WaveCastNet: An AI-enabled Wavefield Forecasting Framework for Earthquake Early Warning

2024-05-30

arXiv.org

AI Summary

Early earthquake warning systems face challenges in rapidly predicting high-dimensional wavefields—especially for rare, large-magnitude events—due to reliance on error-prone source parameter estimation and empirical attenuation models that fail to capture strong heterogeneity in wave propagation. Method: This paper proposes an end-to-end wavefield prediction framework centered on a novel Convolutional Long Expressive Memory (ConvLEM) module. ConvLEM models long-range spatiotemporal dependencies and multi-scale wave propagation via reusable spatiotemporal weight mechanisms, eliminating the need for epicenter localization and empirical attenuation models. Built upon a lightweight sequence-to-sequence architecture, the model enables millisecond-scale inference. Results: Evaluated on synthetic data for the San Francisco Bay Area, it achieves significantly improved accuracy in predicting destructive ground-motion intensity and arrival times. Compared to Transformer baselines, it reduces parameters by 62%, accelerates inference by 3.1×, and demonstrates superior generalization—particularly for out-of-distribution, high-magnitude scenarios.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Large Vision Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Large earthquakes can be destructive and quickly wreak havoc on a landscape. To mitigate immediate threats, early warning systems have been developed to alert residents, emergency responders, and critical infrastructure operators seconds to a minute before seismic waves arrive. These warnings provide time to take precautions and prevent damage. The success of these systems relies on fast, accurate predictions of ground motion intensities, which is challenging due to the complex physics of earthquakes, wave propagation, and their intricate spatial and temporal interactions. To improve early warning, we propose a novel AI-enabled framework, WaveCastNet, for forecasting ground motions from large earthquakes. WaveCastNet integrates a novel convolutional Long Expressive Memory (ConvLEM) model into a sequence to sequence (seq2seq) forecasting framework to model long-term dependencies and multi-scale patterns in both space and time. WaveCastNet, which shares weights across spatial and temporal dimensions, requires fewer parameters compared to more resource-intensive models like transformers and thus, in turn, reduces inference times. Importantly, WaveCastNet also generalizes better than transformer-based models to different seismic scenarios, including to more rare and critical situations with higher magnitude earthquakes. Our results using simulated data from the San Francisco Bay Area demonstrate the capability to rapidly predict the intensity and timing of destructive ground motions. Importantly, our proposed approach does not require estimating earthquake magnitudes and epicenters, which are prone to errors using conventional approaches; nor does it require empirical ground motion models, which fail to capture strongly heterogeneous wave propagation effects.

Problem

• Forecasting high-dimensional seismic wavefields using deep learning
• Reducing computational requirements for real-time earthquake early warning
• Eliminating error-prone magnitude estimation in ground motion prediction

Innovation

• Deep sequence-to-sequence learning for wavefield forecasting
• Convolutional memory architecture models spatiotemporal dependencies
• Weight sharing reduces parameters for faster inference