We've been overestimating the power of the world's biggest earthquakes by using the wrong math.
April 17, 2026
Original Paper
Width-Saturated Fault Scaling and AI-Driven Seismic Hazard: A Global First-Principles Machine Learning Framework
EarthArXiv · 10.31223/X5XF4S
The Takeaway
For years, seismologists assumed that the bigger a fault line gets, the more energy it can release in a predictable, 'self-similar' way. This paper proves that's wrong: once a fault gets wide enough, it 'saturates,' and the energy growth slows down significantly. By analyzing faults globally with AI, they found the scaling factor is 1.08 instead of the 1.5 everyone was using. This means the 'worst-case scenario' for some of our biggest fault lines might not be as cataclysmic as we feared. It changes how we design skyscrapers and insurance models in earthquake zones.
From the abstract
Traditional probabilistic seismic hazard analysis (PSHA) relies on empirical magnitude-area scaling relationships that systematically overestimate energy release in large, geometrically saturated fault systems. This study presents a dynamic, data-driven framework integrating first-principles geophysics with Gaussian Process Regression (GPR) to produce a physics-informed global seismic hazard index across 25 major fault systems. The central empirical result is a seismic moment-fault area scaling