Global investigation of large earthquakes using long-period seismic data

Robust earthquake source models provide key information for a wide range of applications in Earth Sciences, such as in global tectonics, natural hazards and tomographic studies. In this thesis, we study large magnitude earthquakes using the Earth’s normal modes, which are standing waves resulting from the interference of surface waves travelling in opposite directions. We start by carrying out earthquake source validation tests for a recent source inversion technique (SCARDEC), which shows a tendency for steeper fault dip angles than those reported in the widely used Global CMT catalogue (GCMT), for large magnitude (Mw ≥ 7.5) shallow thrust earthquakes. We show that SCARDEC source parameters explain normal mode data as well as GCMT, and that SCARDEC dip angles explain body wave data similarly or slightly better than GCMT solutions. SCARDEC dip angles also agree well with results from previous individual earthquake studies and with geophysical subduction zone constraints, suggesting that SCARDEC is a robust technique for rapid source parameter determinations. A new Monte Carlo earthquake source inversion method based on ultra low-frequency normal mode data (f < 1 mHz) is then developed, providing an independent way to estimate bulk rupture source parameters (rupture length and duration, magnitude, fault strike, dip and rake) of large magnitude earthquakes. Realistic synthetic tests show the importance of accurately modelling lateral heterogeneity, notably for rake angle, rupture length and duration determinations. Moreover, application of the new technique to five real shallow subduction earthquakes (Mw ≥ 8.5) of the past decade and a recent normal faulting, deep earthquake using a 3-D Earth mantle model, show clearly the complementary role of the new approach to classical earthquake source techniques, and the advantages of using normal mode data to study very long-duration, slow-slip earthquake sources, such as the 2004 Mw 9.3 Sumatra-Andaman earthquake.

Name dc:type.qualificationname

phd

Level dc:type.qualificationlevel

doctoral

Grantor dc:publisher.institution

University of East Anglia

Year dc:date.issued

2013

Author dc:creator

• Lentas, Konstantinos

Language dc:language

en