Preprints
https://doi.org/10.5194/egusphere-2022-105
https://doi.org/10.5194/egusphere-2022-105
 
17 May 2022
17 May 2022
Status: this preprint is open for discussion.

Causal deep learning models for studying the Earth system: soil moisture-precipitation coupling in ERA5 data across Europe

Tobias Tesch1,2, Stefan Kollet1,2, and Jochen Garcke3,4 Tobias Tesch et al.
  • 1Institute of Bio- and Geosciences, Agrosphere (IBG-3), Forschungszentrum Jülich, 52425 Jülich, Germany
  • 2Center for High-Performance Scientific Computing in Terrestrial Systems, Geoverbund ABC/J, Jülich, Germany
  • 3Fraunhofer Center for Machine Learning and Fraunhofer SCAI, 53757 Sankt Augustin, Germany
  • 4Institut für Numerische Simulation, Universität Bonn, 53115 Bonn, Germany

Abstract. The Earth system is a complex non-linear dynamical system. Despite decades of research, many processes and relations between Earth system variables are still poorly understood. Current approaches for studying relations in the Earth system may be broadly divided into approaches based on numerical simulations and statistical approaches. However, there are several inherent limitations to current approaches that are, for example, high computational costs, reliance on the correct representation of relations in numerical models, strong assumptions related to linearity or locality, and the fallacy of correlation and causality.

Here, we propose a novel methodology combining deep learning (DL) and principles of causality research in an attempt to overcome these limitations. The methodology combines the recent idea of training and analyzing DL models to gain new scientific insights in the relations between input and target variables with a theorem from causality research. This theorem states that a statistical model may learn the causal impact of an input variable on a target variable if suitable additional input variables are included. As an illustrative example, we apply the methodology to study soil moisture-precipitation coupling in ERA5 climate reanalysis data across Europe. We demonstrate that, harnessing the great power and flexibility of DL models, the proposed methodology may yield new scientific insights into complex, nonlinear and non-local coupling mechanisms in the Earth system.

Tobias Tesch et al.

Status: open (until 29 Jul 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Tobias Tesch et al.

Model code and software

Causal deep learning models for studying the Earth system: soil moisture-precipitation coupling in ERA5 data across Europe - Software Code Tobias Tesch, Stefan Kollet, Jochen Garcke https://doi.org/10.5281/zenodo.6385040

Tobias Tesch et al.

Viewed

Total article views: 256 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
189 60 7 256 10 5
  • HTML: 189
  • PDF: 60
  • XML: 7
  • Total: 256
  • BibTeX: 10
  • EndNote: 5
Views and downloads (calculated since 17 May 2022)
Cumulative views and downloads (calculated since 17 May 2022)

Viewed (geographical distribution)

Total article views: 225 (including HTML, PDF, and XML) Thereof 225 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 02 Jul 2022
Download
Short summary
A recent statistical approach for studying relations in the Earth system is to train deep learning (DL) models to predict Earth system variables given one or several others, and use interpretable DL to analyze the relations learned by the models. Here, we propose to combine the approach with a theorem from causality research to ensure that the deep learning model learns causal rather than spurious relations. As an example, we apply the method to study soil moisture-precipitation coupling.