Evaluating suspended sediment and turbidity reduction projects in a glacially conditioned catchment through dynamic regression and fluvial process-based modelling

Elevated turbidity (T_n) and suspended sediment concentrations (SSC) during and following flood events can degrade water supply quality and aquatic ecosystem integrity. Streams draining glacially conditioned mountainous terrain, such as those in the Catskill Mountains of New York State, are particularly susceptible to high levels of T_n and SSC sourced from erosional contact with glacial-related sediment. This study forwards a novel approach to evaluate the effectiveness of stream restoration best management practices (BMPs) meant to reduce stream T_n and SSC, and demonstrates the approach within the Stony Clove sub-basin of the Catskills, a water supply source for New York City. The proposed approach is designed to isolate BMP effects from natural trends in T_n and SSC caused by trends in discharge and shifts in average T_n or SSC per unit discharge (Q) following large flood events. We develop Dynamic Linear Models (DLMs) to quantify how T_n-Q and SSC-Q relationships change over time at monitoring stations upstream and downstream of BMPs within the Stony Clove and in three other sub-basins without BMPs, providing observational evidence of BMP effectiveness. A process-based model, the River Erosion Model, is then developed to simulate natural, hydrology-driven SSC-Q dynamics in the Stony Clove sub-basin (absent of BMP effects). We use DLMs to compare the modelled and observed SSC-Q dynamics and isolate the influence of the BMPs. Results suggest that observed reductions in SSC and T_n in the Stony Clove sub-basin have been driven by a combination of declining streamflow and the installed BMPs, confirming the utility of the BMPs for the monitored hydrologic conditions.