Improving model calibrations in a changing world: controlling for nonstationarity after mega disturbance reduces hydrological uncertainty

Boardman, Elijah N.; Boisramé, Gabrielle F. S.; Wigmosta, Mark S.; Shriver, Robert K.; Harpold, Adrian A.

doi:10.5194/hess-29-6333-2025

Articles | Volume 29, issue 22

https://doi.org/10.5194/hess-29-6333-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-29-6333-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 29, issue 22

Research article

|

17 Nov 2025

Research article |

| 17 Nov 2025

Improving model calibrations in a changing world: controlling for nonstationarity after mega disturbance reduces hydrological uncertainty

Elijah N. Boardman, Gabrielle F. S. Boisramé, Mark S. Wigmosta, Robert K. Shriver, and Adrian A. Harpold

Supplement

https://doi.org/10.5194/hess-29-6333-2025-supplement

Data sets

Data and Code for Megafire Disturbance Hydrological Model Calibration Study Elijah N. Boardman https://doi.org/10.5281/zenodo.16972670

Data Portal Airborne Snow Observatories, Inc. https://data.airbornesnowobservatories.com/

Database of full natural flow records for the SBF station California DWR https://cdec.water.ca.gov/dynamicapp/staMeta?station_id=SBF

National Land Cover Database (NLCD) 2019 Products Dewitz, J. and U.S. Geological Survey https://doi.org/10.5066/P9JZ7AO3

Rangeland Condition Monitoring Assessment and Projection (RCMAP) Fractional Component Time-Series Across the Western U. S. 1985--2020 Rigge, M. B., Bunde, B., Shi, H., and Postma, K. https://doi.org/10.5066/P95IQ4BT

Soil Survey Staff Soil Survey Geographic (SSURGO) Database https://sdmdataaccess.sc.egov.usda.gov

Model code and software

randomForest: Breiman and Cutlers Random Forests for Classification and Regression Breiman, L., Cutler, A., Liaw, A., and Wiener, M. https://doi.org/10.32614/CRAN.package.randomForest

Short summary

Environmental changes can cause hydrological model biases that vary over time (nonstationarity). We demonstrate a new calibration framework to detect and correct nonstationary streamflow biases after a large wildfire, which reduces predictive uncertainty and constrains parameter equifinality.