Preprints
https://doi.org/10.5194/egusphere-2025-1224
https://doi.org/10.5194/egusphere-2025-1224
07 Apr 2025
 | 07 Apr 2025

How to deal w___ missing input data

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

Abstract. Deep learning hydrologic models have made their way from research to applications. More and more national hydrometeorological agencies, hydro power operators, and engineering consulting companies are building Long Short-Term Memory (LSTM) models for operational use cases. All of these efforts come across similar sets of challenges—challenges that are different from those in controlled scientific studies. In this paper, we tackle one of these issues: how to deal with missing input data? Operational systems depend on the real-time availability of various data products—most notably, meteorological forcings. The more external dependencies a model has, however, the more likely it is to experience an outage in one of them. We introduce and compare three different solutions that can generate predictions even when some of the meteorological input data do not arrive in time, or not arrive at all.

Competing interests: Daniel Klotz is editor at HESS.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Martin Gauch, Frederik Kratzert, Daniel Klotz, Grey Nearing, Deborah Cohen, and Oren Gilon

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CC1: 'Comment on egusphere-2025-1224', Xin Yu, 07 Apr 2025
    • AC1: 'Reply on CC1', Martin Gauch, 24 Apr 2025
  • RC1: 'Comment on egusphere-2025-1224', Juliane Mai, 03 Jul 2025
    • AC2: 'Reply on RC1', Martin Gauch, 05 Aug 2025
  • RC2: 'Comment on egusphere-2025-1224', Anonymous Referee #2, 10 Jul 2025
    • AC3: 'Reply on RC2', Martin Gauch, 05 Aug 2025
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

Model code and software

GitHub repository with analysis code Martin Gauch https://github.com/gauchm/missing-inputs

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

Viewed

Total article views: 693 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
525 147 21 693 18 32
  • HTML: 525
  • PDF: 147
  • XML: 21
  • Total: 693
  • BibTeX: 18
  • EndNote: 32
Views and downloads (calculated since 07 Apr 2025)
Cumulative views and downloads (calculated since 07 Apr 2025)

Viewed (geographical distribution)

Total article views: 677 (including HTML, PDF, and XML) Thereof 677 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 30 Aug 2025
Download
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.
Share