Volcanoes, ore deposits, and the 3d slab geometry along the Andaman-Sumatran subduction system

Abstract

This thesis addresses several topics related to the tectonic history of the Sumatran-Andaman region. Firstly, I investigated the Martabe epithermal gold deposits using high-resolution 40Ar/39Ar geochronology in order to unravel the complex overprinting alteration system. The results indicate that there were five peak periods of alunite growth around the Martabe deposits at 1.40 to 1.70 Ma, 1.90 to 2.08 Ma, 2.12 to 2.51 Ma and 3.22 and 3.48 Ma. Analysis of the Arrhenius plots put the closure temperature of alunite ranging between 390C and 519C, which is above the temperature expected for the formation of the Martabe deposits. This result gives confidence that the measured ages from 40Ar/39Ar geochronology are for the formation of alunite and not an age at which the system has cooled below the closure temperature.

Secondly, I created a 3D model of the subducted slab beneath Sumatra and the Andaman Sea, and restored the modelled slab geometry to the Earth's surface. This enabled recognition that the former spreading centres and transform faults of the Wharton Fossil Ridge localised a potential slab tear, thus circumventing otherwise enormous distortions that would have occurred during subduction. Seismotectonic analysis suggests continuing movement during subduction, in particular on the transform faults that once separated the spreading centres between the Indian and Australian plates. The emanation of fluids from the deforming lithosphere may have localised both the Toba supervolcano and the epithermal gold deposits at the Martabe.

Thirdly, tomography in the Andaman region is not characteristic of a simple subducting slab. Instead there are unusual structures between 11N and 15N that have not been previously discussed. Tomographic anomalies can be misleading and might be erased with improvements in tomographic resolution. Nevertheless, we propose two interpretations of the slab morphology that explain a westward dip anomaly. Model 1 consists of overturning the subducted Indian slab, a geometry common for advancing hinges. However, an advancing Andaman trench is not consistent with the relative motion of the Indian plate. Instead, this morphology could arise from the northward motion of the Indian plate through the mantle rotating the edge of the Andaman slab, overturning the slab and producing slab tears. Model 2 interprets the anomaly as a westward dipping slab, required a slablet derived from the Andaman Sea to have punctured the subducted Indian plate. Model 1 fits within the tectonic reconstructions published for this region but requires a geodynamic mechanism to overturn the slab, while Model 2 requires an unrecognised sutured subduction zone within the Andaman Sea.

Lastly, I report on a geomorphic analysis aimed at assessing a different hypothesis for the evolution of the Sumatran Fault System. Extensional structures in and adjacent to restraining bends on the Sumatran Fault System are unusual in an overall transpressive wrench regime. In addition, NW-SE trending tectonic lineaments appear to connect to offsets of the Sumatran Fault. Such features could be explained by westward rollback of the hinge of the subducting Indian plate driving NW-SE extension across northern Sumatra. Slab-tearing may have localised rollback and created NW-SE trending left-lateral strike-slip faults that offset the Sumatran Fault, eventually requiring the formation of new relay faults to accommodate the ongoing relative motion of the Sunda Block. The new relay faults eliminate the obstacles caused by offsets of the main fault strand. However, this model requires switching between extended periods during which transverse left-lateral strike-slip faults driven by differential slab rollback offset the Sumatran Fault, alternating with periods during which there is a renewal of transpressional wrenching. Inferred switches in the stress trajectories are complementary to those documented in the Andaman Sea. Show more