Affiliation:
1. University of Science and Technology of China
Abstract
Abstract
The occurrence of earthquakes at depths greater than 60 km is an enigma in earthquake science, as rocks at greater depths are anticipated to be ductile. Over the past decades, seismologists have reported a variety of rupture characteristics that are distinguishable among shallow (0–60 km), intermediate-depth (60–300 km) and deep-focus (300–700 km) earthquakes. However, their underlying physical mechanisms remain enigmatic. Using machine learning, we show that the previously observed differences in earthquake moment release processes are caused by the Earth’s depth-dependent elastic properties. Despite the presentation of medium rigidity, earthquakes with different generation mechanisms unanimously have similar moment release processes. Our results support a constant strain drop hypothesis, that is, the ratio of coseismic slip to the characteristic rupture length remains largely unchanged for earthquakes at all depths and mechanisms. This finding contests the conventional stress drop hypothesis, leading to a revised definition of earthquake self-similarity. In particular, the moment release invariant property predicts distinct depth- and rigidity-dependencies of critical earthquake parameters, such as rupture duration, geometric scale, and stress drop. These results have broad implications for understanding earthquake physics and assessing seismic hazards.
Publisher
Research Square Platform LLC