Subduction zones, where one tectonic plate moves under another, are the locations of the Earth's largest earthquakes. The physical and chemical conditions along a fault separating two tectonic plates in a subduction zone are controlled largely by temperature; fault temperatures increase with depth below the surface of the Earth. Studies of friction between rocks suggest that earthquakes can be generated only where fault temperatures are between 150 and 350 C. These temperature limits have been combined with site-specific estimates of fault temperatures to outline potential areas of subduction zone earthquakes. These potential earthquake areas are used to estimate the ground-shaking and tsunami hazards for coastal regions overlying subduction zones (for example, Washington, Oregon, and northern California).
Most existing subduction zone temperature estimates do not include the thermal effects of water flowing through an aquifer in the ocean crust. This flowing water can move large amounts of heat from one part of a subduction zone to another, affecting fault zone temperatures. The proposed study will improve temperature estimates in subduction zones by accounting for the effects of fluid flow. Results of this research will be used to improve seismic hazard estimates for subduction zones (including the U.S. Pacific Northwest) - a direct societal benefit.
Accurate subduction zone thermal models are necessary to understand key metamorphic and seismogenic processes. A recent study of the Nankai margin (southern Japan) shows that previous thermal models had neglected a process that dramatically influences subduction zone temperatures - hydrothermal circulation within the basaltic basement aquifer of subducting crust. For the Nankai margin, hydrothermal circulation explains longstanding, large, enigmatic thermal anomalies and reduces seismogenic zone temperatures by up to 100 C relative to a case without hydrothermal circulation. This study will test the hypothesis that similar fluid circulation in subducting crust is an important control on subduction zone temperatures for four margins capable of producing M9+ earthquakes: Cascadia, Alaska, Chile, and Sumatra. We will develop thermal models for these subduction zones using a conductive proxy to simulate the effects of vigorous fluid circulation in an ocean crust aquifer [e.g., Spinelli and Wang, 2008; 2009]. The numerical models will be constrained by surface heat flux observations and the location of major slab alteration.
This study will result in improved thermal models for subduction zones. This will allow us to test the hypothesis that the seismogenic zone of the plate boundary fault extends from ~150-350 C. The coincidence of the seismogenic zone with temperatures of ~150-350 C has been demonstrated for the Nankai margin, where thermal models include the effects of fluid circulation. Determining if this relationship applies to numerous subduction zones will either advance the concept of a thermally defined seismogenic zone beyond conjecture or demonstrate the inability to delineate a seismogenic zone based on subduction zone temperatures. The proposed research has the potential to transform a number of avenues of subduction zone research, as the predictions of metamorphic reaction progress and interpretation of fault zone processes that followed from earlier thermal models that did not account for fluid circulation in subducting crust may need to be revisited.
This award was supported by the Geophysics Program and EPSCoR.
We have examined how seawater flowing through oceanic crust can affect temperatures within subduction zones. Changes in the physical properties of rocks as they are heated result in temperature-controlled changes in the behavior of fault zones. As a result, estimates of temperatures along the fault separating two tectonic plates in a subduction zone are useful for understanding earthquake behavior. The portion of these faults that are capable of generating earthquakes is likely between the temperatures of 150 and 350 ?C. In this study, we develop thermal models to estimate subduction zone temperatures, with particular interest in temperatures on the plate boundary fault. We find that water can carry heat from subducted oceanic crust seaward, where some of the heat can be lost to the ocean. This results in a cooler subduction zone than if the fluid flow were not occurring. This cooling of subduction zones by fluid flow is most important for systems with young oceanic lithosphere, which tends to be hotter than older oceanic lithosphere. In the Cascadia, southern Japan, southern Chile, Colombia/Ecuador, Mexico, and Solomon Islands subduction zones, fluid flow likely cools the plate boundary fault by at least 50 ?C and potentially up to 150 ?C. Therefore estimates of temperatures for these subduction zones should include the effects of fluid flow in the oceanic crust. For the Cascadia subduction zone in the U.S. Pacific Northwest and southwestern Canada, constraints on temperatures in the margin are most consistent with thermal models that include fluid flow in the oceanic crust. These models show that fluid circulation cools the subduction zone and makes the portion of the plate boundary fault that is between 150 and 350 ?C wider than if the fluid circulation were not active. As a result, the portion of the plate boundary fault that may be capable of generating earthquakes extends ~30-55 km farther landward than previously estimated. This wider seismogenic zone is consistent with recent models of the slow deformation occurring on the margin between large earthquakes.
