Accuracy of firm yield estimates of direct supply reservoirs with limited data

Open Access
Publisher Website
Google Scholar
Abstract
The error of the estimate of the yield of a direct supply reservoir due to limited data was investigated for conditions prevailing in the northwest of Europe using the error of the 2% design drought for durations up to 36 months as a measure of this error. Two methods of estimating the 2% drought were considered: the monthly design drought and an empirical procedure which requires only rainfall data. The errors for the design drought were found by splitting a 500-year monthly flow record into shorter records; the 500-year record was obtained by extending an historic record of 111 years using the Thomas-Fiering model. To estimate the errors of the empirical procedure, a widely used method of Law (1953), which uses rainfall data and the mean annual losses to estimate the design drought, was applied to 13 catchments with historic records greater than 30 years. The error of the empirical procedure was found from the differences between the empirical procedure and the estimate from the historic data. When the results from the two analyses were compared, the accuracy of the empirical procedure was for all durations equal to the accuracy obtained from a record of between 10 and 20 years.