Mechanisms of isostatic compensation in the vicinity of the East African Rift, Kenya

Abstract

Summary The topography and gravity anomalies in the area surrounding the East African Rift in Kenya can be modelled as the sum of the effects of surface and subsurface loading of an elastic plate. Assuming surface and subsurface loading are independent processes, the observed coherence between the 2-D Fourier transforms of Bouguer gravity and topography provides a constraint on the effective elastic thickness or flexural rigidity of the plate. Distinct linear segments of the log gravity power spectrum suggest that components of the gravity field with wavelengths of 250–1000 km are generated predominantly by a density contrast at a depth of about 32 km. Most shorter wavelength gravity anomalies are probably associated with source depths of less than 1 km and indicate variations in thickness of low density sedimentary or volcanic layers or lateral variations in density of the surface rocks. A simple density model based on these estimates and the geology of Kenya consists of a cover layer averaging 0.5 km thick with density 2300 kg m−3, a layer 32 km thick with density 2800 kg m−3, and an underlying half-space with density 3200 kg m−3. Using this density model and assuming loading due to relief on all three density interfaces, the elastic thickness that best predicts the observed coherence in the least squares sense is 25 km (Flexural rigidity of 1.4 × 1023Nm). Amplitudes of loads on the density interfaces can be calculated based on the model response. Topography with wavelength greater than 650 km is locally compensated making surface and subsurface loading indistinguishable. At shorter wavelengths, rift volcanics and volcanic cones including Mts Kenya, Elgon and Kilimanjaro can be identified as surface loads. The largest amplitude subsurface load is an upward directed, regionally supported load beneath the Kenya Dome that may correspond to a region of hot, low density mantle recognized by other geophysical studies. In creating the topography for which surface and sub-surface loading are distinguishable, surface loading is much more important than doming due to subsurface loading.