USGS revises Skagit River flood numbers

Scientists have refined calculations of some historic Skagit River floods so that they are slightly lower than previously calculated, but still within the plus—or—minus 15—percent margin of error of previous calculations, according to a report released today by the U.S. Geological Survey (USGS).

The USGS will present the results of the report to the Skagit County Board of Commissioners on Monday, August 13, from 1:30 to 2:30 p.m. in the Skagit Room of the Continental Building, 1800 Continental Place, Mt. Vernon, Wash.

One of the previous calculations, performed following the 1921 Skagit River flood by USGS engineer James Stewart, is critical for determining the 100—year flood for the river. The 100—year flood, or the flood flow with a one—percent chance of being exceeded in a given year, is used by the U.S. Army Corps of Engineers to delineate areas vulnerable to flooding and design flood—protection infrastructure.

After considering multiple lines of evidence, including new data and new hydraulic calculation procedures, the USGS concluded that Stewart’s calculated 1921 peak flow of 240,000 cubic feet per second (cfs) should be revised downward by about five percent, to 228,000 cfs. The estimates for the floods of 1897, 1909, and 1917 are also to be revised downward because they were based on the 1921 calculation.

“These four historic floods are larger than any flood since our streamgage near Concrete, Wash. began recording flows in 1924,” said Mark Mastin, USGS hydrologist and author of the report. “They have a significant influence on the estimation of the 100—year flood.”

“With new flood data now available, several lines of evidence all suggest that James Stewart’s calculations of the peak flow in 1921 is slightly high, so we have revised the number slightly downward to what we consider to be the best estimate of the historic peak flow.” The old flood estimates were based on an assessment of channel roughness for a site about 30 miles down stream of the streamgage near Concrete while the new estimates employ roughness data that were determined from field measurements made at the streamgage in 1950.

At the center of attention is the magnitude of flooding on the Skagit River at the USGS stream gage near Concrete, Wash. Located in a bedrock gorge where all of the Skagit River funnels through, gaging began there by the USGS in 1924, just a few years after the major flood of December 1921. Soon after the 1921 flood, Stewart began his investigation by surveying flood marks from the current and previous floods. Based on these flood marks, Stewart calculated the magnitude of the 1921 flood at 240,000 cfs.

The report, “Re—evaluation of the 1921 Peak Discharge at Skagit River near Concrete, Washington,” by M.C. Mastin, is published as U.S. Geological Survey Scientific Investigations Report 2007—5159. The report is available on the Internet at pubs.usgs.gov/sir/2007/5159/.