The Malawi earthquake of March 10, 1989 - Deep faulting within the east-African rift system

Since 1964 there have been six earthquakes of Mw ≥ 5.5 in east Africa whose centroid depths have been demonstrated to be in the range of 25–40 km. These depths are significantly greater than the 5- to 15-km range typical of most other regions of continental extension. The March 10, 1989 earthquake (Mw 6.1) in Malaŵi is the first such deep event to have occurred within the main topographic expression of the late Cenozoic east African rift system. Its focal mechanism and depth (32 ± 5 km) allow it to be plausibly associated with slip on a deep part of a major normal fault zone bounding the Malaŵi rift. We cannot determine whether the earthquake occurred in the crust or mantle or whether the postulated fault zone exists as a continuous seismogenic surface from the upper crust to depths of ∼30 km: it is possible that the fault zone exists as an aseismic shear zone in the lower crust. In the Malaŵi rift the width of the half graben (up to 50 km), the effective elastic thickness of the lithosphere (∼35 km), and probably the largest fault segment lengths (>50 km) are greater than is typical in rifts outside Africa. We suggest that these features and the greater earthquake depths are all related to the likelihood that the upper part of the lithosphere is colder and stronger than is typical elsewhere. These observations are consistent with earlier suggestions that normal faulting and significant strength can exist throughout the bulk of the crustal thickness. If this is the case, wide half graben can form without requiring shear strengths on the bounding faults to be greater than 1–10 MPa (10–100 bars), which is the typical level of stress drop observed in earthquakes