Barry P. Rochelle | USGS Science

USGS2007

Relationship between annual runoff and watershed area for the eastern United States

As part of the U.S. Environmental Protection Agency's effort to determine the long-term effects of acidic deposition on surface water chemistry, annual runoff was estimated for about 1000 ungaged sites in the eastern U.S. using runoff contour maps. One concern in using contour maps was that a bias may be introduced in the runoff estimates due to the size of the 1000 ungaged sites relative to the size of the watersheds used in developing the maps. To determine if a bias was present the relationship between the annual runoff (expressed as depth) and the watershed area for the Northeast (NE) and Southern Blue Ridge Province (SBRP) was tested using five regional data bases. One short-term data base (1984 Water Year, n = 531) and two long-term data bases (1940–57, n = 134 and 1951–80, n = 342) were used in the NE. In the SBRP one short-term database (1984 Water Year, n = 531) and one long-term data base (1951–80, n = 60) were used. For the NE and the SBRP, runoff was not directly correlated with watershed area using the five regional databases. Also, runoff normalized by precipitation was not related to watershed area.

northeastern United States↗

USGS1990

Map of mean annual runoff for the Northeastern, Southeastern, and Mid-Atlantic United States, water years 1951-80

A map of mean annual runoff for States within the Northeastern, Southeastern, and Mid-Atlantic United States was prepared by the U.S. Geological Survey for the Direct/Delayed Response Project being conducted by the U.S. Environmental Protection Agency. This map shows mean annual runoff during water years 1951-80. Mean annual runoff from the northeastern region during 1951-80 ranged from less than 12 to greater than 40 inches. Runoff from the southeastern region runoff ranged from less than 12 to greater than 55 inches. In the mid-Atlantic region ranged from less than 10 to greater than 40 inches. Error analysis using 93 gaging stations not used for preparing the runoff map showed that the runoff map could be used to predict runoff with an average error of less than 10%. Errors in runoff estimation averaged about 12% if the locations of the gaging stations were used to estimate runoff instead of computing an area-weighted average over the basin. If the locations of the gaging stations were used to estimate runoff, there also was a significant negative bias of the errors that did not occur if the centroid or a weighted average runoff of the drainage basin were used. The runoff map is expected to be more accurate in areas that have a relatively high concentration of gaging stations and little topographic variability, such as part of the Northeast. Based on these criteria, the least reliably mapped areas would be in the Smokey Mountains along the North Carolina- Tennessee border.

Water-Resources Investigations Report↗