Yellowstone Caldera Chronicles is a weekly column written by scientists and collaborators of the Yellowstone Volcano Observatory. This week's contribution is from David Shelly, seismologist with the U.S. Geological Survey.

Sources/Usage: Public Domain. View Media Details

Yellowstone, like many regions with hydrothermal activity, often exhibits earthquake swarms. But how do we define an earthquake swarm and distinguish it from other seismic activity? And what is the cause of such swarms?

Before we answer these questions, let's take a small step back. Many earthquakes follow a non-swarm pattern known as a mainshock-aftershock sequence. In its simplest form, this means that the largest earthquake occurs first in this sequence, followed by a series of smaller shocks, decaying over a time period ranging from weeks to decades. On average, the magnitude of the largest aftershock tends to be about one magnitude unit smaller than the mainshock.

However, not all earthquake sequences follow this pattern. Sometimes, the largest event might not be the first event in a sequence; instead it might occur in the middle. Sometimes sequences can have many earthquakes with magnitudes similar to the largest earthquake of the sequence. Sometimes sequences don't decay over time, but rather remain steady or even increase in their activity rates over periods of days, weeks, or even months. Sequences that don't fit a mainshock-aftershock pattern are typically considered swarms.

There is no precise definition of when a mainshock-aftershock sequence becomes a swarm. In reality, the distinction is not sharp. Earthquake sequences follow a whole range of behaviors from "very mainshock-aftershock" to "very unlike mainshock-aftershock." The "swarm" designation is typically applied when we observe relatively many earthquakes within a relatively small area, which just don't fit the pattern of a mainshock-aftershock sequence.

Now for the second, more important question: What, physically, causes seismic sequences to behave as swarms rather than mainshock-aftershock sequences? This question is still a subject of active research, both at Yellowstone and elsewhere, as it gets to the heart of our goal of understanding active processes deep in the subsurface. Based on past research, we understand that swarms probably indicate that an extra "ingredient" is involved where the earthquakes are happening—an ingredient that isn't as prevalent in mainshock-aftershock sequences. Sometimes this ingredient might be that a fault is slipping slowly, and small sticky patches are popping off and generating numerous small earthquakes. Sometimes, magma might be pushing up into the crust, opening up a pathway for itself by breaking the rock in front of it. But most often, swarms are probably caused by fluids (dominantly water) interacting with faults.

Unlike magma, which requires a relatively wide pathway to avoid freezing into solid rock, water can move within the subsurface through small cracks. Although this process is incredibly slow in most intact rocks, it speeds up dramatically when larger cracks are present. How do we get larger cracks deep in the Earth's subsurface? Why, earthquakes of course! Right after slipping in an earthquake, a fault tends to be much more permeable than when it started slipping. At the same time, fluids within faults, especially at high pressures, can reduce the effective clamping force on a fault, causing it to slip. Thus, we have the potential for a positive feedback loop, where earthquakes allow fluids to diffuse, which in turn generates more earthquakes! We sometimes see evidence for this process in earthquake locations that begin in a concentrated area of a fault and expand dramatically outward with time. The caveat is that this only happens if the faults are already quite close to failure. This is often true in active tectonic areas like Yellowstone, but swarms will die out as soon as they encounter areas where faults are less stressed.

We still have much to learn about earthquake swarms and their underlying physical processes. Although we have some understanding of a deep water source in a place like Yellowstone (water and gases are expected to be slowly released from underlying magma as it cools and crystallizes), the larger-scale water pathways, ultimately connecting from relatively deep magma storage to the surface, remain largely unknown. We also don't fully understand the fluid-chemical dynamics during earthquake swarms. One intriguing possibility is that swarms could sometimes be given an extra kick by gas bubbles that may form within faults during earthquakes. Similar to shaking a Coke bottle, these bubbles could ultimately increase the overall fluid pressure at depth and trigger more earthquakes. Answering these questions will require a continued commitment to detailed geophysical and geochemical observations at Yellowstone and elsewhere, combined with carefully laboratory and computer modeling studies. We can't wait to learn more!