Styles of post-subduction collisional orogeny: Influence of convergence velocity, crustal rheology and radiogenic heat production

Abstract

A number of models of continental collision already exist, but the role of foregoing oceanic plate subduction or variable coupling between plates still remains to be explored. In addition, heat generation by radioactive decay may be quite variable. For example, our geochemical data from low-to-high grade metamorphic rocks of the Lesser and Higher Himalayan sequences reveal measured heat production values on the order 4–5 μW/m 3, which is considerably higher than normal inferred upper crustal values (< 1.5 μW/m 3). Therefore, we performed 2D numerical experiments of oceanic-continental subduction followed by continental collision for varying continental lower crustal strength, convergence velocity and radiogenic heating rate. Oceanic subduction produces thinning and upwarping of the overriding lithospheric wedge, such as is observed in Andean-type collisional settings (Cascadia and Central Andes). After collision two main orogenic styles develop depending on the values of key parameters: either (a) slow convergence rates (weak shear heating and consequently strong rheological coupling between plates) and/or a weak lower crust generate a two-sided, thick and narrow symmetrical collisional zone, or (b) high convergence rates (strong shear heating and therefore weak rheological coupling between plates) and a strong lower crust produce a one-sided, thin and broad asymmetrical collisional zone. In this case a vertical lithospheric body divides two upper crustal wedges that develop distinct deformation patterns. In both cases, enhanced radiogenic heating favors the formation of midcrustal partially melted channels that propagate toward the surface and the pro-foreland. Partial melting triggers lateral extrusion of metamorphic rocks from different depths (15–23 kbar, > 700 °C) on top of coherent LP-LT units. The geometry of metamorphic isograds (inverted), shapes of P-T paths at different grades and exhumation rates are similar to those observed in parts of the Himalayan chain, indicating that the models capture the essence of real systems. The model results demonstrate that internal heat production rate is a crucial parameter in determining the timing of slab breakoff, polarity reversal and exhumation of the inverted metamorphic sequence with respect to the timing of partial melting. Late stages of some numerical experiments showed interesting geodynamical evolution of the collisional zone, with the appearance of polarity reversal, roll over of both plates and roll back of the subducting slab. Irrespective of these details, hundreds of km of continental lithosphere are recycled into the mantle in all models, supporting the idea that continental crustal consumption is, indeed, a common and considerable process during collision. This affects inferences about extents of convergence and/or shortening during orogenesis. Moreover, crustal recycling into the deep mantle during collisional orogeny needs to be considered in models of crustal growth and evolution.