Evaluating Bank Restoration and Stabilization in the Yellow River in Northern Indiana: Mitigation of Sediment and Nutrient Transport

Aleia Dumond , Brittany Hanrahan, Aubrey Bunch

The U.S. Geological Survey, in cooperation with the Kankakee River Basin and Yellow River Basin Development Commission, evaluated the effects of streambank remediation on the Yellow River in northwest Indiana. Two phases of remediation were completed on the Yellow River between 2021 and 2023, after assessments indicated it was a large contributor of sediment to the Kankakee River. Temporal effects were assessed using pre‑remediation (April 2019 –October 2021) and post‑remediation (February – October 2024) data from two long‑term monitoring sites, USGS 05516665 Yellow River near Oak Grove (Oak Grove) and USGS 05517000 Yellow River at Knox (Knox). Spatial patterns were evaluated among sites from upstream to downstream during the post-remediation period. Monitoring included the pre-existing Oak Grove and Knox sites, an additional monitoring site at USGS streamgage 05516500 Yellow River at Plymouth (Plymouth), and discrete sampling at two additional sites. Discrete sites include USGS 05516663 Yellow River at Upas Rd near Oak Grove (Upas Rd), located upstream of the remediation reach, and USGS 05516667 Yellow River at CR11100E near Oak Grove (CR1100E), located within the remediation reach.
Generalized least squares and linear mixed‑effects regression models were used to evaluate temporal differences in turbidity-to-discharge relations for slope and turbidity units, discharge, and suspended‑sediment in before and after remediation data. Similar analysis using generalized least squares and linear mixed‑effects regression models were conducted for spatial analysis of discharge and sediment and nutrient concentrations and fluxes at each site post-remediation. Temporal comparisons showed that turbidity increased less with discharge increases after remediation, indicated by slope of relation, at both Oak Grove and Knox, indicating improved water clarity during higher flows. Mean turbidity and suspended‑sediment concentrations were not statistically different between remediation periods.
Comparison of the post-remediation continuous water quality data from upstream to downstream sites through the remediated reaches indicated that downstream sites exhibited steeper turbidity‑to‑discharge slopes than the upstream control site at Plymouth. A statistically significant difference was detected between the slopes at Oak Grove and Plymouth, whereas Knox did not differ significantly from either Plymouth or Oak Grove. Downstream sites had significantly lower mean turbidity, suggesting improved water clarity within and downstream of the remediation reach. Nutrient concentrations generally decreased moving downstream and suspended‑sediment concentrations showed no consistent trend. Nutrient and sediment fluxes increased downstream, likely reflecting the influence of greater discharge volumes.
Interpretation of remediation effects was limited by the absence of pre‑remediation data at upstream sites and by the short post‑remediation monitoring period. Continued monitoring could allow for assessment of longer‑term responses to streambank remediation on the Yellow River, as effects may become more pronounced over time.