Chiara Civiero

In the Earth’s interior, thermo-chemical instabilities rise from the core-mantle boundary, resulting in upwelling of buoyant mantle material in the form of mantle plumes. They have been generally seen as hot mushroomed-shaped anomalies with a large round head followed by a narrow tail. However, these idealized mushroom shapes are rarely detected in seismic tomography. This implies that the classical view of how the head-tail structure develops in the mantle, it makes contact with the lithosphere and its consequences for surface volcanism needs to be revised.
In continental settings, mantle plumes have often been held responsible for anomalous high topography and massive magma production that can assist with plate breakup. The East Africa-Middle East system is an ideal place to study the potential connection between these surface phenomena and mantle plumes. From south to north, high domes stand at over ~2 km elevation including the Kenyan dome, the Ethiopian dome, the West Arabian swell, and the Turkish–Iranian plateau. The system shows evidence for active volcanism, which has persisted for at least 45 million years.
Different sources associated with either passive or active mantle upwellings have been advanced to explain the diffuse volcanism in the region. However, a conclusive model reconciling the complex pattern of the volcanism with the plate tectonics, and the underlying mantle flow is yet to be established.
Here, we use a large global waveform dataset to compute a new S-wave velocity tomographic model, which enables us to improve the resolution of the upper mantle imaging, compared to other existing models. The tomographic images reveal a spectacular star-shaped low-velocity anomaly below the rift system composed of three branches. One branch is imaged in the upper mantle below East Africa, where a quite homogeneous ‘curtain-like’ low-velocity anomaly extends from Ethiopia southwards to northern Tanzania. The second branch is a low-velocity channel underlying West Arabia that follows the eastern margin of the Red Sea and extends down to ~250 km depth. A shallower low-velocity anomaly, extending down to ~200 km depth, approximately follows the Gulf of Aden rift and is the shortest branch of the star-shaped anomaly. The anomalously high temperature of the sub-lithospheric mantle throughout the star-shaped structure, the ongoing intraplate volcanism and pronounced uplift across the areas above it suggest that this is an active plume head.
Our tomographic model reveals, for the first time, a low-velocity in the transition zone below Iraq, which bends towards west below Syria at shallower upper-mantle depths and connects with the low-velocity corridor underlying West Arabia . We identify this feature as the Levant Plume that, together with the Afar and Kenya Plumes, interacts with the thinned lithosphere and creates volcanic patterns very different from a classical hotspot track.
Plate reconstructions suggest that the Kenya plume was below Ethiopia at 40-50 Ma. In the same time range, Afar plume was below the Arabian Platform, but volcanism erupted in this period is absent. Taken together, these evidences imply that the Kenya plume may have initiated volcanism in Ethiopia while the Afar plume was still rising in the lower mantle below Arabia. Once Kenya plume impinged upon the base of the Ethiopian lithosphere, it started forming the SE-oriented line of volcanic edifices, and it is currently under Lake Victoria.
Our model brings more into focus these features that were partially seen in some of the previous models. We provide evidence from mantle tomography, plate reconstructions, and eruption ages that the volcanism over a large region of the Earth's surface is related to a single, complex-shaped plume head in the upper mantle. We show that the hot material is ponded exactly below the areas of uplift and volcanism, is inter-connected and thus comprises a single plume head shared between three mantle plumes, the Kenya, Afar and Levant Plumes.