The first recorded surface rupture in southern Africa: the 22 February 2006 M 7.0 Machaze, Mozambique, earthquake

A major (M 7.0) earthquake occurred on 22 February 2006 at 22:19 UTC in the western province of Manica in Mozambique, southern Africa. The epicenter was located in Machaze, the southernmost district of Manica province, about 45 km due south of the district capital of Chitobe. The earthquake was felt throughout eastern southern Africa but caused surprisingly little damage and a very small number of casualties. Surface rupture was mapped along the NNW-striking Machaze fault and liquefaction was observed throughout the epicentral region. This is the largest earthquake to have occurred in Mozambique in historical time. The surface-rupture along the Machaze fault is generally linear, without significant topographic deviation, indicating that the fault plane is relatively steep. Displacement is normal, down-to-the-west on west-dipping fault plane. The fault rupture is marked by a series of west-facing fault scarps developed in Holocene sandy alluvium. Vertical separation across the fault scarp ranges between 0.4 m and 2.05 m. At one location, pre- existing trails crossing the fault form piercing points that indicate a component of left-lateral strike-slip displacement of up to 0.7 m. To date we have only observed part of the surface rupture along a fault length of 15 km. Time constraints and the presence of minefields have prevented full investigation of the surface rupture. The total rupture length is estimated to be on the order of 30-40 km based on the regression equations between earthquake magnitude and surface rupture length and from the time function duration from preliminary waveform inversion for the mainshock. The surface rupture was developed in a broad, flat alluvial plain, with no apparent pre-existing fault-related geomorphology. However, with the benefit of hindsight, very subdued features, including vegetation lineaments and areas of ponded sediment that appear to define a lineament along part of the trace of the surface rupture are observed on pre-earthquake remote sensing imagery. The extremely subtle and limited lateral extent of these features indicates it is unlikely that they would have been identified as being related to a potentially seismogenic source had the earthquake not happened. This begs the perennial question of how many other cryptic seismogenic structures are there waiting to be discovered? Is this an "old, slow fault" similar to those observed in intraplate regions or is this a new structure related to southward propagation of the East African Rift?