This paper demonstrated that the background seismic noise can be used to obtain information about the Earth’s interior. This allows us to increase significantly the efficiency of passive seismic imaging. Rather than waiting for earthquakes, the new method recovers the useful information from ambient seismic noise that is constantly produced by fluctuations in the Earth’s atmosphere and oceans.

“This paper demonstrated that the background seismic noise can be used to obtain information about the Earth’s interior.”

This promises significant improvements in the resolution and accuracy of crustal and upper mantle images down to 100 kilometers or more within the Earth.

This paper presents a novel approach to the analysis of seismic data. Extracting deterministic seismic waves from correlations of seismic noise opens new opportunities to improve the imaging of the Earth and the seismic monitoring of active faults and volcanoes.

Most of continuous seismic recordings are composed of signals excited by non-tectonic sources, related to the coupling of the solid Earth with the atmosphere and the oceans. Because of the traditional orientation of seismology toward studies of earthquakes and tectonics, those non-tectonically excited seismic waves were often categorized as noise and were largely ignored by seismologists.

The present study demonstrates that, by computing cross-correlations of noise records at two separate stations A and B, we can reconstruct a signal that is similar to a seismic record that would be registered at A if an earthquake occurred at B. Application of this technique to the records of the Californian seismic network resulted in a few thousand new seismic measurements that allowed us to construct images of the California crust with a resolution level that is much superior to what could be obtained with traditional earthquake-based methods.

This work is an example of a very fruitful exchange of ideas between different scientific disciplines. My main field is seismology and I have been lucky to become involved in an interdisciplinary research on the physics of random wavefields. After pioneering studies published in the ’60s and ’70s, random seismic waves within the Earth rather disappeared from the "mainstream" seismology, being considered as too complex and not very useful to obtain information about the Earth’s interior.

Recently, some seismologists have challenged this point of view. Among them were Michel Campillo and his colleagues from the University of Grenoble in France, who developed collaborations with physicists and acousticians to bring new ideas and methods of analysis of random wavefields into seismology.

Three years ago, I was working as a research associate with professor Michael Ritzwoller at the University of Colorado, developing new approaches for seismic surface-wave tomography.

At this time, Michel Campillo and his colleagues demonstrated that correlations of randomly scattered seismic waves within the Earth lead to deterministic measurements. A logical continuation in this direction was an application of this technique to the ambient seismic noise. Combination of the new noise-based measurement technique with the surface-wave tomography proved to be very fruitful.

Seismological methods are largely used for evaluation of geologic hazards and, consequently, have important social implications. Improvements to imaging and monitoring of the superficial Earth structure with the new method proposed in this paper can provide valuable information for risk assessment.