| Summary: | This dissertation investigates dynamic processes in the mantle using source parameters, waveforms, and travel times associated with deep-focus earthquakes. First, I systematically determined the focal mechanisms and depths of all large-to-moderate sized earthquakes that occurred along the Kuril-Kamchatka in the past 30 years. My results show that the configuration of the subducting slab and the state of strain within the subducted lithosphere are highly variable along both the strike and dip. In particular, I observed down-dip compressional strain in the northern and central portions of the arc between depths of 200 and 650 km. In these regions, deformation of the subducted slab near the base of the upper mantle is delineated by seismicity, probably in response to resistance encountered near the 670-km discontinuity. Second, I analyzed P and SH waveforms from eight deep-focus earthquakes in subduction zones of the northwestern Pacific to investigate whether complexities in broadband waveforms can be adequately explained by simple source models. Both sub-horizontal and sub-vertical ruptures are found within the same subducting system. At a depth of approximately 480 km, the length of one such rupture constrains the seismogenic thickness of the subducted slab to be approximately 40 km. Finally, using both digital and analog seismograms, I precisely measured approximately 400 arrival times of teleseismic P phases for seven earthquakes, most of which occurred since 1987 in the Kuril-Kamchatka. In this region, fast P residuals of deep-focus earthquakes at azimuths to the north-northeast and south-southwest are often interpreted as evidence for a deep-penetrating slab. Along these azimuths, I also observed that P residuals of a deep-focus earthquake in the northern Kuril-Kamchatka are significantly faster than those of both deep- and shallow-focus earthquakes located beneath the Sakhalin Island near southern Kuril. My observations suggest that seismic wave speeds below depths of approximately 650 km vary along the arc. A synthesis of my results shows a correlation between the state of strain within subducted lithosphere in the upper mantle, as determined from source parameters, and the extent of slab penetration into the lower mantle, as inferred from travel-time analysis. (Abstract shortened by UMI.)
|
|---|