VARIABLE GEOMORPHIC RESPONSES TO DAM REMOVAL: THE EFFECT OF RESERVOIR WIDTH ON POST-REMOVAL EROSION

Dam removals occur in a variety of settings, and site specific conditions influence geomorphic responses in former reservoirs. Recent work suggests wide and narrow reservoirs are categorically different in the percent of stored sediment exported following dam removals. However, the physical controls on these differing responses have not been investigated thoroughly. Here we compare field evidence one month prior to and up to five years following the removal of two contrasting run-of-river dams in New England.

The Monstweag dam in Wiscasset, Maine was 10.7 m high and formed a reservoir ~ 1.2 km long, ~ 60 to 100 m wide, and ~ 10 times the width of the upstream channel. The post-removal channel in the former reservoir initially incised up to 2 m into unconsolidated sediment and subsequently widened to ~ 11 m. The width is similar to the upstream channel width and significantly less than the width of the former reservoir. Channel meandering and avulsions were minimal. The post-removal channel could only access ~ 1/10^th of the former reservoir bed, which greatly impeded export of stored sediment. The geomorphic response was consistent with the channel evolution model of Simon and Hupp (1986).

In contrast, the Homestead dam in West Swanzey, New Hampshire was 4 m high with a reservoir ~ 9.5 km long, ~ 40 m wide, ~ 1.5 times the upstream channel width. In this narrow reservoir, bed incision and bank erosion averaged 10.5 cm and 7.5 cm, respectively, substantially less than at the wide Monstweag reservoir. The post-removal channel in the former Homestead reservoir was approximately the same width as the former reservoir, and the post-removal flows had better access to sediment stored in the reservoir. Here the geomorphic response differed from the channel evolution model and was better characterized by modeled sediment transport. At both sites, the lack of high-precision pre-dam surveys makes it difficult to determine exact values of stored sediment and the percent exported. However, it is clear that access to stored sediment by the post-removal channels influenced the rate and amount of sediment exported from the former reservoirs. This study improves understanding and predictions of the variable geomorphic responses to dam removals and helps evaluate costs and benefits of dam removals as restoration tools.