Last week, a swarm of earthquakes happened below Long Valley's Mammoth Mountain, lasting a little over 12 hours. How do we decide whether or not it's "normal" activity for the area?

Every day, CalVO duty scientists check the number, intensity, and character of earthquakes happening at our volcanoes. It's actually quite common for an active volcano - meaning one that has eruptible magma somewhere in its plumbing - to experience small quakes and shakes. There's always something going on, whether it's magma gurgling, gases and fluids flowing through a hydrothermal system, or the bulk of the volcano settling and shifting. The trick is deciding which of these things are "background" and happen often without eruptions, and which are a sign of volcanic unrest.

To start, we look at which stations a signal appears on. Our first test is to make sure we're actually seeing what we think we're seeing, and not some odd environmental noise that looks suspicious. If a seismic signal only appears in one place, there's a good chance its source is very small and near the seismometer, or related to the equipment itself - for instance, a wobbly telemetry tower or a vibrating overpass.

Next we check the depth, location, and signal character. Recordings of tectonic earthquakes and volcanic earthquakes often look different - for instance, a seismologist might describe a long-period earthquake related to water or steam in a crack as "broad and wobbly," while an earthquake from rocks breaking would be "sharp and distinct." In addition, if we know the depth to certain features of a volcanic area - a hydrothermal system, an old caldera floor, a region of melt accumulation - then we can associate seismic activity with a particular source.

We also look at other data streams. Earthquakes can be the first heralds of volcanic activity, but they have to be correlated with other signals at some point. Magma pushing into the subsurface has to displace the rocks around it, and will cause deformation. The gases that come with it want to rise through the earth, and eventually have to come out somewhere. If we don't see either of those things, the source may be too deep and probably not a concern - or it isn't magmatic and won't be moving upward.

Finally, we watch and wait. Time will always tell with a volcano, because magma moves. Are the signals getting shallower and stronger or migrating through the subsurface, or are they staying in one place? Are they getting stronger and more numerous, or weaker and fewer? Are they tailing off at a rate that looks more like a tectonic earthquake (remember those exponential decays!), or coming in waves and bursts?

If a swarm were to become something unusual, the first step for the USGS is to talk with partners and issue an information statement about the activity. This is the first heads up we give before changing a volcano's alert level - kind of a "hey, we're keeping an eye on this just in case." And because it's entirely possible for the activity to stop and not come back, it gives us a buffer before we issue a new alert. And most of the time, it does - so we can go back to "normal" business!