Present seismic evidence for a boundary layer at the base of the mantle

Abstract

To improve the global coverage needed in investigating the observational consequences of boundary layer models, the existing data set has been extended with long‐period diffracted P and SH from the SRO network, and short‐period diffracted P from the NORSAR array. It was found that on average, the amplitude decay of diffracted SH is relatively small, in accordance with the majority of previously published WWSSN data, and consistent with the observed weakness in dispersion of this phase. These results, in combination with the apparent absence of anomalously high dT/dΔ, can be explained by a relatively thin (≤100 km) low‐velocity zone at the base of the mantle. Long‐period diffracted P wave data probably lack the required resolution for detecting a relatively thin transition zone. Such a zone, with relatively low velocity, is allowed by an average of travel time and amplitude data across the core shadow boundary and appears to be required by the NORSAR observations of short‐period diffraction far into the shadow. However, recent results of individual short‐period P wave studies pertaining to geographically limited regions only (including also the NORSAR observations) are partly conflicting unless lateral variations exist. On balance, current thermal models appear to be consistent with (but not required by) the present seismic evidence.