Refine fault geometry with broadband waveform modeling for earthquake source parameters
About the Event:
Global earthquakes catalogs have provided one of the most important evidences for plate tectonics. However, the resolution in the modern global catalogs, both location and focal mechanism, are usually not good enough to well resolve the geometry of the faults where mega-quakes occur. Here we conduct broadband waveform modeling for medium size earthquakes to refine their source parameters, which include location, in particular for depth, and fault plane solutions. We proposed a novel method that allows us to model the teleseismic P-waves up to 1.5Hz. This method uses well-constrained medium size earthquakes to establish amplitude amplification corrections for teleseismic stations that were later used to invert the high frequency waveform of other nearby events. The new approach has the power to resolve some of the early aftershocks, which is not possible in current global catalogs. We also conduct high frequency waveform modeling of the depth phases that provide strong constraints on earthquake depths. For the earthquakes with good signal to noise ratio waveform records, we performed long period waveform inversion along with the bootstrapping method to estimate the uncertainty of strike, dip and rake, as well as horizontal location. Our approaches are applied to the 2015 Nepal earthquake sequence and the Mentawai earthquakes since 1990. For the Nepal sequences, our results reveal about a dozen events more than the Global CMT catalog. The refined earthquake locations delineate a ramp-flat-ramp Main Himalaya Thrust (MHT) geometry that is in agreement with the structure geology, receiver functions and gravity analysis. We also found that the Mw7.2 aftershock in the sequence did not rupture the MHT but a shallower fault with dip angle around 10 degrees. For the Mentawai sequence, our refined catalog shows that the 2009 and 2005 “back-thrust” sequences actually occurred on the steeper dipping fault that contradicts with the previous studies. These events probably have reactivated the splay faults that were formed at the earlier formation stage of the accretionary wedge. We also found that the dip angles of the main plate boundary earthquakes are systematic 5 to 10 degree larger than the fault dip delineated by the seismicity. Such discrepancy could either explained by the irregular topography of the plate interface (e.g. seamount) or the fault zone structure above the plate boundary.
The application of our approaches to both Nepal and Mentawai sequences shed some new lights on reconsidering the fault geometry around subduction zone plate interface.
About the Speaker:
Wang Xin obtained his Ph.D. in geophysics from the Institute of Geology and Geophysics, Chinese Academy of Sciences in Dec. 2016. He then jointed in Earth Observatory of Singapore as a research fellow in seismology. His research interests include in seismic velocity anomalous and discontinuities, and earthquakes source properties, in particular, those associated with the subduction zones. His recent research focuses on understanding the crustal-scale velocity structure in Myanmar and the fault geometry in Sumatra.