How to deal w___ missing input data

Gauch, Martin; Kratzert, Frederik; Klotz, Daniel; Nearing, Grey; Cohen, Deborah; Gilon, Oren

doi:10.5194/hess-29-6221-2025

Articles | Volume 29, issue 21

https://doi.org/10.5194/hess-29-6221-2025

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

https://doi.org/10.5194/hess-29-6221-2025

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

Articles | Volume 29, issue 21

Research article

|

13 Nov 2025

Research article |

| 13 Nov 2025

How to deal w___ missing input data

Martin Gauch, Frederik Kratzert, Daniel Klotz, Grey Nearing, Deborah Cohen, and Oren Gilon

Data sets

Models, configs, and predictions Martin Gauch https://doi.org/10.5281/zenodo.15008460

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

CAMELS Extended NLDAS Forcing Data Frederik Kratzert https://doi.org/10.4211/hs.0a68bfd7ddf642a8be9041d60f40868c

Model code and software

Code Martin Gauch https://doi.org/10.5281/zenodo.17362593

Interactive computing environment

Code Martin Gauch https://doi.org/10.5281/zenodo.17362593

Short summary

Missing input data are one of the most common challenges when building deep learning hydrological models. We present and analyze different methods that can produce predictions when certain inputs are missing during training or inference. Our proposed strategies provide high accuracy while allowing for more flexible data handling and being robust to outages in operational scenarios.