A new study demonstrates how increasing fluid pressure and crystal size affects subduction zone earthquakes.
By Davitia James, Temblor Earthquake News Extern (@davitiaa)
Citation: James, D., 2021, Fluids and tiny minerals play a big role in subduction zones, Temblor, http://doi.org/10.32858/temblor.213
The Earth’s surface is constantly moving as tectonic plates slide past, rip away from, and bump into each other. All this commotion commonly creates earthquakes at plate boundaries, but these seismic events are not evenly spaced, geographically speaking. Why do earthquakes happen more often along some parts of a plate boundary? And why do some places seem to escape large seismic events?
Geologists John Hooker of University of the Incarnate Word and Donald Fisher of Penn State University, honed into one area on the northern part of the Pacific Plate to look for answers. In their recently published research, they examined the Aleutian margin, a seismically active subduction zone off the coast of Alaska marked by a trail of volcanic islands.
The pair recreated the plate boundary in a numerical model and evaluated how fluids and tiny, pore-filling crystals, known as cement, could affect seismicity in a subduction zone. They found that changes in fluid pressure — linked to how easily fluids can flow through the tectonic boundary — modulate the magnitude, frequency and location of earthquakes. Where rocks have less cement, fluids can flow more easily, resulting in smaller quakes.
Subduction setting, mineral modeling
The Aleutian subduction zone extends across much of the northern end of the Pacific Plate where the crust of the Pacific Ocean dives under North America. The zone stretches from Alaska to the Russian Kamchatka Peninsula.
Previous work in the Aleutian subduction zone focused on rocks samples brought to the surface from depth via natural processes. In that work, Fisher and his colleagues discovered patterns in crystals that suggest that mineral precipitation occurs at roughly the same time as the earthquakes that created the cracks in the first place. This observation, says Hooker, suggested that the process of mineral precipitation, which involves fluids flowing through these zones, could play a role in the timing and magnitude of earthquakes.
In light of these findings, Hooker and Fisher decided to look at the ways fluids helped fuse rocks via cementation, a process in which fluids move within and between rocks, precipitating minerals in pore spaces. In the Aleutians, that cement is predominantly made of quartz crystals. As quartz cement forms, it reduces the space available for fluids to flow, increasing the pressure applied by fluids on the rock. Cementation also gradually seals flow pathways. This means fluids cannot flow as freely, which increases pressure along faults and reduces friction, making earthquakes more likely.
To understand how these processes are related, Hooker and Fisher added to an existing computer model that recreates what takes place below the surface. They used the Mineralization, Earthquake, and Fluid-Flow Integrated Simulator (MEFISTO) to explore how changing rates of cementation and fluid production during subduction affect the possibility of rupture. To do so, they simulated the physical and chemical processes preceding and during earthquakes, using observations from both field studies and samples. In the computer model, they divided the blocks of rock along the fault surface into a grid of kilometer-scale cells, each with its own default porosity (amount of pore space) and permeability (interconnectedness of pore space) based on rock type. By changing the rate of cementation, the available pore space changes. By changing the rate of fluid production, the fluid pressure changes. The computer model lets stress from fluid activity increase over time until one or multiple “blocks” move, resulting in an “earthquake.” After each simulation, they compared their model results to records of actual earthquakes that occurred at the Aleutian margin. They explored both decade- and century-long timescales to see if there were any patterns in earthquake activity over time, or at different geographic points along the subduction zone. As it turns out, cementation, fluid pressure, fluid flow and earthquakes, says Hooker, “are all dynamically interlinked.”
Fluid flow and sticky minerals influence earthquakes
Hooker and Fisher’s results support connections between cementation, the freeness of fluid flow and motion at the plate boundary. They found that although cementation affects fluid pressure, and both components affect earthquake occurrence, the process of growing new minerals works at longer time scales than the movement and build-up of fluids. In other words, short-term seismic events are usually tied to fluid activity because it works on a faster timescale, said Hooker.
Though cementation works at different time scales, changing its rate changed fluid properties because there is less space to store or transport fluids in the rock, which can cause earthquakes. Changing the rate of cementation also produced patterns in earthquake size and timing. The numerical model showed that clusters of earthquakes smaller than magnitude-7.0 occurred when cementation reduced up to 10 percent of pore space. Similar earthquake clusters are regularly observed in nature along the Aleutians. Simulated earthquake aftershock patterns matched natural ones, both in how quickly they occurred after the mainshock, and in their tendency to occur on the same part of the fault. More modeled aftershocks occurred with higher levels of fluid pressure in the system, resulting from either increasing rate of fluid production or increased cementation that reduced the space available for fluid flow.
Future, better modeling
Hooker and Fisher’s modeling simplifies a complicated environment, says Ake Fagereng, a geologist at Cardiff University who was not involved in the new study. Every aspect of subduction cannot be represented in the model, he says, but this study provides an important early step in understanding earthquake potential in subduction zones. The cementation and fluid processes discussed in this study happen too far below the surface to be directly observed, and some assumptions had to be made about fluids sources and production, he says. Future studies could refine the ranges of cementation and fluid production rates, for example, Fagereng suggests.
Regardless of the exact direction of future studies, the end goal is to improve understanding of earthquake hazard to protect lives and reduce damages in populated areas, Hooker says. Anticipating where an earthquake might strike is, he says, “notoriously difficult.”
Fisher, D.M., Hooker, J.N., and Oakley, D.O.S., 2019a, Numerical models for slip on the subduction interface motivated by field observations: Lithosphere, v. 11, p. 322–332, https://doi .org/10.1130/L1008.1.
Huang, J., Narkounskaia, G., and Turcotte, D.L., 1992, A cellular-automata, slider-block model for earthquakes: II. Demonstration of self-organized criticality for a 2-D system: Geophysical Journal International, v. 111, p. 259–269, https://doi .org/10.1111/ j.1365-246X.1992.tb00575.x