Uncertainties in streamflow measurement under winter ice conditions a case study: The Red River at Emerson, Manitoba, Canada

Abstract

Computed streamflows on the Red River at Drayton in North Dakota, United States, have been generally higher under winter conditions than computed streamflows on the Red River at Emerson in Manitoba, Canada. The station at Drayton is located 85 km upstream from Emerson. Moreover, historically, the U.S. Geological Survey (USGS) has measured higher streamflows under ice conditions than the Water Survey of Canada (WSC) on the Red River at Emerson. The difference between their discharge measurements under ice conditions is generally between +10 and 20%. To explain these differences, field and laboratory experiments were conducted in 1986, which included an international field metering program at Drayton and Emerson, rating of current meters and suspension equipment at national calibration services in Canada and the United States, and testing of current meters in flowing water. Through the analysis of the field and calibration data it was determined that significant differences exist between a current meter rated on rod and on cable suspension, and that this suspension effect is responsible for the large differences in measurements between the USGS and WSC. Changes in discharge measurement techniques under ice conditions are suggested, including the sampling of vertical velocity curves and increasing the time of exposure for current meter observations. Several investigations are suggested to improve the overall quality of streamflow records for the winter ice period.