Abstract
Deep-imaging multi-channel seismic reflection data show that volcanic centers along the offshore part of the Cameroon Volcanic Line (CVL) are composed of uplifted, Aptian to Late Cretaceous oceanic crust, >4 km of sedimentary overburden, and Neogene igneous rocks, with volcanic material forming a cap < 1.5 km thick over pre-uplift sedimentary deposits. At Príncipe Island, the underlying oceanic basement has been uplifted by as much as 3 km to form a crustal arch less than 200 km wide perpendicular to the CVL trend. Vertical faults having small offsets and dikes are common across this arch. Reflection Moho shallows parallel to the uplifted crust along margins of the arch, but is not observed directly below the arch axis where volcanism and faulting are pervasive. The episode of crustal uplift is marked by a prominent reflection unconformity. This unconformity occurs at other CVL islands and seamounts and represents a synchronous period of crustal uplift and volcanism. Reflectors from this unconformity have been correlated to offshore boreholes indicating a Miocene age. Gravity modelling indicates that an elongate wedge of relatively less dense lithospheric mantle (Δρ = −0.1 g/cm
3) underlies Príncipe Island to a depth of 40 km. This interpreted zone of lighter mantle material may form by a combination of intruded mafic partial melt and reheating of the lithosphere. Dynamic support from asthenospheric upwelling may also have contributed to uplift. Other NE-trending volcanic chains and rises off West Africa (Canary Islands, Cape Verde Rise, Sierra Leone Rise and Walvis Ridge) display similar features to CVL islands. These volcanic chains exhibit crustal uplift unconformities and intraplate volcanism occurring during the Miocene and later; Miocene and older marine sediments crop out on most of the islands; there are no flexural depressions surrounding volcanic centers; their ocean island basalts (OIB) have similar geochemical characteristics; the OIB does not appear to be the main construction material of each chain; anomalously high modern heatflow occurs along their lengths; and hotspot-like age progression of volcanism is not clearly defined along their lengths. It is apparent that the CVL is not the product of a single mantle plume or hotspot, and we speculate that the CVL and possibly other NE-trending volcanic chains off West Africa (and perhaps linear belts of Neogene volcanism on the African continent) are the result of linear, mantle upwelling zones or ‘hotlines’. These hotlines are suggested to form above upwelling flow currents in between cylindrical Rayleigh-Bernard convection rolls in the upper mantle. Such convection may be driven by heat transfer across and/or shear along the 670 km discontinuity as a result of convection in the lower mantle.
