Implications of fault array evolution for synrift depocentre development: insights from a numerical fault growth model

We investigate the effects of interaction between growing normal faults on the creation of accommodation in extensional half grabens. Fault evolution is simulated using a numerical model in which we calculate both the stress field around each fault and the changes in stress level on neighbouring faults caused by individual slip events (earthquakes). These stress changes govern the interaction and determine both the direction and rate of lateral fault propagation and the accumulation of displacement. The spatial distribution of subsidence resulting from fault growth is examined parallel and transverse to strike to document the three-dimensional evolution of accommodation creation. The numerical experiment permits analysis of the geometry, inception and growth history of individual fault-controlled depocentres during development of a linked normal fault array. Model results indicate that the pattern of fault-controlled subsidence, and hence hangingwall accommodation creation, is spatially and temporally variable as the timing and rates of fault movement vary considerably. In general, there is a progression from many relatively small and isolated subbasins initially, to the development of a larger laterally continuous half graben in the hangingwall to a major through-going linked fault system. Many smaller faults initially active in footwall and hangingwall areas become inactive as the extension localizes onto the major structure. This switching-off of some faults, combined with focusing of the extensional deformation, leads to an increase in the rate of displacement accumulation on the remaining active fault. As dip-slip displacement is a proxy for hangingwall subsidence, we interpret this prominent rate change in terms of the 'rift-initiation' to 'rift-climax' transition, previously recognized in synrift stratigraphy. A general picture of depocentre development relative to timing of linkage emerges from the simulated fault evolution that provides some simple conceptual models to be tested. However, the important result of this paper is that it shows the degree to which fault activity can vary in space and time both along the same fault zone and also across strike. We briefly discuss the implications of model results for stratigraphic architecture in rift basins. Our conclusion is that some of the stratigraphic complexity of rifts previously ascribed to other controls (e.g. sediment supply, eustasy, etc.) may be tectonically controlled and that with improved three-dimensional imaging of rift basins these effects may be recognized.