Preprints
https://doi.org/10.5194/hess-2021-154
https://doi.org/10.5194/hess-2021-154

  14 Apr 2021

14 Apr 2021

Review status: a revised version of this preprint is currently under review for the journal HESS.

Uncertainty Estimation with Deep Learning for Rainfall–Runoff Modelling

Daniel Klotz1, Frederik Kratzert1, Martin Gauch1, Alden Keefe Sampson2, Johannes Brandstetter1, Günter Klambauer1, Sepp Hochreiter1, and Grey Nearing3 Daniel Klotz et al.
  • 1Institute for Machine Learning, Johannes Kepler University Linz, Linz, Austria
  • 2Upstream Tech, Natel Energy Inc.; Alameda, CA, USA
  • 3Google Research, Mountain View, CA, USA

Abstract. Deep Learning is becoming an increasingly important way to produce accurate hydrological predictions across a wide range of spatial and temporal scales. Uncertainty estimations are critical for actionable hydrological forecasting, and while standardized community benchmarks are becoming an increasingly important part of hydrological model development and research, similar tools for benchmarking uncertainty estimation are lacking. This contributions demonstrates that accurate uncertainty predictions can be obtained with Deep Learning. We establish an uncertainty estimation benchmarking procedure and present four Deep Learning baselines. Three baselines are based on Mixture Density Networks and one is based on Monte Carlo dropout. The results indicate that these approaches constitute strong baselines, especially the former ones. Additionaly, we provide a post-hoc model analysis to put forward some qualitative understanding of the resulting models. This analysis extends the notion of performance and show that learn nuanced behaviors in different situations.

Daniel Klotz et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on hess-2021-154', John Quilty, 15 May 2021
    • AC1: 'Reply on RC1', Daniel Klotz, 04 Jul 2021
  • RC2: 'Comment on hess-2021-154', Anonymous Referee #2, 18 May 2021
    • AC2: 'Reply on RC2', Daniel Klotz, 04 Jul 2021
  • RC3: 'Comment on hess-2021-154', Anna E. Sikorska-Senoner, 21 May 2021

Daniel Klotz et al.

Daniel Klotz et al.

Viewed

Total article views: 25,588 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
12,989 12,572 27 25,588 16 10
  • HTML: 12,989
  • PDF: 12,572
  • XML: 27
  • Total: 25,588
  • BibTeX: 16
  • EndNote: 10
Views and downloads (calculated since 14 Apr 2021)
Cumulative views and downloads (calculated since 14 Apr 2021)

Viewed (geographical distribution)

Total article views: 22,986 (including HTML, PDF, and XML) Thereof 22,986 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 01 Dec 2021
Download
Short summary
This contribution evaluates distributional runoff-predictions from Deep Learning based approaches. We propose a benchmarking setup and establish four strong baselines. The results show that accurate, precise, and reliable uncertainty estimation can be achieved with Deep Learning.