Crustal structure from gravity and seismic measurements

Abstract

Gravity data indicate that there is a regular relationship between crustal structure, crustal density (composition), and surface elevation. Earthquake and surface seismic refraction and reflection evidence as to the composition and structure of the earth's crust have not yielded a simple, unambiguous relationship to the surface elevation. The velocity dispersion of earthquake surface waves, on the other hand, indicates variations in the thickness and composition of the crust that are in general accord with the variations in surface elevation and the Bouguer gravity anomalies. Why seismic refraction measurements have not agreed everywhere with gravity and surface wave indications of crustal structure appears to be a result of masking of crustal layering. On the basis of the slope of the curve that describes the relationship between the seismic depth to the Mohorovicic discontinuity and Bouguer gravity anomalies, the density difference between the crust and the mantle appears to decrease as the thickness of the crust increases. On the assumption that the mantle has a constant mean density of 3.32 gm/cc, the mean crustal density would appear to increase from a minimum value of 2.86 gm/cc in the oceans to about 3.08 gm/cc beneath the high plateaus and mountains. If the mean crustal density is essentially constant, the effective density of the mantle must decrease by a comparable amount. The existence of a low-density zone in the upper part of the mantle, as suggested by the velocity dispersion of very long period Rayleigh waves, would explain the relationships observed. Isostatic relationships suggest that the mean density of the continental crust is essentially constant (2.85 gm/cc to 2.88 gm/cc). These values imply that a basaltic layer is present everywhere. That there is possibly an increase in mean crustal density as the crust thickens is suggested by U.S.S.R. seismic studies in Central Asia. These show that the intermediate (basaltic) layer is usually thicker beneath areas of uplift. Although the origin of the basaltic layer can only be surmised, its general inhomogeneity, as indicated by variations of seismic velocity from 6.4 to 7.3 km/sec, and its varying thickness suggest that it may be a zone of phase transformation within the underlying mantle rock. Despite the lack of homogencity in the crust, it appears possible that empirical relationships may be used to predict approximate crustal thickness from the regional Bouguer gravity anomalies or from surface elevations with a reliability approaching that for seismic measurements.