A Numerical and Analogue Study of Dike Ascent in Asymmetric Continental Rift Zones

Thursday, 17 December 2015: 08:30
304 (Moscone South)
Jana Schierjott1,2, Francesco Maccaferri1, Valerio Acocella3 and Eleonora Rivalta1, (1)Deutsches GeoForschungsZentrum GFZ, Potsdam, Germany, (2)ETH Swiss Federal Institute of Technology Zurich, Zurich, Switzerland, (3)Univ. Roma Tre, Rome, Italy
In continental rift zones, tectonic extension generates deep topographic depressions, typically graben or half-graben structures, confined by large border faults. Volcanism may be distributed within, at the border and outside of the depressions, and the mechanisms controlling this distribution are debated. Recently, Maccaferri et al. (2014) proposed that the reorientation of the principal stresses linked to crustal thinning and overall crustal mass redistribution in rift zones modifies the expected trajectory of ascending magma pockets and plays a fundamental role in the distribution of volcanism at the surface. However, the model does not explain why volcanism is asymmetric in most continental rift zones. The goal of this study is to investigate the relation between the characteristic distribution of volcanism at the surface, the distribution and geometry of magma storage at depth, and the observed geometric asymmetry of the grabens at most rift zones. By using a boundary element model for dike propagation and analogue laboratory experiments we evaluate the ascent path of magmatic dikes in asymmetric continental rifts.

We find that the position of the magma source along the cross section of the rift and its spatial extent and the asymmetry of the graben cross section are the most important factors controlling one-sided volcanic activity at surface. For dikes starting beneath the rift's center, the more asymmetric the rift structure the more likely is asymmetric volcanic activity. Dikes are deflected to the shallow rift side and no volcanism develops on the deep side or only focused in one spot. However, if the position of the magma ponding region is offset towards the deep side of the graben, the dikes tend to emerge on the rift shoulder adjacent to such deep side. To a minor extent, also the starting depth of the dikes, any topographic loading on the graben flanks due to flank uplift and the background tectonic stress impact the surface distribution of volcanism. Mostly, all the latter mechanisms influence the distance between the graben's edge and dike arrival at surface. Our analogue experiments show similar results supporting that the graben geometry and the location where the dikes nucleate are major controls on the surface distribution of the magmatism in rifts.