EGU2020-6851
https://doi.org/10.5194/egusphere-egu2020-6851
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Neural Network Applications in Earthquake Prediction (1994-2019): Meta-Analytic & Statistical Insights on their Limitations

Arnaud Mignan1,2 and Marco Broccardo3
Arnaud Mignan and Marco Broccardo
  • 1Institute of Risk Analysis, Prediction and Management, Academy for Advanced Interdisciplinary Studies, Southern University of Science and Technology, Shenzhen, China
  • 2Institute of Geophysics, Swiss Federal Institute of Technology, Zurich, Switzerland (arnaud.mignan@sed.ethz.ch)
  • 3Institute of Structural Engineering, Swiss Federal Institute of Technology, Zurich, Switzerland (broccardo@ibk.baug.ethz.ch)

In the last few years, deep learning has solved seemingly intractable problems, boosting the hope to find approximate solutions to problems that now are considered unsolvable. Earthquake prediction, the Grail of Seismology, is, in this context of continuous exciting discoveries, an obvious choice for deep learning exploration. We reviewed the literature of artificial neural network (ANN) applications for earthquake prediction (77 articles, 1994-2019 period) and found two emerging trends: an increasing interest in this domain over time, and a complexification of ANN models towards deep learning. Despite the relatively positive results claimed in those studies, we verified that far simpler (and traditional) models seem to offer similar predictive powers, if not better ones. Those include an exponential law for magnitude prediction, and a power law (approximated by a logistic regression or one artificial neuron) for aftershock prediction in space. Due to the structured, tabulated nature of earthquake catalogues, and the limited number of features so far considered, simpler and more transparent machine learning models than ANNs seem preferable at the present stage of research. Those baseline models follow first physical principles and are consistent with the known empirical laws of Statistical Seismology (e.g. the Gutenberg-Richter law), which are already known to have minimal abilities to predict large earthquakes.

How to cite: Mignan, A. and Broccardo, M.: Neural Network Applications in Earthquake Prediction (1994-2019): Meta-Analytic & Statistical Insights on their Limitations, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-6851, https://doi.org/10.5194/egusphere-egu2020-6851, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 26 Apr 2020