The main focus of my research is studying earthquakes using 3D numerical modeling.  I am interested in how to better constrain source parameters, such as the distribution of slip in time and space, and using them to learn about the physics of earthquakes.

 

Errors in Global CMT parameters due to unmodeled 3D structure

We investigate errors in centroid source parameters due to unmodeled structural hetero-

the estimated source parameters and the parameters used to compute the synthetic data

set as errors due to unmodeled structural heterogeneity and the presence of noise. We

expect that the errors obtained in this study are representative of the errors in the Global

CMT catalogue.  We find that the errors in scalar moment, moment-tensor elements and

location are small on average. The depth and centroid time are, however, biased by a

small amount.

 

 

Bias in depth and centroid time due to unmodeled 3D structure.

 

 

 

We investigate the March 25, 1998, Balleny Islands (Antarctic plate) earthquake.  The event is the largest intraplate earthquake observed in the last 30 years and it exhibited several unusual features.  The centroid-moment tensor (from the Global CMT project) has a large non-double couple suggesting that the event did not occur on a single planar fault interface.  Other investigators (Henry et al) found that the shear waves and the intermediate-period surface waves could be well matched by slip on two similarly oriented fault planes with a 100 km wide unbroken patch in between.  Our results, however, indicate that a single fault plane with continuous slip can better explain the combined body-wave and long-period surface wave data.  This is important, as the unbroken patch would indicate that dynamic triggering over the distance of 100 km would have occurred. Our results indicate that the more standard model of crack propagation due to stresses at the crack tip can explain the rupture of this event.  This example shows that in order to obtain an accurate picture of the slip history of an earthquake it is important to include many different types of data in the modeling.

 

 

Moment rate as a function of time and frequency for three different models

 

A collaborative effort headed by Mohamed Chlieh.

 

 

A collaborative effort headed by Ozgun Konca.  

 

The distribution of slip for the Great Sumatra earthquake of 2004

 

A collaborative effort including many researchers.  I worked together with Chen Ji and Alex Song to obtain a slip distribution that could explain many types of data simultaneously.  In particular the size of this great event was of interest.  We found that the event had a magnitude of close to Mw 9.1.  However, this estimate is dependent on the assumed dip angle of the fault plane.

 

Snapshot of the vertical velocity of the Earth’s surface during the last stage of rupture of the great Sumatra earthquake.  Superimposed is the ground displacement recorded at PALK, Sri Lanka (made together with Santiago Lombeyda, CACR Caltech, Jeroen Tromp, Caltech, and Rick Aster, New Mexico Tech).