Improving predictions of hydrological low-flow indices in ungaged basins using machine learning. (March 2018)

• Record Type: Journal Article
• Authors: Worland, Scott C.; Farmer, William H.; Kiang, Julie E.
• Is Part Of: Environmental modelling & software. Volume 101(2018)
• Page Start: 169
• Page End: 182
• Publication Date: 2018-03
• Subjects: Low streamflow -- Ungaged basins -- 7Q10 -- Machine learning -- Censored regression -- Variable importance; 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.2017.12.021
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: We compare the ability of eight machine-learning models (elastic net, gradient boosting, kernel-k-nearest neighbors, two variants of support vector machines, M5-cubist, random forest, and a meta-learning ensemble M5-cubist model) and four baseline models (ordinary kriging, a unit area discharge model, and two variants of censored regression) to generate estimates of the annual minimum 7-day mean streamflow with an annual exceedance probability of 90% (7Q10) at 224 unregulated sites in South Carolina, Georgia, and Alabama, USA. The machine-learning models produced substantially lower cross validation errors compared to the baseline models. The meta-learning M5-cubist model had the lowest root-mean-squared-error of 26.72 cubic feet per second. Partial dependence plots show that 7Q10s are likely moderated by late summer and early fall precipitation and the infiltration capacity of basin soils. Highlights: Machine learning outperforms baseline models for predicting 7Q10s in ungaged basins. M5-cubist models result in the best predictions of 7Q10s in ungaged basins. Stacked-ensemble methods can improve predictions of 7Q10s in ungaged basins. Partial dependence functions can relate 7Q10 values to basin characteristics.