CENOZOIC PALEOSTRESS AND KINEMATIC EVOLUTION OF THE RUKWA - NORTH MALAWI RIFT-VALLEY (EAST-AFRICAN RIFT SYSTEM)

The kinematic evolution of the Tanganyika - Rukwa - Malawi rift zone of the western branch of the East African Rift System has been intensely debated, but there is still a major controversy about the role of dip-slip versus strike-slip faulting in rift development, as well as about the principal extension direction. A detailed field survey was carried out in Western Tanzania, along the northern part of the Lake Malawi rift valley, and at its triple junction with the Rukwa rift valley and the transversal Usangu trough. Paleostress tensors have been determined from fault-slip data, for a total of 976 minor faults on 29 sites distributed along major border faults and in rift sediments and volcanics. From field data, and as evidenced by previous works, it is clear that the general NW-SE fault trend of the Rukwa-North Malawi rift segment is strongly influenced by the Precambrian structural trend, and partly inherited from a Karoo (Permo-Triassic) rifting phase. This work focuses on the Cenozoic rifting period, demonstrating two major stages of paleostress state : I : an initial near-radial extensive stress regime (principal compression vertical, magnitude of intermediate stress close to magnitude of minimum stress), with two dominant extension directions (ENE-WSW and NW-SE), which generated or reactivated major rift border faults with a dominant dip-slip component; II : a subsequent strike-slip stress regime (intermediate compression vertical), with dominantly N-S principal compression, which mainly caused strike-slip reactivation of major rift border faults. Stratigraphic constraints, provided by dated lava flows, fix the stress inversion from radial extensive to strike-slip regime in the Middle Pleistocene, between 0.55 My and 0.42 My. Phase I represents the kinematic evolution from Late-Miocene to Pleistocene, It could also account for the major part of basin development during the Cenozoic (reactivation of the North-Malawi and Rukwa rift basins and neoformation of the Usangu basin) and it is associated with two magmatic pulses in the Rungwe Volcanic Province. The second kinematic phase started in Middle Pleistocene following a stress inversion caused by the permutation of the compressive and intermediate stress axis, leading to a strike-slip regime. Once established, this new regime evolved rapidly by successive clockwise rotations of the NNW-SSE maximum compression, first to N-S (dominant regime), and then to NNE-SSW (last paleostress regime recognized).The stress inversion is also marked by the onset of a third magmatic pulse in the Rungwe Volcanic Province. Phase II caused dextral strike-slip reactivation of the major NW-SE trending normal rift faults, and sinistral strike-slip reactivation of a system of NE-SW subvertical joints of pre-Karoo origin. A comparison with earthquake focal mechanisms in the area indicates that the present stress conditions of its horizontal compressive and extensive axes show similarities to the last recorded paleostress state. This kinematic evolution appears to be at least on a regional scale, as indicated by similar observations made by other teams in the Central Kenya Rift and along the western coast of Lake Malawi. The causes for this paleostress evolution have not yet been established, but it could be related to modifications in the kinematics of the African lithospheric plate.