A new constraint on the composition of the topmost continental mantle-anomalously different depth increases ofPandSvelocity

Abstract

S U M M A R Y Compared to clear crustal P-and S-waves as well as a strong refracted P,,-wave propagating in the topmost mantle (i.e., 10 to 15 km below the Moho), an anomalously weak appearance of the corresponding refracted shear wave S,, is observed on refraction seismic profiles in SW Germany, France, coastal Maine (NE United States) and Fennoscandia. It is possible that the observation of Pn/PMP ratios close to unity combined with small S,/SMS ratios (i.e., small S, , amplitudes) is typical for the continental uppermost mantle since this was observed for quite different types of continental crust. It is shown that the observed phenomenon is not a source effect. Several possible causes to explain the observations are studied, also with the aid of synthetic seismograms. The effects of temperature, attentuation and anisotropy have been investigated. Their influence cannot explain the observations. The most likely candidate is a depth-increasing vp/vs ratio (an increase in the order of 0.027 over a depth range of lOkm), forcing P-and S-waves to propagate on different paths in the topmost mantle. The observed phenomenon provides new constraints on the change of composition with depth for the crust mantle transition zone with basalt depletion as the most likely explanation.

Investigations ofSnandPnphases in the area of the upper Rhine Graben and northern Switzerland

“…The attenuation in the crust is fixed to a Q, of 600. Considering the results of Gajewski et al (1990), who proposed a Q, of around 300 for the uppermost mantle and the Q, of 500, which Durst (1981) has derived for the crust in the Rhine Graben area, the Q, of the mantle in our model is varied from Q, = 200 to Q, = 600. Of course, a higher Q value for S waves in the uppermost mantle leads to an increase in S,, amplitudes ( Fig.…”

“…For better identifying of the main body phases, the near-surface layers in the model are defined as zones without reflectivity. We have chosen a Qs of 400 for the crust as well as for the mantle, which is only slightly higher than the Q, of around 300 given by Gajewski et al (1990) for the topmost mantle. We varied the submoho velocity gradient in a stepwise approach to fit the observed amplitude ratios of refracted to direct and reflected phase.…”

Lithospheric structure of the Qiangtang Terrane, northern Tibetan Plateau, from complete regional waveform modeling: Evidence for partial melt

“…In order to estimate attenuation, we computed synthetic seismograms for lithospheric Qp = 50, 100, 200, 300, 400, and 500. For each model, S wave attenuation was related to P wave attenuation by the relation QP = 2.25 Qs [e.g., Gajewski et al, 1990]. Elastic structure was fixed to that described above (Table 2).…”

Observation of converted Moho reflections in the north-west of the Iberian Peninsula

“…This behaviour of phase S,, amplitudes not being detectable in the presence even of clear P, arrivals, which is common for refraction seismic profiles in north-west Spain (Tellez 1993), has also been observed in other parts of the Variscan Belt (e.g. Haggag 1980Gajewski & Prodehl 1985 and in other areas with very different types of continental crust (El-Isa, Mechie & Prodehl 1987;Stangl et al 1989;Gajewski et al 1990).…”