DOWNSTREAM CHANNEL RESPONSE TO DAM REMOVAL AT REACH AND BEDFORM SCALES, WHITE SALMON RIVER, WASHINGTON

Dam removals provide an opportunity to better understand the forcings, sensitivities, and changes in transport capacity that dictate river response to large sediment pulses. On October 26, 2011, the 98-year-old Condit Dam on the White Salmon River, Washington, was abruptly breached by an explosives blast. The ensuing rapid drainage of water and sediment from the reservoir produced a sediment pulse that draped the downstream river with silt and sand within hours. We investigate the evolution of the White Salmon River as a result of this disturbance, from pre-breach conditions to the summer (~9 months) after the breach, using repeat, pre- and post-breach surveys of topography and grain size, combined with calculations of transport capacity. We test the hypothesis that, at the reach scale, a 4-km, steep, confined reach immediately downstream of the dam will return to pre-breach conditions of bed elevation, slope, and grain size more rapidly than a 1.3-km, low-gradient, less-confined reach that is influenced by its confluence with the Columbia River. We also test the hypothesis that, at the bedform scale, riffles will return to pre-breach conditions more rapidly than pools. In the confined reach, median grain size from pre-breach surveys in August 2011 to post-breach surveys in August 2012 decreased from 160 to 50 mm, suggesting an increase in transport capacity that will facilitate sediment evacuation from the confined reach. Water surface slopes over pool-riffle sequences became more uniform after the breach, and sediment deposition in pools has persisted through summer 2012. Average slopes increased from 0.001 to 0.006 over pools and decreased from 0.032 to 0.012 over riffles, indicating a reduction in bedform roughness that may also increase transport capacity. Comparative surveys of bed aggradation at the former dam site and the mouth of the White Salmon show a readjustment toward pre-breach bed elevations near the dam, but persistent aggradation near the mouth, supporting our hypothesis that the less confined reach will be more sensitive than the confined reach.