Causes of systematic over‐ or underestimation of low streamflows by use of index‐streamgage approaches in the United States

Abstract

Abstract:Low-flow characteristics can be estimated by multiple linear regressions or the index-streamgage approach. The latter transfers streamflow information from a hydrologically similar, continuously gaged basin ('index streamgage') to one with a very limited streamflow record, but often results in biased estimates. The application of the index-streamgage approach can be generalized into three steps: (1) selection of streamflow information of interest, (2) definition of hydrologic similarity and selection of index streamgage, and (3) application of an information-transfer approach. Here, we explore the effects of (1) the range of streamflow values, (2) the areal density of streamgages, and (3) index-streamgage selection criteria on the bias of three information-transfer approaches on estimates of the 7-day, 10-year minimum streamflow (Q 7,10 ). The three information-transfer approaches considered are maintenance of variance extension, base-flow correlation, and ratio of measured to concurrent gaged streamflow (Q-ratio invariance). Our results for 1120 streamgages throughout the United States suggest that only a small portion of the total bias in estimated streamflow values is explained by the areal density of the streamgages and the hydrologic similarity between the two basins. However, restricting the range of streamflow values used in the index-streamgage approach reduces the bias of estimated Q 7,10 values substantially. Importantly, estimated Q 7,10 values are heavily biased when the observed Q 7,10 values are near zero. Results of the analysis also showed that Q 7,10 estimates from two of the three index-streamgage approaches have lower root-mean-square error values than estimates derived from multiple regressions for the large regions considered in this study. Published in

A comparison of methods for low streamflow estimation from spot measurements

“…The number of total measured flows (two per recession) examined in this experiment was 4, 6, 8, 10, and 12 days. Eng et al () used longer recessions (at least an 8‐day drop) and more baseflow measurements (20 measurements to estimate 10 values of τ) for regional regression with τ. A less stringent baseflow criterion and fewer measurements to reduce the total duration of sampling needed to obtain low streamflow estimates were used.…”

“…Baseflow correlation and scaling methods rely on a gauged donor site; selection of the donor site can have a large influence on the performance of low flow estimators (Eng et al, ; Laaha & Blöschl, ; Zhang & Kroll, ). There are a wide range of techniques that can be used to select donor sites such as the nearest neighbour (NN), drainage area within a certain range, annual precipitation within a certain range, map correlation method, and various other basin characteristics and spatial methods (Archfield & Vogel, ; Clark & Evans, ; Laaha & Blöschl, ; Ries & Friesz, ; Zhang & Kroll, ).…”

“…They explored the impact of the range of streamflow used with the index approaches, the areal density of gauges, and two donor site selection methods. Eng et al () found that baseflow correlation and a maintenance of variance method produced 7Q10 estimators with a lower bias than regional regression and that only a small portion of this bias is explained by the areal density of stream gauges and hydrologic similarity. Eng et al () used very large hydrologic regions, which may have adversely impacted the performance of regional regression.…”

“…Eng et al () found that baseflow correlation and a maintenance of variance method produced 7Q10 estimators with a lower bias than regional regression and that only a small portion of this bias is explained by the areal density of stream gauges and hydrologic similarity. Eng et al () used very large hydrologic regions, which may have adversely impacted the performance of regional regression. In addition, they only examined the use of 10 spot measurements at the partial record gauge.…”

Estimating minimum streamflow from measurements at ungauged sites in regions with streamflow‐gauging networks

“…Discrete streamflow measurements at ungauged sites have been used to improve regional estimation of low-flow statistics, but their value depends on the timing of the measurement and the hydrological similarity of streams in a region (Hirsch, 1982;Stedinger & Thomas, 1985;Laaha & Blӧschl, 2005;Eng & Milly, 2007;Seibert & Beven, 2009;Eng et al, 2011;Stagnitta, Kroll, & Zhang, 2018). We develop a method for using a single streamflow measurement at a site in conjunction with a network of continuous streamflow gauges to estimate minimum daily mean streamflow (minimum streamflow) for a specified period.…”