Evaluation of Stream Bank Restoration to Improve Water Quality in a Semi-Arid Stream

Human watershed activities such as converting land cover to agriculture and livestock grazing have negatively impacted stream water quality worldwide. One such case is Utah’s Upper Sevier River where a loss of woody bank vegetation (reduced shading) and accelerated bank erosion (increased fine sediment inputs) has led to increased stream temperature and water turbidity. As a result, the state of Utah sought to improve water quality conditions using streambank restoration. While commonly recommended and performed, the effectiveness of this sort of restoration has rarely been quantified. Here, I evaluated a restored reach of the Upper Sevier River near Hatch, UT using continuous monitoring data and a historical photo analysis. As Utah wishes to continue performing this type of restoration in additional locations on the Upper Sevier River, I applied a simple sediment budget model to test its value in informing future streambank restoration decisions. Continuous monitoring data at the upstream and downstream extent of restoration showed that both stream temperature and turbidity increased downstream along the restored reach. In addition, I found that stream temperature violated Utah’s cold-water stream threshold at both sites but did not violate thresholds for rainbow trout. Turbidity violated state and biological thresholds at both sites. I was unable to conclude whether the streambank restoration directly altered water quality because I lacked monitoring data before restoration occurred. Results of the historical aerial photo analysis showed that restoration practitioners were successful in reducing cut bank erosion. My use of SIAM as a simple sediment budget model proved insufficient due to poor data quality and quantity. Overall, streambank restoration was successful at reducing cut bank erosion, and I recommended monitoring future restoration before and after project completion, identifying and monitoring upstream sources of fine sediment, and pursuing more comprehensive sediment models to inform future streambank restoration