. The Rappahannock River’s possession of only one monitoring station and how both it and Hazel Run have degraded water quality make it appropriate to conduct a pilot study. We will be utilizing in-situ sensors, which collect and store data, compared to manual samples. Using in-situ measurements allows data to be frequently collected, even during high-transport events like storms and floods, giving better records of the water quality. The goal of this capstone is to detail a pilot study project and complete a grant proposal to apply for the Jeffress Trust Awards Program in Interdisciplinary Research grant. The project in question will be a pilot study utilizing a proxy-model methodology using the deployment of many in-situ sensors to monitor and predict sediment and nutrient load in Hazel Run, a tributary of the Rappahannock River.