Âé¶¹ÒùÔº


Why seismic waves are slower shortly after an earthquake

Why seismic waves are slower for shortly after an earthquake
Rock sample of Bentheim sandstone, inside a protective neoprene jacket and with all the sensors glued to its surface, before going into GFZ's experimental press. Credit: Manuel Asnar/GFZ

Solid as they are, rocks are not static materials with constant properties. Even small loads are enough to alter their mechanical properties; their reaction to being deformed is a loss of stiffness. Rocks which have been damaged in such a way are then less able to withstand loads, such as gravity or tectonic stresses. This phenomenon is therefore of relevance for understanding the occurrence of material failure, as in landslides or earthquakes.

Such changes in are commonly observed in heterogeneous and granular materials such as rocks, concrete or sediments. As a result, they play a role in geotechnics as well as in the stability of man-made structures.

These effects have been observed in laboratory experiments for years using acoustic methods. The development of seismic interferometry made such observations possible in the field by exploiting the so-called "seismic noise."

A key observation using these methods is the sudden decrease in the velocity of seismic waves in the subsurface in the wake of an earthquake (damage). This decrease is followed by a slow re-increase, which can extend over several years (recovery).

Despite these studies and many years of research, the physical origins of these processes have still not been clearly determined. It is, however, commonly agreed that the contrast between the very stiff grains and vastly softer grain contact planes, as well as the stress concentrations at these contact points, is responsible for these effects.

Manuel Asnar and a team of collaborators from the GFZ, the University of Edinburgh in the UK, and the Université de Lorraine in France, have managed to make a breakthrough in laboratory experiments carried out at the GFZ's High-Pressure Labs. Their allowed them to measure wave velocities in a 10-centimeter sandstone cylinder along various directions of propagation to an extremely high degree of precision. The findings are in the journal Nature Communications.

The sample was subjected to varying levels of stress along the cylinder's axis. Doing so showed that, as expected, the static effects of the loading strongly impacted waves along the main axis, while leaving the waves along the diameter relatively unaffected. The time-dependent effects, however, that is, the sudden damage and long healing, were consistently observed along every direction of propagation.

These results show that the time-dependent effects are not caused by grain contacts that are being more or less compressed against each other. Rather, these effects can be traced back to contact planes sliding against each other, irrespective of whether the load is being applied or released.

The effects of friction along contacts within the material have long been suggested as being responsible for these time-dependent changes in wave velocities; but this study of the anisotropy of velocity changes—meaning, their directional dependence—provides meaningful evidence in favor of this interpretation. Based on those findings, models can be developed to better describe and predict the time-dependent changes of in rocks and geotechnical materials.

More information: Manuel Asnar et al, Anisotropy reveals contact sliding and aging as a cause of post-seismic velocity changes, Nature Communications (2025).

Journal information: Nature Communications

Citation: Why seismic waves are slower shortly after an earthquake (2025, August 29) retrieved 29 August 2025 from /news/2025-08-seismic-slower-shortly-earthquake.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

A 2022 Pacific volcano eruption made a deep dive into Alaska

0 shares

Feedback to editors