Upper mantle structure beneath Alaska: implications for inner core anisotropy studies from analysis of PKPdf, bc, ab absolute and differential measurements

Abstract:

Differential travel times of the inner core phase PKPdf measured relative to either of the two outer core phases, PKPbc or PKPab, are often used to study inner core anisotropy. The differential measure is assumed to eliminate contamination by crust and upper mantle structure, and source mislocation, which are common to both two phases. Differential travel times measured on quasi-polar paths between sources in the South Sandwich Islands and stations in Alaska are generally included in models of inner core anisotropy and contribute significantly to the estimate of the anisotropy strength, increasing the maximum velocity anomaly by more than a factor of 1.5 compared to when these data are excluded. Notably, these data present a much larger scatter (±3 seconds) than those obtained from other paths in the global dataset.

Here we use array methods to identify the source of the excess travel time anomaly in Alaska. We show that the contribution from strong velocity heterogeneity, including the Alaskan slab, in the upper mantle beneath the Alaskan stations is not fully eliminated in the observed differential travel time and slowness anomalies and can explain a significant portion of the scatter. Using predictions from recent regional P tomographic models of the upper mantle beneath Alaska, we show that the signature of the slab is present in trends of differential time and slowness anomalies observed both as a function of distance and azimuth across the Alaska USarray deployment. More accurate models of inner core anisotropy can be obtained by correcting core phases and, in particular, differential travel times with improved, higher resolution upper mantle tomographic models.