Evaluation of the Acoustic Doppler Velocity Meter for Computation of Discharge Records at Three Sites in Colorado, 2004-2005

摘要

The U.S. Geological Survey (USGS), in cooperation with the Colorado Water Conservation Board, conducted a study in 20042005 at three sites in Colorado: Bear Creek at Morrison, Clear Creek near Empire, and Redlands Canal near Grand Junction. The study was done to evaluate acoustic Doppler velocity meter (ADVM) technology in different hydrologic settings that are characteristic of many Colorado streamflow-gaging sites. ADVMs have been tested and used extensively in many parts of the United States by USGS but not in Colorado where relatively small, shallow, clear, coarse-bed streams that ice up in the winter may affect the ADVM suitability. In this study, ADVM instrumentation was successfully used and discharge computations compared favorably, generally within 5 to 10 percent, with conventional USGS stage/discharge methods at the three Colorado sites. However, two factors, encountered in this study, may adversely affect the use of ADVM technology in Colorado. First, for some streams, the depth required (about 1.5 feet for a side-looking instrument) cannot be met during low-flow periods of the year. Second, cold temperatures and freezing-thawing cycles can produce ice effects that could prevent collection of usable ADVM (and stage) data. Some Colorado streamflow-gaging sites have conditions where ADVM measurements may enhance accuracy or ease of computation. Variable backwater from tributary inflow, beaver dams, ice effect, diversion dams, canal operations, or debris are problematic for stage/discharge records, but can be handled by ADVM methods using stage as a variable in the index-velocity equation if necessary. This technique allows for more than one discharge at a given gage height and can reduce or eliminate the need for shift correction for changes in the control. Study results at the three Colorado sites indicate that ADVM technology is a viable, though more expensive alternative to conventional USGS stage/discharge instrumentation. Much of the ADVM discharge computed for this study was within 5 percent of the USGS conventional stage/discharge, except at Clear Creek, where part of the ADVM discharge data were generally in the 5 to 10 percent range of percentage difference.