Yellowstone Caldera Chronicles is a weekly column written by scientists and collaborators of the Yellowstone Volcano Observatory. This week's contribution is from Madison Myers, assistant professor of Igneous Processes at Montana State University, and Megan Saalfeld, PhD student at MSU.
What does a magma chamber look like? At first thought, many of us would imagine a large cavern in the crust filled with molten rock. While this has been the traditional model for a number of decades, geophysical imaging (using seismic waves to look for areas of increased heat and melt, which result in slower seismic wave speeds) of the regions below volcanic systems have never found evidence for this style of magma storage.
As more data became available from geochemical and geophysical studies, our idea of what a magma chamber looks like has continued to evolve. By understanding what a magma system looks like prior to an eruption, we can start to understand how volcanic reservoirs are built and sustained over time. This information can then be used to help us to monitor and prepare for future eruptions.
At a depth of several miles (many kilometers), magma has two main components: solid crystals and liquid melt (there may also be some gas, but this is typically minor). How these components are distributed can drastically change their stability in the crust, as well as what that geophysical image might look like. Many studies have proposed that magma is stored as a "mush zone" — a large, semi-rigid region that is composed of about 50-70 percent crystals and with smaller amounts of melt distributed within this crystalline framework. Researchers have shown that magma stored in these melt pockets within the mushier framework can merge over the decades to centuries prior to being erupted, presumably from a central vent. In other cases, melt bodies may not fully merge at depth, and instead could be erupted from multiple vents. This might explain why large bodies of melt have never been imaged, even though large eruptions have obviously occurred at places like Yellowstone.
This knowledge has important implications for how we monitor volcanic systems, especially when interpreting geophysical data, since it redefines what we might consider an active or eruptible magma reservoir. Additionally, this new model of magma storage as a network of smaller, disconnected pockets of melt, rather than a giant pocket of molten rock, is important for understanding how large volumes of magma can be stored stably in the crust for long periods of time.
One such example for these isolated melt pockets comes from the Huckleberry Ridge eruption — the largest of Yellowstone's three caldera-forming events. The Huckleberry Ridge eruption occurred 2.08 million years ago and produced 2,500 km3 of material (for comparison, Mount St. Helens produced only 1 km3), and dispersed ash that probably touched most of the United States in one way or another. Recent research has found subtle variations in the compositions of glass (which is basically frozen magma) and minerals produced during the earliest part of the eruption that suggest the magma was stored as four separate melt bodies. Because there is no evidence that these melt pockets mixed with each other, it is inferred that there were several active vents, each erupting magma from different melt pockets (instead of a single central vent).
After about 50 km3 of material had been erupted, the magma reservoir became underpressurized, meaning it was no longer able to support the weight of the overlying crust. At this point, caldera collapse (the roof of the magma system falling inward) began, similar to a piston cylinder dropping down and pushing the remainder of melt out of the crust. During this phase, not just four, but eight different melt pockets were tapped from vents around the perimeter of the caldera.
Overall, this model of complex magma storage, rather than a single magma chamber, can explain how large volcanic systems are built in areas without one massive magma reservoir. Fortunately, seismic surveys of the Yellowstone magma system do not show any evidence of substantial melt pockets. In fact, geophysical imaging shows that the mush zone consists of only about 5-15 percent melt, which indicates that a large eruption is not likely to happen anytime soon.