Polet, Jascha (1999) Seismological observations of upper mantle anisotropy [pt. 1] ; Source spectra of shallow subduction zone earthquakes and their tsunamigenic potential [pt. 2]. Dissertation (Ph.D.), California Institute of Technology. http://resolver.caltech.edu/CaltechTHESIS:02092011-083952190
One of the most important developments in observational seismology in the last 10 years is the worldwide increase in the number of broadband instruments and seismic networks, as well as the improved access to the data-set that these seismometers provide. A data-set of this magnitude offers nearly unlimited possibilities for research into earthquake source processes and Earth structure. The work presented in this thesis involves the application of different methods to seismological recordings, as well as an interpretation and discussion of the results. In Chapter I, I take advantage of the very broadband nature and small spacing of the stations of TERRAscope, one of the first digital broadband seismic networks, to determine dispersion curves for long period surface waves. This enables us to invert for an upper mantle S-wave velocity model for southern California. The Rayleigh wave, SV, model is about 4% slower than the model developed for tectonic north America. If the correction for higher modes I performed on our Love wave data measurements is accurate, the resulting SH velocity model shows about 5% anisotropy (transverse isotropy) in the upper mantle beneath southern California. In Chapter 2, I perform measurements of shear-wave splitting on a unique data-set obtained from temporary arrays located above the Nazca subduction zone in South America. Data from SKS, and local S-wave data from deep and intermediate depth earthquakes, were used to develop a model of the anisotropy in this region. The above slab component of anisotropy in the western region, where the slab is at a depth of about 300 km and up is oriented NS and its delay time is limited to about 0.3 sec. This direction agrees with the shortening direction of the Andes and is orthogonal to that predicted by a comer flow model. To the east, the stations have EW aligned fast directions and possibly sample the Brazilian craton. The below slab component samples a zone of EW aligned anisotropy, as well as trench parallel aligned anisotropy. The trench parallel directions can be explained by the retrograde motion of the slab in south America, and I speculate that the EW direction could suggest a tear in the slab, or a local EW orientation because of local buckling of the slab under NS compression. In Chapter 3 I use this same method for the data of the TriNet array. Here I find an overall pattern of consistent directions of the polarization direction of the fast SKS waves, the fastest P-wave velocities and the World Stress Map maximum horizontal compressive stress directions. This suggest that the pattern of anisotropy is generally uniform in the crust and lithospheric mantle, in a layer with an overall thickness of 100 to 150 km. The alignment of most fast directions can be explained by plate-tectonic, extensional and compressional events. We also examine the detailed lateral and vertical variations of anisotropy in this region. Chapter 4 is focussed on the differences in source spectra and tectonic setting between tsunami earthquakes, which excite anomalously great tsunamis and 'regular' shallow subduction earthquakes. We find that these unusual events have several characteristics in common: low energy release at short periods, centroid location close to the trench, updip rupture, relatively small accretionary prism, sediment subduction and a well-developed horst and graben structure of the oceanic plate close to the trench. We speculate that these events can nucleate in an unusually shallow part of the subduction zone, where sediments normally exhibit stable sliding behavior, because of the contacts between the horsts and the overriding plate. Because the earthquakes are so shallow, and there is some sediment being subducted, part of the rupture goes through sediments, making the source process slow. The true displacement (and thus the tsunami height) of these events may be underestimated because the elastic constants of the fault zone are not taken into account when converting seismic moment into displacement.
|Item Type:||Thesis (Dissertation (Ph.D.))|
|Degree Grantor:||California Institute of Technology|
|Division:||Geological and Planetary Sciences|
|Thesis Availability:||Restricted to Caltech community only|
|Defense Date:||29 June 1998|
|Default Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Dan Anguka|
|Deposited On:||09 Feb 2011 17:23|
|Last Modified:||26 Dec 2012 04:33|
- Final Version
Restricted to Caltech community only
See Usage Policy.
Repository Staff Only: item control page