An interdisciplinary collaboration investigates one of earth science’s cryptic mysteries: the origin of deep earthquakes.
By Meghan Zulian, University of California, Davis, Temblor science writing extern (@meghan_zulian)
Citation: Zulian, M., 2021, A fluid’s journey into deep earth may explain deep quakes, Temblor, http://doi.org/10.32858/temblor.183
In 2017, seismologists at the Carnegie Institution for Science’s Earth and Planets Laboratory seminar were discussing the source of deep earthquakes — the very existence of which has long confused scientists. Geochemist and diamond petrologist Steven Shirey listened with interest. But when seismologists asserted that sudden pressure changes associated with mineral transformations in the mantle were likely the cause of deep earthquakes, and not fluids — because water could never make it greater than 300 kilometers (about 190 miles) below Earth’s surface, they reasoned — Shirey sat up and took notice. That’s not true, he thought. Shirey couldn’t explain how fluids got to the great depths at which these mysterious earthquakes tend to nucleate, but, he says, “diamonds are there, and they have to form from fluids — so there have to be fluids down there.”
After the seminar, Shirey, who’s a senior staff scientist at the Carnegie Institution for Science, sought out colleague and seismologist Lara Wagner to discuss how fluids might travel to such depths. Shirey and Wagner assembled a team of scientists, including geodynamicist Peter van Keken and experimental petrologist Michael Walter, both of Carnegie, plus diamond petrologist Graham Pearson of the University of Alberta. Their collaboration resulted in a new study published in AGU Advances, where the researchers present multiple lines of evidence demonstrating how descending oceanic plates carry fluids to the depths at which deep earthquakes nucleate.
Shirey’s and the diamond community’s logic about deeply derived fluids come from growth patterns in diamonds and the way most diamonds are carried volcanically to the surface via explosive magma punching its way through hundreds of kilometers of Earth’s mantle. These magmas, termed kimberlites, contain the diamonds as exotic crystals picked up from mantle rock hosts as the kimberlite passes through. The growth patterns in the diamonds preserve a prior history of fluid passage through the host as the diamond grew there. These diamonds can only form in the presence of fluids through sequential dissolution and precipitation. “They prove that water can make it as deep as these deep earthquakes and is moving freely at those depths,” says Shirey.
Although the presence of these diamonds indicates fluids, it does not prove the fluids initially came from Earth’s surface, and it does not prove that these diamonds originated at the specific depths relevant to deep earthquakes. So for those answers, the team analyzed the diamonds’ chemistry and minerals trapped in their crystal structure called inclusions.
Analyses revealed that the carbon isotopic composition of the diamond and the oxygen and iron isotope composition of some mineral inclusions could only come from seawater and materials found on the seafloor. Yet, other minerals trapped in the diamonds, like ringwoodite, breyite and bridgmanite, form hundreds of kilometers below Earth’s surface at depths of 300-700 kilometers (about 190-440 miles). The isotopic analysis of the diamonds and their mineral inclusions prove that these diamonds formed in the presence of fluids derived originally from Earth’s surface, and are now at the exact range of depths where deep earthquakes originate. But how did the fluids get down there?
Goldilocks slab paths
Scientists had long known that water could make it about 300 kilometers down with these subducting plates but did not think it could be deeper because they thought the temperatures were high enough that the minerals would release any trapped water before traveling deeper. To figure out how, where and why fluids could remain in a subducting slab, Wagner and van Keken modeled thermal profiles from 23 subduction zones. In the model, the 23 slab transects were constructed where the geometries, plate motions, convergence directions and velocities at each transect led to defined pressure-temperature paths with depth. When they then projected the seismicity onto pressure-temperature paths, the models revealed that some older, colder, descending plates could remain at low-enough temperatures and high-enough pressures to retain water in hydrated minerals until between 450 to 650 kilometers below Earth’s surface — the exact range of depths where these deep earthquakes nucleate.
In an opinion article written about this study, also published in AGU Advances, seismologist Douglas Weins at Washington University in St. Louis pointed out that the presence of fluids at depth and a path for them to get there does not prove they cause deep earthquakes. Minerals proposed as the fluid source (hydrous magnesium silicates) release volatiles at much higher temperatures than those predicted at deep earthquake origins, he wrote. According to Shirey, descending slabs could stall and warm at intermediate depths until they release fluids. However, Weins noted, most deep earthquakes originate in the core of slabs that move continuously.
Shirey says he hopes the results convince the seismological community to consider the presence and impact of fluids in their work. “It will also end up changing how diamond petrologists think too; diamond petrologists now understand how fluids survive to great depths at which diamonds can form.”
For some researchers, like doctoral candidate and deep earthquake researcher Becky Fildes of the University of California, Davis, she says the “whole” of collaborative studies like this paper is greater than the parts. “[With] a shallow earthquake you can go look at a fault scarp, look at materials, take materials,” she says, “but with these deep slab earthquakes, we can’t do that.” Multiple lines of evidence supporting the same story are important, she says, “when you’re looking at something that is 500 kilometers below the surface of the Earth.“
Rakovan, J., Gaillou, E., Post, J., Jaszczak, J., & Betts, J. (2014). Optically Sector-Zoned (Star) Diamonds from Zimbabwe. Rocks & Minerals, 89, 173–178. https://doi.org/10.1080/00357529.2014.842844
Shirey, S. B., Wagner, L. S., Walter, M. J., Pearson, D. G., & Keken, P. E. van. (2021). Slab Transport of Fluids to Deep Focus Earthquake Depths—Thermal Modeling Constraints and Evidence From Diamonds. AGU Advances, 2(2), e2020AV000304. https://doi.org/10.1029/2020AV000304
Smith, E. M., Ni, P., Shirey, S. B., Richardson, S. H., Wang, W., & Shahar, A. (2021). Heavy iron in large gem diamonds traces deep subduction of serpentinized ocean floor. Science Advances, 7(14), eabe9773. https://doi.org/10.1126/sciadv.abe9773
Wiens, D. A. (2021). Diamonds Hold Clues About the Cause of Deep Earthquakes. AGU Advances, 2(2), e2021AV000434. https://doi.org/10.1029/2021AV000434
- Hawaiʻi quake likely related to tectonic plate bending - October 19, 2021
- Opinion: Science communication on trial following White Island disaster - October 15, 2021
- Fluids and tiny minerals play a big role in subduction zones - October 12, 2021