Answers are grounded in this paper and cite their sources.
A regression line has been fitted to times of diffracted S observed at Canberra from 11 earthquakes. When corrections are made to the observed times to compensate for probable mislocations of the earthquakes, a value of 8.9 sec/deg is obtained for the travel time gradient, with a 95 per cent confidence interval estimate of 0.2 sec/deg. Assuming a core radius of 3473 km, this corresponds to an S velocity at the core-mantle boundary of 6.8 km/sec. This is 0.5 km/sec lower than the value given by Jeffreys (1939). The analysis also provides an estimate of the S time at 100° distance which is about 12 sec greater than that given in the Jeffreys-Bullen tables. The results suggest that current models of the lowest part of the mantle are significantly in error. Revision of the model in this region should enable a more satisfactory reconciliation with free oscillation data.
No takes yet. Share an insight, caveat, or question.
John Cleary (1969) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: