V33B-4843:
3D Thermal Structure of the Alaska-Aleutian Arc with Predictions for the Metamorphic Structure and Seismic Velocities in the Subducting Slab

Wednesday, 17 December 2014
Kathryn Volk1, Peter E Van Keken1, Bradley R Hacker2 and Geoffrey A Abers3, (1)University of Michigan Ann Arbor, Ann Arbor, MI, United States, (2)University of California Santa Barbara, Santa Barbara, CA, United States, (3)Cornell University, Ithaca, NY, United States
Abstract:
An understanding of the thermal structure of subduction zones is important to gain insight into metamorphic dehydration reactions, the flow of fluids, generation of intermediate depth seismicity and the production of melts that generate arc volcanism. A tradition approach to subduction zone modeling relies on 2D cross-sections. While this approach appears adequate for subduction systems that have near-normal convergence and small changes in structure along-strike, it fails to fully represent the thermal and velocity structure of subduction systems with curved trenches or oblique subduction (van Keken & Bengston, JGR, 2003). Two-dimensional models will also likely fail near slab edges and in regions with strong changes in convergence parameters such as slab velocity, age of incoming slab, and variations in overriding plate structure. The Alaska-Aleutian arc is a prime example of a subduction zone with strong along-arc variations, which made it of key interest for interdisciplinary studies with the GeoPRISMS program (geoprisms.org). As the Pacific Plate subducts beneath North America, the Aleutian trench curves southward. Subduction also trends from trench normal near Alaska to nearly trench parallel in the far western Aleutians. These changes in the subduction zone geometry are accompanied by changes in volcanic activity, arc size, and changes in the arc to slab distance.

We have developed a full 3D model of the Aleutian-Alaska subduction system using a kinematic slab surface (following the geometry from Syracuse and Abers, G-cubed, 2006). In a comparison of 2D models we found that where subduction is most oblique the thermal structure of 2D cross-sections can differ up to ~60°C depending on whether the cross-section is normal to the trench or parallel to convergence. Due to the predicted strong 3D flow in the mantle wedge the discrepancies between 2D and 3D models will be higher.

We use the 3D model to investigate the nature of the metamorphic changes in the slab using Perplex modeling, which also provides insights into the dehydration of the slab and the seismic velocity structure. The resulting thermal-petrological-seismological structure will be made available for general use and comparisons seismological, petrological and geochemical investigations of this complicated subduction system.