A data-centric perspective on the information needed for hydrological uncertainty predictions

Auer, Andreas; Gauch, Martin; Kratzert, Frederik; Nearing, Grey; Hochreiter, Sepp; Klotz, Daniel

doi:https://doi.org/10.5194/hess-28-4099-2024

Articles | Volume 28, issue 17

https://doi.org/10.5194/hess-28-4099-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-28-4099-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 28, issue 17

Research article

|

12 Sep 2024

Research article |

| 12 Sep 2024

A data-centric perspective on the information needed for hydrological uncertainty predictions

Andreas Auer, Martin Gauch, Frederik Kratzert, Grey Nearing, Sepp Hochreiter, and Daniel Klotz

Data sets

Models and Model States - A data-centric perspective on the information needed for hydrological uncertainty predictions A. Auer https://doi.org/10.5281/zenodo.10653863

CMAL - Non PUB - A data-centric perspective on the information needed for hydrological uncertainty predictions A. Auer https://doi.org/10.5281/zenodo.10654345

CMAL - PUB - A data-centric perspective on the information needed for hydrological uncertainty predictions A. Auer https://doi.org/10.5281/zenodo.10654399

Catchment attributes for large-sample studies data repository: Boulder, CO N. Addor et al. https://gdex.ucar.edu/dataset/camels/file.html

CAMELS Extended Maurer Forcing Data F. Kratzert https://doi.org/10.4211/hs.17c896843cf940339c3c3496d0c1c077

Model code and software

Code - A data-centric perspective on the information needed for hydrological uncertainty predictions A. Auer https://doi.org/10.5281/zenodo.10674231

Short summary

This work examines the impact of temporal and spatial information on the uncertainty estimation of streamflow forecasts. The study emphasizes the importance of data updates and global information for precise uncertainty estimates. We use conformal prediction to show that recent data enhance the estimates, even if only available infrequently. Local data yield reasonable average estimations but fall short for peak-flow events. The use of global data significantly improves these predictions.