Table of Contents

I am a data-driven experimental-computational seismologist. My research revolves around observational natural- and applied-geophysics to learn about the time-varying properties of the earth, at a variety of scales, and for a variety of earth and environmental issues.

… inevitably 'under construction' down here …

Seismic Noise Gradiometry

The emergence of large and dense seismic arrays that are deployed long-term, or even permanently, offer tremendous opportunities for imaging that conventional approaches fail to exploit. Dense measurements of the seismic wavefield, as recorded by today's large seismic array's, provide a direct measurement of the wavefield's derivatives in time and space. For surface waves, we can obtain the phase-velocity directly from the ratio between second-order temporal and spatial derivatives of the wavefield. This relationship holds for any kind of surface-wave wavefield, a plane wave, seismic noise, and even standing waves. Unlike seismic interferometry, this technique does not rely on cross-correlations to obtain the Green's function between two seismic receivers.

Publications and Proceeding

Love and Scholte Wave Ambient Seismic Imaging at Valhall


Select Proceedings

Azimuthal Anisotropic Ambient Seismic Imaging at Ekofisk



Reservoir-Scale Seismic Interferometry

Online Lecture


Select Proceedings

Seismoelectric Modeling and Interferometric Green's Function Recovery

Interferometric Green's function representations can be used to retrieve a Green's function between two receiver stations, effectively turning one receiver into a source. Through reciprocity theorems of the convolution and correlation types, I derived interferometric Green's function representations for coupled electromagnetic and seismic wave propagation in 1D. These representations express a symmetrized Green's function in terms of correlations of sources distributed throughout the domain of reciprocity and on its boundary.

The main challenge for practical implementation is the necessity of sources throughout a domain. Numerical examples show how this constraint can be relaxed for different configurations. In a configuration of two layers bounded by a vacuum, seismic noise sources behind the interface can be used to recover seismoelectric reflection responses that suffer from small amplitude losses, but are not corrupted by spurious events.


Conference Proceeding

Green's Function Retrieval by Coda Wave Cross-Correlations

Seismic coda are the chaotic arrivals of energy that follow after the first break of seismic waves. They result from complex subsurface structure that cause a long string of reverberation after the balistic wave passes. The coda waves cary information about the earth, but it is more difficult to image than the information in balistic waves. The angular distribution of energy flow in the coda is much more varied than in baslistic waves. This can be utilized with cross-correlations.

A unique dataset of the coda of an event the size of a small earthquake recorded with high spatial resolution is the Non Proliferation Experiment (NPE) recording made by Sam Allen and the Subsurface Exploration Company of Pasadena, CA.


Further Reading
What Jon Claerbout has to say about this dataset.
About the Comprehensive Nuclear Test Ban Treaty (CTBT).

Near Surface Geophysics

I have an interest in near surface seismic imaging. It got me involved in an attempt to image an actively burning coal fire just 15 meters deep.
Left figure: Airplane to bring the team and equipment to the field site in Utah. Right figure: Me creating a shot record. sageep_figure.jpg Figure: Two shot records at the Southern Ute Nation Coal Fire Site; a direct P-wave event is annotated '1', two refracted events (possibly of the burning coal seam) are annotated '2' and '3', possible reflected events are annotated 'R1' and 'R2', and dispersive ground roll is annotated 'G'.

Proceedings and Publications