Impact of thermal inheritance on South China Sea rifting : insight from 3D thermo-mechanical simulations

The South China Sea margins have been studied in depth due to the considerable amount of available seismic lines, and displays evidence of variable composition such as granitoids, as well as a clear structure of crustal boudinage. The continental margin has been stretched over a large distance (1000Km), in a wide rift style. The stretching took place within the subducting plate, thus implying a weak  hot lithosphere . As the crust is injected by plutons since the Triassic that migrated toward the SE until the late Early Cretaceous  and those plutons are mostly located on the NW margin of the SCS .  To the west of the Cretaceous volcanic arc, the crust is devoid of plutons except for isolated Miocene ones, and is inferred to be the former crust of a continental block (Luconia Block) docked against the margin in the Early Cretaceous. The orogen formed by this time correlates with the Yenshanian Orogen which extends from Vietnam to NE China.

Here we concentrate the modelling effort on the effect of heterogeneous heat production distribution on continental rifting by focusing on the SCS a wide rift that succeed to continental break up across a post-orogenic crust that display lateral variation in lithologies and heat production. All simulations are solved with pTatin a method that solves for conservation of energy and momentum in a nonlinear incompressible viscous fluid which viscosity depends on strain rate, temperature, pressure and stress.

The model geometry is simple. It consists of a 1000 km wide by 250 km deep domain constituted of a 40 km mechanically homogeneous crust that might be coupled (modelled with Diorite flow law) or decoupled (modelled with Quartz flow law) from the lithospheric mantle. The mantle rocks (Dry Olivine) extend down to 250km depth. In order to mimic the presence of plutons, we use a gaussian distribution of radiogenic heat production in map view and affect its value to the whole column of upper crust. The Gaussian characteristic half-width (sigma) is set to 6 km. The position in x-z of the gaussian/pluton is random but the number of plutons is chosen to ensure an average spacing of 50 km.

We find that the presence of the pluton belt influence significantly the strain rate distribution in the early stage of rifting both at lithospheric scale, by defining zone of distributed vs localized deformation, and at crustal scale by influencing the location of faults and basins. In the later stage of rifting the imprint of plutons belt become less important and the mechanical layering of the lithosphere rules which basins are abandoned and which basins continue their activity. None the less the initial presence of the plutonic belt oblique to the direction of extension induces a typical en-echelon pattern that we obtain compares well with the data in the continental rifting stage of the SCS. We find that, given the configuration of the simulation, a change in kinematics is necessary to explain the orientation of the magnetic anomaly in the west subbasin.