USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2007–5159

Back to Table of Contents

Summary

In 1952 and 1954, the U.S. Geological Survey investigated a revision of the historic floods at Skagit River near Concrete based on a revision of the 1921 peak-flow discharge that used a revised channel roughness coefficient calculated from the 1949 peak discharge. At the time, the recomputations were not used to change the published values because the proposed revisions were all less than 10 percent. Recently, a new channel roughness coefficient verification was made using the water-surface slope of the 2006 flood at Skagit River near Concrete along with channel geometry and high-water marks from the 2003 flood. The recalculation of the 1921 peak discharge using this new roughness coefficient and a subdivision of the cross sections also indicated that the original computation of the 1921 peak discharge by James Stewart in 1923 was too high although well within the margin of error. However, the island/bar in the slope-area reach thought to have been barren during the 1921 and 1949 peak discharges has re-vegetated into a dense forest, and it is likely altering the flow patterns during high flows. The slope-area computation does not solve for these complex flow patterns in two dimensions, and therefore, the recalculation based on a verified channel roughness coefficient using the 1949 flood data and channel characteristics and subdividing techniques is thought to more accurately reflect 1921 conditions. The recalculated peak discharge for the 1921 flood using the channel roughness coefficient verified using 1949 flood data is 5.0 percent less than the published value computed in 1923. The current stage-discharge relations at the station also indicate that the lower peak discharge value is more reasonable than the published value. With these several lines of evidence that all indicate that the published value of the 1921 peak is too high, the USGS will revise the historical peak discharges. The 1921 peak discharge will be revised to 228,000 ft³/s or 5.0 percent less than the published value, and correspondingly, the other historical peak discharges will be revised from straight-line extension of a revised stage-discharge rating that passes through the revised 1921 peak-discharge data point.

Back to Table of Contents