Refinement of a regression-based method for prediction of flow-duration curves of daily streamflow in the conterminous United States

Over, Thomas M.; Farmer, William; Russell, Amy M.

doi:10.3133/sir20185072

Cited by 8 publications

(12 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Streamflow data and basin characteristics from the Gages-II data set from all reference gages within a 400 km radius of the target site were used to fit the regression equations. Regression equations for the 27 exceedance probabilities were fitted at each validation site using an automated regression fitting and selection process as described by [24]. This method considers a wide range of covariates for each streamflow quantile, building candidate regression based on an exhaustive "best-subsets" procedure.…”

Section: Streamflow Estimation Methodsmentioning

confidence: 99%

Evaluation of Uncertainty Intervals for Daily, Statistically Derived Streamflow Estimates at Ungaged Basins across the Continental U.S.

Levin

Farmer

2020

Water

Self Cite

View full text Add to dashboard Cite

Streamflow estimation methods that transfer information from an index gage to an ungauged site are commonly used; however, uncertainty in daily streamflow estimates are often not adequately quantified. In this study, daily streamflow was simulated at 1331 validation streamgauges across the continental United States using four transfer-based streamflow estimation methods. Empirical 95 percent uncertainty intervals were computed for estimated daily streamflows. Uncertainty intervals were evaluated for reliability, sharpness, and overall ability to accurately quantify the uncertainty inherent in the estimated daily streamflow. Uncertainty intervals performed reliably in the Eastern U.S. and Pacific Northwest regions of the country, containing a median of 96 and 99 percent of the observed values respectively. Uncertainty intervals were less reliable in the Great Plains and arid Southwest regions, where uncertainty intervals contained a median of 83 and 94 percent of the observed streamflows respectively. Uncertainty interval performance was correlated with gage density and hydrologic similarity near the validation site, as well as the aridity and baseflow indices at the site.

show abstract

Section: Streamflow Estimation Methodsmentioning

confidence: 99%

Evaluation of Uncertainty Intervals for Daily, Statistically Derived Streamflow Estimates at Ungaged Basins across the Continental U.S.

Levin

Farmer

2020

Water

Self Cite

View full text Add to dashboard Cite

show abstract

“…NNDAR, MCDAR, NNQPPQ, and MCQPPQ estimates were computed following methods described by [1] with updates to the flow-duration curve modeling which is described in [37]. OKDAR estimates were computed following methods in [17].…”

Section: Data From Statistical Transfer Modelsmentioning

confidence: 99%

Comparing Trends in Modeled and Observed Streamflows at Minimally Altered Basins in the United States

Hodgkins

Dudley

Russell

et al. 2020

Water

Self Cite

View full text Add to dashboard Cite

We compared modeled and observed streamflow trends from 1984 to 2016 using five statistical transfer models and one deterministic, distributed-parameter, process-based model, for 26 flow metrics at 502 basins in the United States that are minimally influenced by development. We also looked at a measure of overall model fit and average bias. A higher percentage of basins, for all models, had relatively low trend differences between modeled and observed mean/median flows than for very high or low flows such as the annual 1-day high and 7-day low flows. Mean-flow metrics also had the largest percentage of basins with relatively good overall model fit and low bias. The five statistical transfer models performed better at more basins than the process-based model. The overall model fit for all models, for mean and/or high flows, was correlated with one or more measures of basin precipitation or aridity. Our study and previous studies generally observed good model performance for high flows up to 90th or 95th percentile flows. However, we found model performance was substantially worse for more extreme flows, including 99th percentile and annual 1-day high flows, indicating the importance of including more extreme high flows in analyses of model performance.

show abstract

“…Yokoo and Both statistical and process-based approaches have been used for estimating FDCs and for understanding their climatic and landscape controls. Statistical methods rely on fitting a statistical distribution function to many observed FDCs in gauged sites in a region and then building regional regression relationships between the fitted parameters of the distribution and catchment properties (Castellarin et al, 2004(Castellarin et al, , 2007Li et al, 2010;Over et al, 2018;Singh et al, 2001;Vogel & Fennessey, 1994). For example, Castellarin et al (2004Castellarin et al ( , 2007 related the parameters of an index flow stochastic model fitted to the FDCs to the geomorphoclimatic characteristics of catchments (Castellarin et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

A New Framework for Exploring Process Controls of Flow Duration Curves

Ghotbi

Wang

Singh

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

The flow duration curve (FDC) is effectively the cumulative distribution function of streamflow. For a long time, hydrologists have sought deeper understanding of the process controls on the shape of FDC, which has been a challenge due to contrasting processes controlling the fast flow and slow flow components of streamflow and their interactions. In this paper, we outline a new framework for exploring the process controls of the FDC, which involves studying fast and slow flow components of FDC separately and combining them statistically, explicitly accounting for their dependence. We illustrate the potential of the framework by constructing empirical fast flow duration (FFDC) and slow flow duration (SFDC) curves from the flow components obtained by traditional baseflow separation. Streamflow time series data from 245 MOPEX catchments across the continental United States are used. The dependence of FFDC and SFDC components is captured by the Gumbel copula, the strength of which varies regionally. In catchments where FFDC and SFDC are independent, FDC can be approximated by a simple convolution of FFDC and SFDC. The proposed conceptual framework opens the way for future modeling studies to explore process controls of fast and slow components of streamflow separately, their dependence, and their relative contributions to the shape of the FDC.

show abstract

Refinement of a regression-based method for prediction of flow-duration curves of daily streamflow in the conterminous United States

Abstract: Cover. Map of conterminous United States showing study regions and streamgages, and sample graphs showing unit flow-duration curves, associated standard deviation functions and root-meansquare error values, and flow-regime breakpoints computed from streamgages in the region.

Cited by 8 publications

References 7 publications

Evaluation of Uncertainty Intervals for Daily, Statistically Derived Streamflow Estimates at Ungaged Basins across the Continental U.S.

Evaluation of Uncertainty Intervals for Daily, Statistically Derived Streamflow Estimates at Ungaged Basins across the Continental U.S.

Comparing Trends in Modeled and Observed Streamflows at Minimally Altered Basins in the United States

A New Framework for Exploring Process Controls of Flow Duration Curves

Contact Info

Product

Resources

About