Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Integrating Artificial Intelligence and Geophysical Insights for Earthquake Forecasting: A Cross-Disciplinary Review

Zhang Ying, Wen Congcong, Sornette Didier, Zhan Chengxiang

TL;DR

The importance of interdisciplinary collaboration is emphasized, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology, so that more accurate, reliable, and societally impactful earthquake forecasting tools are developed.

Abstract

Earthquake forecasting remains a significant scientific challenge, with current methods falling short of achieving the performance necessary for meaningful societal benefits. Traditional models, primarily based on past seismicity and geomechanical data, struggle to capture the complexity of seismic patterns and often overlook valuable non-seismic precursors such as geophysical, geochemical, and atmospheric anomalies. The integration of such diverse data sources into forecasting models, combined with advancements in AI technologies, offers a promising path forward. AI methods, particularly deep learning, excel at processing complex, large-scale datasets, identifying subtle patterns, and handling multidimensional relationships, making them well-suited for overcoming the limitations of conventional approaches. This review highlights the importance of combining AI with geophysical knowledge to create robust, physics-informed forecasting models. It explores current AI methods, input data types, loss functions, and practical considerations for model development, offering guidance to both geophysicists and AI researchers. While many AI-based studies oversimplify earthquake prediction, neglecting critical features such as data imbalance and spatio-temporal clustering, the integration of specialized geophysical insights into AI models can address these shortcomings. We emphasize the importance of interdisciplinary collaboration, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology. By bridging these disciplines, we can develop more accurate, reliable, and societally impactful earthquake forecasting tools.

Integrating Artificial Intelligence and Geophysical Insights for Earthquake Forecasting: A Cross-Disciplinary Review

TL;DR

The importance of interdisciplinary collaboration is emphasized, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology, so that more accurate, reliable, and societally impactful earthquake forecasting tools are developed.

Abstract

Earthquake forecasting remains a significant scientific challenge, with current methods falling short of achieving the performance necessary for meaningful societal benefits. Traditional models, primarily based on past seismicity and geomechanical data, struggle to capture the complexity of seismic patterns and often overlook valuable non-seismic precursors such as geophysical, geochemical, and atmospheric anomalies. The integration of such diverse data sources into forecasting models, combined with advancements in AI technologies, offers a promising path forward. AI methods, particularly deep learning, excel at processing complex, large-scale datasets, identifying subtle patterns, and handling multidimensional relationships, making them well-suited for overcoming the limitations of conventional approaches. This review highlights the importance of combining AI with geophysical knowledge to create robust, physics-informed forecasting models. It explores current AI methods, input data types, loss functions, and practical considerations for model development, offering guidance to both geophysicists and AI researchers. While many AI-based studies oversimplify earthquake prediction, neglecting critical features such as data imbalance and spatio-temporal clustering, the integration of specialized geophysical insights into AI models can address these shortcomings. We emphasize the importance of interdisciplinary collaboration, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology. By bridging these disciplines, we can develop more accurate, reliable, and societally impactful earthquake forecasting tools.

Paper Structure

This paper contains 76 sections, 101 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction and Motivation
  2. Fundamentals of earthquake forecasting
  3. Differences between an earthquake forecast and an earthquake prediction
  4. Testing the performance of earthquake forecasting models
  5. Benchmarks for earthquake forecasting models
  6. Recommended resources for non-experts in earthquake forecasting
  7. A quick review of AI algorithm used for earthquake forecasting
  8. AI algorithms used in earthquake prediction
  9. Traditional Machine Learning Models
  10. Decision Tree
  11. Support Vector Machines (SVM)
  12. K-Nearest Neighbor (KNN)
  13. Naïve Bayes
  14. Polynomial Regression
  15. Mixture of regressions
  16. ...and 61 more sections

Figures (3)

  • Figure 1: (a) Spatial distribution of earthquakes with magnitude larger than 0 that occurred within the time period from 1 January 1981 to 31 December 2020 in the Regional Earthquake Likelihood Models (RELM) polygon. (b) Frequency-magnitude distribution of earthquakes. The black line indicates the magnitude $M_c$ of completeness over the whole time interval estimated using the method by Clauset et al.R43(c) Black, blue and red lines are the logarithms of the cumulative numbers for events with magnitudes larger than 2, 3 and 4 respective. Solid lines are the numbers of events in the full sequence, while the dotted lines are the numbers of independent events. The sum of the independent probabilities of all events gives the number of independent events, where the independent probability refers to the probability that the event is independent (also called a background event), which is provided by the ETAS modelnandan2021seismicity.
  • Figure 2: Cumulative frequency of different types of outputs.
  • Figure 3: Number of times different types of inputs have been used in published papers for different outputs. The histogram on the right shows the total number of papers that have used different types of inputs.