POSEIDON: Physics-Optimized Seismic Energy Inference and Detection Operating Network

TL;DR

POSEIDON proposes a physics-informed energy-based framework for unified seismic hazard prediction, addressing aftershock, tsunami, and foreshock tasks within a shared representation. The Poseidon dataset provides a globally comprehensive catalog of $2.8\times 10^{6}$ events over $30$ years with energy features derived from the Gutenberg-Richter relation, enabling learnable incorporation of seismological laws. Key findings include learned parameters that align with established theory, such as the Gutenberg-Richter $b$-value and Omori-Utsu $p$ and $c$, alongside strong predictive performance (high F1 and AUC) across tasks and interpretable physics signals. The work demonstrates that physics-informed constraints can enhance predictive accuracy while preserving interpretability, and the Poseidon dataset offers a valuable benchmark for physics-informed seismic research with practical implications for hazard assessment and early warning.

Abstract

Earthquake prediction and seismic hazard assessment remain fundamental challenges in geophysics, with existing machine learning approaches often operating as black boxes that ignore established physical laws. We introduce POSEIDON (Physics-Optimized Seismic Energy Inference and Detection Operating Network), a physics-informed energy-based model for unified multi-task seismic event prediction, alongside the Poseidon dataset -- the largest open-source global earthquake catalog comprising 2.8 million events spanning 30 years. POSEIDON embeds fundamental seismological principles, including the Gutenberg-Richter magnitude-frequency relationship and Omori-Utsu aftershock decay law, as learnable constraints within an energy-based modeling framework. The architecture simultaneously addresses three interconnected prediction tasks: aftershock sequence identification, tsunami generation potential, and foreshock detection. Extensive experiments demonstrate that POSEIDON achieves state-of-the-art performance across all tasks, outperforming gradient boosting, random forest, and CNN baselines with the highest average F1 score among all compared methods. Crucially, the learned physics parameters converge to scientifically interpretable values -- Gutenberg-Richter b-value of 0.752 and Omori-Utsu parameters p=0.835, c=0.1948 days -- falling within established seismological ranges while enhancing rather than compromising predictive accuracy. The Poseidon dataset is publicly available at https://huggingface.co/datasets/BorisKriuk/Poseidon, providing pre-computed energy features, spatial grid indices, and standardized quality metrics to advance physics-informed seismic research.

Figures (14)

• Figure 1: Overview of Poseidon Dataset.
• Figure 2: PI-EBM Architecture Overview.
• Figure 3: Performance comparison across baseline methods showing F1 scores.
• Figure 4: ROC curves for all prediction tasks.
• Figure 5: Gutenberg-Richter law validation showing frequency-magnitude distribution, b-value convergence, and regional variation.