Predicting streamflow drought in the conterminous United States using machine learning and a donor-gage approach, 1982–2020

Heldmyer, Aaron; Sando, Roy; Simeone, Caelan; Wieczorek, Michael; Hamshaw, Scott; Goodling, Phillip; McShane, Ryan; Diaz, Jeremy; Watkins, David; Pulver, Bryce; Shastry, Apoorva; Hafen, Konrad; Hammond, John

doi:10.5194/hess-30-5925-2026

Articles | Volume 30, issue 18

https://doi.org/10.5194/hess-30-5925-2026

Copyright waived. This work has been dedicated to the public domain (Creative Commons Public Domain Dedication).

https://doi.org/10.5194/hess-30-5925-2026

Copyright waived. This work has been dedicated to the public domain (Creative Commons Public Domain Dedication).

Articles | Volume 30, issue 18

Research article

|

23 Sep 2026

Research article |

| 23 Sep 2026

Predicting streamflow drought in the conterminous United States using machine learning and a donor-gage approach, 1982–2020

Aaron Heldmyer, Roy Sando, Caelan Simeone, Michael Wieczorek, Scott Hamshaw, Phillip Goodling, Ryan McShane, Jeremy Diaz, David Watkins, Bryce Pulver, Apoorva Shastry, Konrad Hafen, and John Hammond

Supplement

https://doi.org/10.5194/hess-30-5925-2026-supplement

Data sets

Random forest and donor-based simulated streamflow drought for 3,198 gages across the CONUS, 1982-2020 Aaron Heldmyer and Roy Sando https://doi.org/10.5066/P1FSRPMU

Streamflow Drought Metrics for Select United States Geological Survey Streamgages for Three Different Time Periods from 1921-2020 C. E. Simeone https://doi.org/10.5066/P92FAASD

Short summary

We used machine learning to explore what causes streamflow droughts across the U.S. We found that different regions are influenced by different factors like temperature, snow, and rainfall. Our new method can also predict droughts in areas without streamflow data, helping improve water resource planning.