Streambank erosion downstream of a flood control dam: processes, rates, and management

Abstract

For nearly a century, large dams have played a major role in protecting and supporting human life by storing runoff for water supply, hydropower, and flood control. But as with exploiting any natural resource for anthropogenic purposes, dam installation comes with an environmental cost. This research answers the question: Can stream practitioners manage the unintended impact to channel instability downstream of large dams and if so, how? A case study was conducted along a 111 km reach downstream of Kanopolis Dam, a 1948-installed flood control dam located on the Smoky Hill River in central Kansas, USA. This study had three objectives, evaluate the effects Kanopolis Dam closure has had on downstream: (1) flow and channel bed stability (2) channel planform, and (3) streambank erosion processes and rates. Statistical analysis of historical flow data from four USGS stream gages on the Smoky Hill River showed that peak flows had dampened, resulting in a more step-like versus flashy hydrograph. Additionally, an increase in moderate daily flows between a 5 to 40% daily exceedance probability following dam installation up to 111 river km downstream was observed. Flow rating curves from USGS stream gages were used to assess bed stability at a low flow stage overtime. The bed stability analysis showed aggradation at the USGS gage upstream of the dam following dam closure, with degradation observed immediately downstream of the dam. Based upon a statistical forecasting method, the bed elevation has stabilized throughout the 111 km reach over the last decade. Channel planform was assessed using two metrics: (1) channel migration rates and (2) sinuosity. Based on GIS analysis of historical aerial imagery and mapping, channel migration rates, or the area of floodplain reworked per stream length per time, have decreased by 69% while channel sinuosity has increased by 5% since Kanopolis Dam closure. An increase in sinuosity is likely caused by an increase in lateral migration rates caused by bank erosion. Finally, using a HEC-RAS 1D quasi-unsteady hydraulic, sediment transport, and bank erosion model, it was shown that toe erosion is the dominant streambank erosion process for a site 86 river km downstream of Kanopolis Dam. Furthermore, if the dam was removed, predicted bank erosion would be higher under a no dam flow condition versus a with dam flow condition. Coupling model results with measured cross sections and observation, it was also demonstrated that toe erosion is occurring in other locations of the stream where bank erosion is typically less common, such as the inside of meander bends and along straight reaches, likely attributed to the step-like hydrograph brought on by flow regulation. System wide toe erosion occurring over longer periods of time along 111 river km downstream of Kanopolis Dam may also explain the observed increase in sinuosity since dam closure. Understanding the geomorphic context of a stream is invaluable to proper and effective management by stream practitioners. Two stream succession scenarios were observed on the Lower Smoky Hill River: (1) Rosgen (2014) stream succession scenario 1 within the degradational zone downstream of Kanopolis Dam and (2) a proposed stream succession scenario downstream of the degradational zone still influenced by flow regulation. Stream succession scenarios can be applied to assist in erosion management strategies and techniques following dam closure. Streambank erosion management techniques were also discussed and evaluated using the same HEC-RAS 1D model. Modeling results showed that a woody bankfull bench, in combination with 3H:1V bank shaping, would provide stability for at least a 40-year post-dam return interval flow event on the Smoky Hill River. This research demonstrates application of a popular and free numerical model to assist with assessing geomorphic change, as well as potential restoration techniques. Since closure in 1948, Kanopolis Dam has prevented nearly 2.5 billion dollars in downstream flood damage, proving its value and need. While large dams are and will continue to be an important part of our landscape and livelihood, it is imperative that the environmental cost of dam operation is understood and evaluated. Future research and monitoring efforts should continue to assess and valuate damages induced by dams, especially under a changing climate, to inform and adapt river management and dam operation through time.