Thermal models of the Mexico subduction zone: Implications for the megathrust seismogenic zone

Abstract

[1] It has been proposed that the seismogenic zone of subduction thrust faults is primarily controlled by temperature or rock composition changes. We have developed numerical models of the thermal structure of the Mexico subduction zone to examine the factors that affect the temperature of the subduction thrust fault. Although the oceanic plates subducting beneath Mexico are young, the top of the oceanic plate at the trench is cool, because of the lack of a thick cover of insulating sediments. Marine heat flow observations suggest that hydrothermal circulation may further cool the oceanic plate. This results in a cool subduction thrust fault, where the brittle part of the fault extends to depths of over 40 km. At these depths, even slight frictional heating may have significant effects on temperature along the thrust fault, particularly for regions with a high convergence rate and shallow plate dip. With the addition of a small amount of frictional heating, the temperatures of the deep (30-40 km) thrust fault are increased by over 200°C. As the observed downdip limit of rupture in recent well-constrained megathrust earthquakes is confined to depths above the intersection of the thrust fault and the continental Moho, a temperature of 350°C may control the downdip extent of the seismogenic zone. Thus, in order to be consistent with the observed shallow rupture areas, it is necessary to include a small amount of frictional heating, corresponding to an average shear stress of 15 MPa.INDEX TERMS: 3015 Marine Geology and Geophysics: Heat flow (benthic) and hydrothermal processes; 3040 Marine Geology and Geophysics: Plate tectonics (8150, 8155, 8157, 8158); 7230 Seismology: Seismicity and seismotectonics; 8130 Tectonophysics: Evolution of the Earth: Heat generation and transport; KEYWORDS: thermal models, subduction zone, megathrust earthquakes, seismogenic zone, heat flow, Mexico subduction zone Citation: Currie, C. A., R. D. Hyndman, K. Wang, and V. Kostoglodov, Thermal models of the Mexico subduction zone: Implications for the megathrust seismogenic zone,

Low heat flow from young oceanic lithosphere at the Middle America Trench off Mexico

“…This result is consistent with observations further south on the Mexican margin, where values de-crease rapidly landward [31,32]. Conversely, the application of critical taper theory [33] would suggest high basal friction since the wedge angle is large.…”

Distribution of hydrous minerals in the subduction system beneath Mexico

“…In central Mexico, significant reduction in S wave velocity by more than 20% (Song et al, 2009) at the shallow-to-flat transition of the top interface of the Cocos plate ( $ 20245 km depth) are difficult to explain simply by velocity difference predicted for plausible subducted oceanic crust lithologies such as lawsonite blueschist and epidote blueschist (Currie et al, 2002), which is only less than 7% (Hacker et al, 2003a). In comparison to previously reported values for central Mexico subduction zone, the Vp/Vs ratios (or Poisson's ratios) shown in Audet et al (2009) and Peacock et al (2011) for Cascadia subduction zone are quite extreme, and both Audet et al (2009) and Peacock et al (2011) argued that such high estimates cannot be due to the presence of hydrous minerals alone but are due to free-water at high porepressure.…”

Seismic variability of subduction thrust faults: Insights from laboratory models

“…It is generally accepted that the temperature range at which interplate seismogenic zones develop is from 100-150°C to 350-450°C [i.e., Hyndman and Wang, 1993;Wang, 1995;Hyndman et al, 1997;Currie et al, 2002]. On this basis, the updip limit of seismogenic thrust faulting is alternatively explained by the transition illite-smectite [Pytte and Reynolds, 1988] or by fault gouge lithification processes [Saffer and Marone, 2003;Marone and Saffer, 2007], while the downdip limit is possibly controlled by the brittle-ductile transition of the crustal material [i.e., Brace and Kohlstedt, 1980] or by the intersection of the slab with the forearc mantle wedge [Peacock and Hyndman, 1999].…”