HR: 0800h
AN: H31E-1350    [Abstracts]
TI: Evaluation of the SHOALS 1000T Bathymetric LIDAR System for Monitoring Channel Sediment Within the Colorado River in Arizona
AU: * Davis, P A
EM: pdavis@usgs.gov
AF: U.S. Geological Survey, 2255 N Gemini Drive, Flagstaff, AZ 86001 United States
AU: Gonzales, F M
H31E-1350 AF: U.S. Geological Survey, 2255 N Gemini Drive, Flagstaff, AZ 86001 United States
AU: Brown, K M
H31E-1350 AF: U.S. Geological Survey, 2255 N Gemini Drive, Flagstaff, AZ 86001 United States
AU: Melis, T S
H31E-1350 AF: U.S. Geological Survey, 2255 N Gemini Drive, Flagstaff, AZ 86001 United States
AB: The Grand Canyon Monitoring and Research Center of the U.S. Geological Survey monitors sediment transport and storage within the Colorado River ecosystem in Arizona in order to develop flow protocols for the Glen Canyon dam that preserve or restore aquatic and terrestrial habitats. Currently, monitoring the channel sediments is accomplished using acoustic multi-beam surveys, which are very time-consuming. We explored more efficient collection systems that could provide a 2-m point spacing and a vertical accuracy of 25 cm or better. The dual-beam SHOALS 1000T LIDAR system, which simultaneously collects bathymetric and topographic data, could meet these requirements if flown at a 300 m altitude. This low altitude required this fixed-wing system to be modified for helicopter collection to navigate the sinuous, steep-walled canyon. We tested the helicopter-based SHOALS on two segments of the Colorado River - the San Juan River confluence at Lake Powell and the southern portion of Glen Canyon near Lees Ferry. The test flights occurred in late November after a high-flow dam experiment. Early winter storms injected such large volumes of sediment into the unmanaged San Juan River that the SHOALS green laser could not penetrate the water's surface. The water at Lees Ferry was relatively clear (Secchi depths of 7-7.5 m) because there are no tributaries between Lees Ferry and Glen Canyon dam and because the dam maintained a low steady flow for a week following the high-flow experiment in order for ground and aerial surveys to collect monitoring data. At a 300 m altitude, the SHOALS scanner produced a 60-m ground swath. Seven separate flight lines were collected - more than necessary to cover the 100-m-wide channel. Three GPS stations were operated within 30 km of the test flights. Within the areas of overlap between each pair of the seven SHOALS flight lines we found the reproducibility of the SHOALS data to be 19 cm in the channel and 21 cm on land. At Lees Ferry, 20 land sites were surveyed with vertical accuracies of 3-5 cm. The river channel was surveyed using an acoustic multibeam system, which produced normally distributed data with a vertical RMSE of 12-25 cm (all RMSE values reported herein are at the 95% confidence level). Our comparisons of the SHOALS land and channel data to our ground truth data showed that the SHOALS ellipsoid-height data had a vertical RMSE of 35 cm on bare ground with slopes less than 12 degrees and a vertical RMSE of 26 cm in the channel. The frequency distribution of vertical error in the SHOALS bathymetric data had a kurtosis of 0.7 and a skewness of 0.01. Vertical error was largest near the shallowest and deepest parts of the channel. We found that the accuracy of the SHOALS test data to be adequate for monitoring channel sediment in the Grand Canyon and that the SHOALS system provided more rapid data collection and processing than our acoustic approach. Although SHOALS did not provide accurate data near the river banks in water depths less than 1 m, due to the combined inherent inaccuracies of both of its lasers in shallow water, the acoustic system has the same limitation. We found that SHOALS could accurately map the channel substrate in the Lees Ferry study area down to a depth of 20 m or 3 Secchi, which is at the upper end of the Secchi range reported previously by researchers. However, the channel only exceeds this depth in small, isolated pockets. The main limiting factor to routine deployment of SHOALS in the Grand Canyon is the 300 m flight altitude necessary to provide our required 2 m spot spacing. At this low altitude, the noise from the helicopter is a serious concern in this wilderness area.
DE: 1855 Remote sensing (1640)
DE: 1862 Sediment transport (4558)
DE: 1895 Instruments and techniques: monitoring
SC: Hydrology [H]
MN: Fall Meeting 2005