thumbnail

Evaluation of Machine Learning Approaches for Predicting Streamflow Metrics Across the Conterminous United States

Scientific Investigations Report 2022-5058
By:  and 
https://doi.org/10.3133/sir20225058

Metrics

4
Crossref references
Web analytics dashboard Metrics definitions

Links

Abstract

Few regional or national scale studies have evaluated machine learning approaches for predicting streamflow metrics at ungaged locations. Most such studies are limited by the number of dimensions of the streamflow regime investigated. This study, in contrast, provides a comprehensive evaluation of the streamflow regime based on three widely available machine learning approaches (support vector regression, random forest, and cubist regression) and on multiple linear regression to predict 106 natural streamflow metrics at ungaged locations. This evaluation is done for 545 streamgages across the northwest United States for recurrence-interval flood metrics and for 1,851 sites in the conterminous United States for non-flood metrics. The results indicate that for flood metrics, predictions by cubist regression and support vector regressions have substantially less error than the other approaches. For all the remaining streamflow metrics, random forest models outperform the other methods.

Suggested Citation

Eng, K., and Wolock, D.M., 2022, Evaluation of machine learning approaches for predicting streamflow metrics across the conterminous United States: U.S. Geological Survey Scientific Investigations Report 2022–5058, 27 p., https://doi.org/10.3133/sir20225058.

ISSN: 2328-0328 (online)

Study Area

Table of Contents

  • Abstract
  • Introduction
  • Study Area and Basin Attributes
  • Methods
  • Performance Evaluation
  • Discussion on Performance of Approaches
  • Summary
  • Acknowledgments
  • References Cited
  • Appendix 1. 176 Basin Attributes and Corresponding Descriptions
Publication type Report
Publication Subtype USGS Numbered Series
Title Evaluation of machine learning approaches for predicting streamflow metrics across the conterminous United States
Series title Scientific Investigations Report
Series number 2022-5058
DOI 10.3133/sir20225058
Publication Date August 31, 2022
Year Published 2022
Language English
Publisher U.S. Geological Survey
Publisher location Reston, VA
Contributing office(s) WMA - Integrated Modeling and Prediction Division
Description Report: iv, 27 p.; Data Release
Country United States
Online Only (Y/N) Y
Additional Online Files (Y/N) N
Additional publication details