UNCERTAINTY ANALYSIS OF RUNOFF ESTIMATES FROM A RUNOFF CONTOUR MAP¹

Abstract

The U.S. Environmental Protection Agency (EPA) in cooperation with the U.S. Geological Survey (USGS) conducted an analysis to quantify the uncertainty associated with interpolating runoff to specific sites using a runoff contour map. We interpolated runoff to 93 gaged watersheds from a runoff contour map using (1) hand interpolation to the watershed outlet, (2) a computer interpolation to the watershed outlet, and (3) hand interpolation to the watershed centroid. We compared the interpolated values to the actual gaged values and found that there was a bias in the average interpolated value for runoff estimated at basin outlets, with interpolated values being less than the actual. We found no significant difference between the hand interpolation method and the computer interpolation method except that the computer method tended to have higher variability due to factors inherent to the software used. There were no strong spatial correlations or regional patterns in the runoff interpolations, which indicates that there are no regional biases introduced in the development of the contour map. We determined that we could estimate runoff, on the average, within approximately 8.9 cm (3.5 in; 15 percent) of the measured value using the three methods. The results of this work indicate that runoff contour maps can he used in regional studies to estimate runoff to ungaged systems with quantifiable uncertainty.

Regional estimation of base flow for the conterminous United States by hydrologic landscape regions

“…Interpolation was also used by Gebert et al (1987), when they developed a spatially complete flow map of the US. Rochelle et al (1989) have also concluded that interpolation from a total flow contour map provides adequate estimates for un-gauged systems and recommended using total flow contour maps for environmental issues at regional scale.…”

“…Base flow volume was then estimated by multiplying total flow volume map with digital filter base flow index map. For the total flow interpolation, a study by Rochelle et al (1989) on the uncertainty analysis of total flow estimates from the flow contour map has shown that uncertainty could be about 15% due to interpolation. Interpolation was also used by Gebert et al (1987), when they developed a spatially complete flow map of the US.…”

Ordinary kriging as a tool to estimate historical daily streamflow records

“…As researchers gained access to higher-powered computers, efforts were made to automate the development of maps of mean annual runoff (Langbein and Slack, 1982). In both Europe and the United States, maps of mean annual runoff generated by geostatistical techniques were found to be as accurate as their hand-drawn predecessors (Rochelle et al, 1989;Domokos and Sass, 1990;Church, 1992, 1995). Mapping techniques have also been used to explore other streamflow statistics (Gottschalk et al, 2006;Archfield et al, 2013) and to assess the accuracy and performance of hydrologic models (Sauquet and Leblois, 2001).…”

EFFECTS OF IMPROVED PRECIPITATION ESTIMATES ON AUTOMATED RUNOFF MAPPING: EASTERN UNITED STATES1

“…To assess the accuracy of the various runoff maps, we conducted a quantitative uncertainty analysis employing the techniques and data used by Rochelle et al (1989) and Bishop and Church (1995) in their analysis of runoff estimates made from long-term average runoff contour maps. The analysis consisted of estimating runoff by interpolation from the maps to the watershed centroids of a set of gage sites withheld from the creation of the maps to be tested and then comparing the interpolated and actual values.…”

“…Our goal was to determine what, if any, improvements in accuracy could be derived from using the 5-km resolution estimates. We use two of the evaluation techniques employed by Rochelle et al (1989), and Bishop andChurch (1992, 1995) in similar studies: visual comparison and a quantitative uncertainty analysis using gaged sites withheld from the creation of any of the maps examined. Long-Term (1951-1980 Runoff for the Part of the Eastern United States Examined in this Study (source: Krug et al, 1990).…”