A bootstrap method for estimating uncertainty of water quality trends. (November 2015)

• Record Type: Journal Article
• Authors: Hirsch, Robert M.; Archfield, Stacey A.; De Cicco, Laura A.
• Is Part Of: Environmental modelling & software. Volume 73(2015:Nov.)
• Page Start: 148
• Page End: 166
• Publication Date: 2015-11
• Subjects: Water quality -- Bootstrap -- Trend -- Uncertainty analysis; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2015.07.017
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Estimation of the direction and magnitude of trends in surface water quality remains a problem of great scientific and practical interest. The Weighted Regressions on Time, Discharge, and Season (WRTDS) method was recently introduced as an exploratory data analysis tool to provide flexible and robust estimates of water quality trends. This paper enhances the WRTDS method through the introduction of the WRTDS Bootstrap Test (WBT), an extension of WRTDS that quantifies the uncertainty in WRTDS-estimates of water quality trends and offers various ways to visualize and communicate these uncertainties. Monte Carlo experiments are applied to estimate the Type I error probabilities for this method. WBT is compared to other water-quality trend-testing methods appropriate for data sets of one to three decades in length with sampling frequencies of 6–24 observations per year. The software to conduct the test is in the EGRETci R-package. Graphical abstract: Highlights: Block bootstrap approach for water quality trends is developed. Used in conjunction with a flexible statistical model for river water quality. Trends in concentration and trends in flux can be evaluated. Confidence intervals can be estimated for trend magnitude. Based on WRTDS: Weighted Regressions on Time, Discharge, and Season.