K.D. Koper, M.L. Pyle, and J.M. Franks
Department of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, MO.
The seismic component of the International Monitoring System (IMS) is distinguished from other global seismic networks in that a significant fraction of the stations are arrays of short-period seismometers. The advantages of array stations in event detection and location are well-known, however array stations also offer unique opportunities in deep earth structure studies. We are currently mining the IMS array database for core reflected body waves (precritical PKiKP, PcP, ScP, and associated depth phases) to map the fine and aspherical structure of Earth's core. We have constructed a data set of 302 joint observations of PcP and PKiKP phases from IMS array stations. Precritical PKiKP is notoriously difficult to detect and our data set is approximately twelve times larger than all previous published reports combined. We measured PKiKP-PcP differential travel times using a combination beamforming and cross-correlation algorithm that has a precision of about 0.1~s. The resulting differential times are generally robust and are insensitive to small variations in the slowness vector used to construct the beams. The PKiKP-PcP travel time residuals have a near-zero mean with respect to PREM and about 95\% of the residuals are between -1.5~s and 1.5~s. Most of the geographically coherent residuals can be explained by 3D mantle heterogeneities and ellipticity corrections, thus limiting lateral variations in outer core thickness to be less than 3-4~km. However, there are at least three areas that seem to require some sort of aspherical core structure. The most plausible explanation for these anomalous data is long wavelength topography on the CMB varying from a 2~km depression beneath Japan, to a 1~km depression beneath western Canada, to a 1.5~km elevation beneath northwestern Australia. This topography is most likely maintained by isostatic compensation of lateral variations in the thickness of a distinct layer at the base of the mantle (D''). We are currently calculating PKiKP/PcP amplitude ratios for the 302 confirmed observations in the IMS data base and will present our most recent findings at the conference. PKiKP/PcP amplitude ratios constrain the jump in density across the ICB, which bears on issues such as the partitioning of light elements between the inner and outer core. Past observations of PKiKP/PcP amplitude ratios have been interpreted cautiously because of substantial scatter in the data and the assumption that precritical PKiKP is only observed when it has been focused by anomalous deep Earth structure. The large size of the IMS derived data set should give a more accurate estimate of the ICB density jump as variations in PKiKP/PcP amplitude ratios caused by anomalies at the CMB are averaged out. Furthermore, the extensive geographical coverage of precritical PKiKP observations from the IMS suggest that focusing by deep Earth heterogeneities is not an observational prerequisite. The widespread observation of precritical PKiKP on the short period IMS seismometers also implies that a simple, sharp (less than 5-10~km) ICB is a global feature. This conclusion is additionally supported by the lack of precritical PKiKP observations in the IMS data base for distances of $60.0^{\circ} < \Delta < 90.0^{\circ}$. Waveform modeling studies have shown that this type of ``transparent zone'' exists only for very thin ($<$ 5-10~km) ICB models.