QuakeFormer: A Uniform Approach to Earthquake Ground Motion Prediction Using Masked Transformers

TL;DR

QuakeFormer is a unified deep learning Transformer model that simultaneously predicts earthquake ground motion, provides early warnings, and interpolates data, outperforming existing models by effectively capturing spatial dependencies from seismic data.

Contribution

The paper introduces QuakeFormer, a novel unified Transformer-based architecture that models multiple earthquake ground motion prediction tasks simultaneously, incorporating spatial dependencies directly from seismic recordings.

Findings

Outperforms state-of-the-art models in all three tasks

Pretraining improves early warning performance

Effectively models spatial dependencies from seismic data

Abstract

Ground motion prediction (GMP) models are critical for hazard reduction before, during and after destructive earthquakes. In these three stages, intensity forecasting, early warning and interpolation models are corresponding employed to assess the risk. Considering the high cost in numerical methods and the oversimplification in statistical methods, deep-learning-based approaches aim to provide accurate and near-real-time ground motion prediction. Current approaches are limited by specialized architectures, overlooking the interconnection among these three tasks. What's more, the inadequate modeling of absolute and relative spatial dependencies mischaracterizes epistemic uncertainty into aleatory variability. Here we introduce QuakeFormer, a unified deep learning architecture that combines these three tasks in one framework. We design a multi-station-based Transformer architecture and a flexible masking strategy for training QuakeFormer. This data-driven approach enables the model to learn spatial ground motion dependencies directly from real seismic recordings, incorporating location embeddings that include both absolute and relative spatial coordinates. The results indicate that our model outperforms state-of-the-art ground motion prediction models across all three tasks in our research areas. We also find that pretraining a uniform forecasting and interpolation model enhances the performance on early warning task. QuakeFormer offers a flexible approach to directly learning and modeling ground motion, providing valuable insights and applications for both earthquake science and engineering.