Using seismic array-processing to enhance observations of PcP waves to constrain lowermost mantle structure

Monday, 15 December 2014: 5:15 PM
Sergi Ventosa1 and Barbara A Romanowicz1,2, (1)Institut de Physique du Globe de Paris, Paris, France, (2)University of California Berkeley, Berkeley, CA, United States
The topography of the core-mantle boundary (CMB) and the structure and composition of the D” region are essential to understand the interaction between the earth's mantle and core. A variety of seismic data-processing techniques have been used to detect and measure travel-times and amplitudes of weak short-period teleseismic body-waves phases that interact with CMB and D”, which is crucial to constrain properties of the lowermost mantle at short wavelengths. Major challenges in enhancing these observations are: (1) increasing signal-to-noise ratio of target phases and (2) isolating them from unwanted neighboring phases. Seismic array-processing can address these problems by combining signals from groups of seismometers and exploiting information that allows to separate the coherent signals from the noise.

Here, we focus on the study of the Pacific large-low shear-velocity province (LLSVP) and surrounding areas using differential travel-times and amplitude ratios of the P and PcP phases, and their depth phases. We particularly design scale-dependent slowness filters that do not compromise time-space resolution. This is a local delay-and-sum (i.e. slant-stack) approach implemented in the time-scale domain using the wavelet transform to enhance time-space resolution (i.e. reduce array aperture). We group stations from USArray and other nearby networks, and from Hi-Net and F-net in Japan, to define many overlapping local arrays. The aperture of each array varies mainly according (1) to the space resolution target and (2) to the slowness resolution required to isolate the target phases at each period. Once the target phases are well separated, we measure their differential travel-times and amplitude ratios, and we project these to the CMB. In this process, we carefully analyze and, when possible and significant, correct for the main sources of bias, i.e., mantle heterogeneities, earthquake mislocation and intrinsic attenuation.

We illustrate our approach in a series of regional studies of the CMB and D” using P and PcP observations with unprecedented resolution, for events with magnitude Mw>5.4 and distances up to 80 degrees. Regions sampled span Alaska and the north of Canada, inside and outside of the northwest border of the Pacific LLSVP, and up to its eastern border from central America.