Improving predictions of hydrological low-flow indices in ungaged basins using machine learning
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.
Citation Information
Publication Year | 2018 |
---|---|
Title | Improving predictions of hydrological low-flow indices in ungaged basins using machine learning |
DOI | 10.1016/j.envsoft.2017.12.021 |
Authors | Scott C. Worland, William H. Farmer, Julie E. Kiang |
Publication Type | Article |
Publication Subtype | Journal Article |
Series Title | Environmental Modelling and Software |
Index ID | 70198127 |
Record Source | USGS Publications Warehouse |
USGS Organization | Lower Mississippi-Gulf Water Science Center |
Related Content
7Q10 records and basin characteristics for 224 basins in South Carolina, Georgia, and Alabama (2015)
William H Farmer, Ph.D.
Acting Director, Northeast Climate Adaptation Science Center
Research Physical Scientist
Related Content
- Data
7Q10 records and basin characteristics for 224 basins in South Carolina, Georgia, and Alabama (2015)
This data release provides the data and R scripts used for the 2017 submitted publication titled "Improving predictions of hydrological low-flow indices in ungaged basins using machine learning". There are two .csv files and 14 R-scripts included below. The lowflow_sc_ga_al_gagesII_2015.csv datafile contains the annual minimum seven-day mean streamflow with an annual exceedance probability of 90% - Connect
William H Farmer, Ph.D.
Acting Director, Northeast Climate Adaptation Science CenterResearch Physical ScientistEmailPhone