This is a question that gets asked regularly and in particular when there is a lot of earthquake activity going on. Surely this must “mean something”! Well, yes, but not necessarily something volcanic. In fact, most of it is not volcanic at all but tectonic as the North American and Eurasian Plates slowly separate, driven by the Mid-Atlantic Rift (MAR). Let us take a closer look at the Vatnajökull earthquakes for the preceding 48 hours!
As earthquakes occur, they are detected by the “SIL stödvar” network (don’t you just love Icelandic!). Then, based on the times and amplitudes with which the earthquake-shockwave arrives at several of these stations, an automated process, an algorithm, calculates the approximate parameters of the earthquake – time, location, depth, strength (magnitude) and quality (accuracy of the calculation).
One should always be very sceptical of any quake assigned a depth of 1.1 km, the default setting when the algorithm cannot decipher the seismic data, and those with a “quality” of less than 90%, especially those below about M2.5! Wait until the seismologist on duty has verified the calculation – which is signified by the quake being assigned a 99.0% “quality”!
Now, which earthquakes are of possible volcanological interest in the table above? The trio of earthquakes close together in both space and time at great depth ~6 km NW of Dreki (below Askja). “Why are those of interest? They’re only M1.0 and there were substantially larger quakes that day, well do I remember it!” Well, take a look at the picture below:
Earthquakes can and do occur at almost any depth from zero down to 150 km or so. The greater the depth, the more ductile or malleable the rock is due to increasing temperature, thus it takes much greater force to result in even a moderate earthquake. Also, magma sitting in a magma chamber at five or ten km depth does nothing except cool (and become even less prone to erupt) unless there is an influx of fresh, hot magma from below. Thus, earthquakes at depths above 15 km are most likely to be tectonic in origin. If they occur in a known zone of hydrothermal activity such as 0-3 km below most Icelandic volcanoes such as Katla or Grímsfjall, they are probably hydrothermal in origin, superheated water flashing into steam.
However, the closer an earthquake is to the mantle, the more likely it is to be associated with upwards movement of magma. If a rift opens up at the bottom of the crust, as is likely along the MAR, it results in a volume of space where the pressure suddenly falls. Now think about boiling water! At sea level, water boils at 100C. But at the top of Mount Everest it is only 71C because of the substantially lower air pressure. It is a similar thing with magma. If a rift opens up, a process called “decompression melt” occurs and more magma is generated thus widening the rift further which results in even more decompression melt of magma. This is why really deep Earthquakes are so significant as they could be the signal of fresh magma moving into the system and potentially cause an eruption.
At this point, we have to take the varying thickness of the Icelandic crust into account. The closer a cluster of earthquakes are to the bottom of the crust at that location, the more significant they are. A couple of months ago, there was a great swarm of earthquakes on the Reykjanes ridge close to Eldey and Eldeyadrangur. The crust there is no more than 10-12 km thick and the earthquakes ran from about 10-12 km and moved upwards to finally about 5 km below the surface – and over a large area too. From “privileged information from anonymous sources”, we knew that an eruption was seen as a 50 – 50 probability.
As you can see, in order to catch an eruption about to happen (as Carl did with Bardarbunga) requires some knowledge of what to look for and a lot of patience. But take heart! My first attempts were, in retrospect, rather embarrassing. Back in 2010 before the eruption at Fimvörduhals started, I was convinced I saw an eruption start on the glacier. At the time, I was completely unaware of the meteorological phenomenon whereby clouds can seemingly materialise out of thin air, thus when clouds were continuously generating at the same spot, I was utterly convinced that a subglacial eruption had started…