PKP TRAVEL TIMES AT NEAR ANTIPODAL DISTANCES: IMPLICATIONS FOR INNER CORE ANISOTROPY AND LOWERMOST MANTLE STRUCTURE

Abstract

Previous studies from PKP(AB–DF) differential travel times at large distances suggest that the central part of the inner core is very anisotropic. These differential times, however, can be affected greatly by strong heterogeneity in the lowermost mantle. Here we examine a unique data set of PKP travel times from global digital and analog stations at near antipodal distances, where the effects of both inner core anisotropy and mantle heterogeneity are the greatest. Our results show that the AB–DF residuals for the polar paths are consistently larger than those of the equatorial paths by over 3–4 standard deviations. We also measured DF and AB absolute times, and found that the DF residuals are negatively correlated with the AB–DF residuals while the AB residuals have a much weaker correlation with the AB–DF residuals. We compare several mantle models with the data. Our results suggest that the mantle structure can explain part of the residuals of the equatorials paths, but cannot explain the polar path anomalies. These results strongly suggest that most of the AB–DF anomalies for the polar paths are likely from the inner core anisotropy and not from mantle heterogeneity. Assuming a uniform cylindrical anisotropy model, the average inner core anisotropy amplitude is about 2.5%. On the other hand, equatorial paths from events in the west Pacific and recorded at South America show a steep azimuthal change in AB–DF times (about 4 s over a 60° azimuthal range). The sharp change is well predicted by existing P and S tomographic models: the azimuthal change occurs as the AB paths sweep across the great slow anomaly in the central Pacific. The high correlation between the P and S velocities suggests that the central Pacific slow anomaly may be of thermal origin.