Modeled temperatures and fluid source distributions for the Mexican subduction zone: Effects of hydrothermal circulation and implications for plate boundary seismic processes

Abstract

In subduction zones, spatial variations in pore fluid pressure are hypothesized to control the sliding behavior of the plate boundary fault. The pressure‐temperature paths for subducting material control the distributions of dehydration reactions, a primary control on the pore fluid pressure distribution. Thus, constraining subduction zone temperatures are required to understand the seismic processes along the plate interface. We present thermal models for three margin‐perpendicular transects in the Mexican subduction zone. We examine the potential thermal effects of vigorous fluid circulation in a high‐permeability aquifer within the basaltic basement of the oceanic crust and compare the results with models that invoke extremely high pore fluid pressures to reduce frictional heating along the megathrust. We combine thermal model results with petrological models to determine the spatial distribution of fluid release from the subducting slab and compare dewatering locations with the locations of seismicity, nonvolcanic tremor, slow‐slip events, and low‐frequency earthquakes. Simulations including hydrothermal circulation are most consistent with surface heat flux measurements. Hydrothermal circulation has a maximum cooling effect of 180°C. Hydrothermally cooled crust carries water deeper into the subduction zone; fluid release distributions in these models are most consistent with existing geophysical data. Our models predict focused fluid release, which could generate overpressures, coincident with an observed ultraslow layer (USL) and a region of nonvolcanic tremor. Landward of USLs, a downdip decrease in fluid source magnitude could result in the dissipation in overpressure in the oceanic crust without requiring a downdip increase in fault zone permeability, as posited in previous studies.

Insights into the causal relationship between slow slip and tectonic tremor in Guerrero, Mexico

“…This relation implies that changes of dilatancy every 20 days are instantaneously translated into changes of pore pressure. Since the permeability is low within the oceanic crust [ Audet et al ., ; Peacock et al ., ; Perry et al ., ] where fluids are likely to be present in Guerrero [ Song et al ., ; Kim et al ., ], fluid diffusion is slow enough to make our approach valid for the time span of the SSE [ Villafuerte , ]. To confine fluids within the top 5 km of the subducted slab we considered B = 0.9 in that layer and B = 0 elsewhere.…”

Diagenetic, metamorphic, and hydrogeologic consequences of hydrothermal circulation in subducting crust

“…Also in this region, Song et al (2009) identify ultraslow seismic-velocity zones in the subducting crust, attributed to excess fluid pressures. Perry et al (2016) generate thermal models of the margin and link them with petrologic models to predict the distribution of subducting slab alteration and dehydration. Models that include hydrothermal circulation predict regions of focused fluid release from subducting sediment in the areas of tremor, slow slip, and the ultraslow layer (Perry et al, 2016).…”

“…Perry et al (2016) generate thermal models of the margin and link them with petrologic models to predict the distribution of subducting slab alteration and dehydration. Models that include hydrothermal circulation predict regions of focused fluid release from subducting sediment in the areas of tremor, slow slip, and the ultraslow layer (Perry et al, 2016). In models without hydrothermal circulation, subducting sediment heats and dehydrates substantially before reaching the flat section of the slab, leaving very little fluid for release in the areas of tremor, slow slip, and the ultraslow layer (most dehydration of the subducting basement rock is predicted to occur deeper than the flat-slab section) (Perry et al, 2016).…”

The 19 September 2017 (Mw7.1) Intermediate‐Depth Mexican Earthquake: A Slow and Energetically Inefficient Deadly Shock

“…Figure 4d shows the temperature rise for different widths predicted by equation (2). According to thermal models of the subducted Cocos plate, the temperature of the slab where the earthquake took place is between 600 and 700 • C (Manea & Manea, 2011;Perry et al, 2016). The melting temperature of peridotite and gabbroic rocks at those depths (i.e., at 1.5-2 GPa) range between 1400 and 1800 • C (Katz et al, 2003;Nielsen et al, 2010;Philpotts & Ague, 2009;Takahashi & Scarfe, 1985).…”