Quantifying uncertainty in historic long-profiles of the Colorado River: fine sediment evacuation downstream from Glen Canyon Dam, 1965-2009
Location
Eccles Conference Center
Event Website
http://water.usu.edu/
Start Date
3-30-2011 10:15 AM
End Date
3-30-2011 10:20 AM
Description
Glen Canyon Dam has fundamentally changed the source, supply, and caliber of sediment carried by the Colorado River through Glen, Marble, and Grand Canyons. The first survey of the thalweg over the 225 river miles (365 km) within these three canyons was led by Luna Leopold in 1965. We digitized this survey and other surveys acquired in 1976, 1984, 2008, and 2009 in order to characterize downstream changes in pool depths (and thus changes in sand storage on the bed). Paper traces were photographed, rectified, and converted to vectors in ArcScan and then rescaled by tying to persistent debris flow formed rapids. Prior to 2009, surveys were located using aerial photographs-GPS was not operable in the canyon until the most recent survey. We accounted for uncertainties in boat location (both down- and cross-channel) by randomly sampling high-resolution digital elevation data sets corresponding to ten long-term monitoring sites surveyed up to seven times between August, 2000 and May, 2009 and fluctuating around the path recorded in 2009. Depths were determined by converting each bed elevation to the modern datum, and subtracting this from modeled water surface profiles. We selected the instantaneous discharge at Lees Ferry for the sample time and then routed this to the sample location using a reach-averaged model of wave propagation specifically developed for the Colorado River. Mean and maximum depths were then calculated for thirty-two pools (defined as above the reach average depth), with between 1-7 pools per site. Despite the high degree of uncertainty associated with these historic datasets, we were able to resolve patterns of change in pool sand storage downstream of Glen Canyon Dam. Our analysis corroborates observations made in Glen Canyon that pools were evacuated first by the pulse flows released in 1965 and then completely in 1983 following high flows released to prevent overtopping of Glen Canyon Dam. Almost all of the pools reached a maximum depth between 1965 and 1984-since then, there is no long-term trend observed in 66% of the pools, and the remaining 33% have filled nearly to 1965 conditions. The 1965 long-profile was collected during the final of fourteen 'flushing' flows released by the Bureau of Reclamation to scour the bed of the Colorado River. Our analysis suggest that these pulse releases had already scoured out much of the pre-dam sand once stored in pools in the main channel of the Colorado.
Quantifying uncertainty in historic long-profiles of the Colorado River: fine sediment evacuation downstream from Glen Canyon Dam, 1965-2009
Eccles Conference Center
Glen Canyon Dam has fundamentally changed the source, supply, and caliber of sediment carried by the Colorado River through Glen, Marble, and Grand Canyons. The first survey of the thalweg over the 225 river miles (365 km) within these three canyons was led by Luna Leopold in 1965. We digitized this survey and other surveys acquired in 1976, 1984, 2008, and 2009 in order to characterize downstream changes in pool depths (and thus changes in sand storage on the bed). Paper traces were photographed, rectified, and converted to vectors in ArcScan and then rescaled by tying to persistent debris flow formed rapids. Prior to 2009, surveys were located using aerial photographs-GPS was not operable in the canyon until the most recent survey. We accounted for uncertainties in boat location (both down- and cross-channel) by randomly sampling high-resolution digital elevation data sets corresponding to ten long-term monitoring sites surveyed up to seven times between August, 2000 and May, 2009 and fluctuating around the path recorded in 2009. Depths were determined by converting each bed elevation to the modern datum, and subtracting this from modeled water surface profiles. We selected the instantaneous discharge at Lees Ferry for the sample time and then routed this to the sample location using a reach-averaged model of wave propagation specifically developed for the Colorado River. Mean and maximum depths were then calculated for thirty-two pools (defined as above the reach average depth), with between 1-7 pools per site. Despite the high degree of uncertainty associated with these historic datasets, we were able to resolve patterns of change in pool sand storage downstream of Glen Canyon Dam. Our analysis corroborates observations made in Glen Canyon that pools were evacuated first by the pulse flows released in 1965 and then completely in 1983 following high flows released to prevent overtopping of Glen Canyon Dam. Almost all of the pools reached a maximum depth between 1965 and 1984-since then, there is no long-term trend observed in 66% of the pools, and the remaining 33% have filled nearly to 1965 conditions. The 1965 long-profile was collected during the final of fourteen 'flushing' flows released by the Bureau of Reclamation to scour the bed of the Colorado River. Our analysis suggest that these pulse releases had already scoured out much of the pre-dam sand once stored in pools in the main channel of the Colorado.
https://digitalcommons.usu.edu/runoff/2011/Posters/29