MANTLE LAYERING FROM SCS REVERBERATIONS .1. WAVE-FORM INVERSION OF ZEROTH-ORDER REVERBERATIONS

This is the first paper in a four-part sequence that examines the nature of mantle layering required by the multiple-ScS phases and internal reflections observed within the reverberative interval of SH-polarized seismograms. Here we present path-averaged estimates of the two-way travel times through the crust+mantle (tau(ScS)) and crust alone (tau(M)), the whole mantle quality factor (Q(ScS)). and the reflection coefficient at the M discontinuity (R0(z(M))) obtained by iterative waveform inversion of ScS reverberations. We show that Q(ScS) estimates derived from direct waveform inversion can be biased significantly if reverberations originating within the crust are ignored, especially in continental regions where the crust is thick. Applying our technique, which explicitly models crustal reverberations, we substantially expand the geographical coverage of tau(Scs) and Q(ScS) measurements available from previous studies. The tau(ScS) data correlate well with whole mantle travel times predicted by the Harvard tomographic models. Both the tau(Scs) data and the Q(ScS) data support the hypothesis that variations in velocity and attenuation structure between old continents and oceans persist to depths of several hundred kilometers. In particular, the value of Q(ScS) derived for paths crossing stable continent is very high (280 +/- 30 at 30 mHz) compared with that for mature ocean basin (189 +/- 16), consistent with the thick-plate model of cratons requiring that differences in thermal structure extend to several hundred kilometers. The crustal thicknesses implied by the observations of path-averaged tau(M) are consistent with those inferred from short-range refraction data, gravity measurements, and surface topography. The path-averaged reflectivity of the M discontinuity is also tectonically correlated, becoming brighter as the percentage of continental crust encountered along the path increases. This result implies that, on average, the impedance at the base of the continental crust is significantly lower than at the base of the oceanic crust.