Gregory obtained his bachelor's degree from University of California, Santa Cruz in 1982 and his PhD degree from MIT in 1989. He became a faculty in Stanford geophysics in 1990 after being a post-doc researcher at MIT.[5]
Research
Fingerprint And Similarity Thresholding (FAST)[6][7]
The Fingerprint And Similarity Thresholding algorithm was developed by Beroza group to "efficiently detect previously overlooked microquakes".[7] This method can analyze week-long seismic information in less than 2 hours, 140 times faster than the traditional autocorrelation method. Furthermore, the new technique would help better monitor and categorize earthquakes.
Greg's team measured stress drops in a number of human-induced and natural earthquakes in central US. They found the ground motions in induced and natural earthquakes are largely the same. The results suggest the ground motion prediction equations can be also applied to human-induced earthquakes and can be used to reduce the earthquake hazards in central US.