STREAM CHANNEL ADJUSTMENTS TO DAM REMOVAL IN HEADWATER CATCHMENTS

Understanding the geomorphic impacts of dam removal has become increasingly important as the pace of removals in the US has accelerated to ~60 per year. Most dams are removed to improve ecological habitat and/or to address a safety concern. These benefits may be offset by the detrimental effects on downstream reaches of sediment released from the former impoundment, which can be minimized by the design and construction of a stable channel through the former impoundment when the dam is removed. In this study we seek to predict the type and magnitude of geomorphic change and the locations where channel adjustments, both up- and downstream of dam removals, are most likely to occur. We hypothesize that stable channels are configured such that the ratio of characteristic water discharge, channel slope, width, and bed grain size approaches a characteristic value of the Shields parameter. It is well established that for gravel-bedded rivers the characteristic value corresponds to the conditions required for the onset of significant sediment transport. We compare our predictions of instability to field observations of two recent removals of small dams with sediment-filled reservoirs. We are able to document both immediate and sustained geomorphic adjustments. Pelham Dam on Amethyst Brook in Pelham, MA, a 6 m tall structure, was removed in fall 2012. Kendrick Dam, a 4 m structure on Sugar Hollow Brook in Pittsford, VT was removed in summer 2014. Both sites are gravel-bedded streams with drainage areas of ~25 km², slopes of 1-2%, and bankfull widths of ~10 m. One year after the removal on Amethyst Brook, median grain size had increased above the dam and decreased by 30-60% below the dam. Knickpoint migration above the former reservoir in both locations indicates instability of the channel following dam removal. On Amethyst Brook, the knickpoint migration ceased ~120 m upstream of the former dam at an older crib structure. We document sediment movement and channel adjustments in the field with repeat surveys of monumented cross-sections, longitudinal profiles, LIDAR imagery, embeddedness surveys, and pebble counts. Our results have important implications for river restoration and will provide another tool for designing future small dam removals.