RATIO OF RELATIVE S-TO-P-VELOCITY HETEROGENEITY IN THE LOWER MANTLE

Citation
Gs. Robertson et Jh. Woodhouse, RATIO OF RELATIVE S-TO-P-VELOCITY HETEROGENEITY IN THE LOWER MANTLE, J GEO R-SOL, 101(B9), 1996, pp. 20041-20052
Citations number
39
Categorie Soggetti
Geochemitry & Geophysics
Journal title
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
ISSN journal
21699313 → ACNP
Volume
101
Issue
B9
Year of publication
1996
Pages
20041 - 20052
Database
ISI
SICI code
2169-9313(1996)101:B9<20041:RORSHI>2.0.ZU;2-P
Abstract
The parameter nu = partial derivative lnv(s)/partial derivative lnv(p) , the ratio of relative S to P velocity heterogeneity, can be measured seismically and is important in determining the mineral physics of th e deep mantle. However, seismic results are inconclusive because they have generally been based on P and S velocity models which are similar , but not proportional, and which are derived from seismic data sets i n different frequency ranges and sample the mantle in different ways. Hence current values of nu are subject to inconsistencies which arise from comparing models of differing resolution. Furthermore, heterogene ity within the Earth may be frequency dependent, and this characterist ic may also bias v. We overcome these difficulties by computing P and S models from data sets with the same period and similar data coverage s and by deriving the ratio in its most rigorous form, whereby we assu me proportionality of P and S models and calculate nu as a proportiona lity factor. We proceed by jointly inverting P and S arrival time data from the International Seismological Centre between the years 1964 an d 1993 for ratios and compressional models expanded to at most degree 8 laterally in terms of spherical harmonics and up to degree 6 in dept h in terms of Legendre polynomials. We also quantify the sensitivity o f our results to data weighting and three-dimensional model parameteri zation and determine to what depth we are justified in assuming propor tionality of P and S heterogeneity. We parameterize nu in terms of Leg endre polynomials and find that there is an insignificant improvement in variance reduction beyond degree 1. Hence we find that nu increases linearly from 1.7 to 2.6 between 700 and 2000 km depth, which is the greatest depth to which we believe our results are reliable. We also c ompare our results with those from theoretical mineral physics and fin d that seismically derived values are significantly larger and increas ing at a greater rate. Reconciliation of these results requires at lea st modifications to the physical mechanisms currently postulated.