10 August 2012

Mini maars

Posted by Jessica Ball

A while back, I wrote about UB’s exciting new facility for experimental volcanology, which is part of our Center For Geohazards Studies. The facility itself isn’t anything like a big fancy laboratory – it’s out in the country and is mainly open space. But that’s a perfect setting for making holes in the ground, which (in a very basic sort of way) was the whole point of the most recent test. Volcanologists and engineers from UB, Italy, and New Zealand were all present for this explosive event:

The goal of this test was to try and simulate the volcanic eruptions that produce maars (although I must admit, blowing things up is pretty entertaining in and of itself). So what’s a maar?

A maar is a special type of volcanic edifice that forms because of the interaction of groundwater with magma (i.e., an explosion). A maar is what we might call a destructive feature; instead of building a cone, a maar crater cuts into the existing geologic units and makes a diatreme (a cone-shaped vent blasted into existing layers of rock). The diatremes are often filled with breccias from the explosions which formed them, and may be surrounded by rings of tephra. Some classic examples are the Lunar Crater maar in Nevada and the Lago Albano maar in the Colli Albani, Italy. But one of the main questions that volcanologists have about maars is how this crater is excavated. Currently, there are a few competing ideas about this. One suggests that the hydromagmatic explosions occur at deeper and deeper levels as a crater is excavated and shallow groundwater is used up – a progressively deepening diatreme and a progressively widening crater, called the incremental growth model.

The incremental growth model of maar diatreme formation, after Lorenz 1986, from Valentine et al. 2011

Alternatively, in the major-explosion dominated model, the crater dimensions of a maar are directly related to the most powerful hydromagmatic explosion that occurs during maar formation. The implications of this model are that the scale of the maar is then directly related to the energy released in the eruption.

The major-explosion dominated model (figure from Valentine et al. 2011).

This experiment was a preliminary setup, meant to see if each it was feasible to simulate each of these situations in the first place; later experiments will work on testing the ideas more rigorously. The experimental setup involved three 12 x 12 foot test pits (engineers were involved, so some of the measuring happened in English rather than metric units), filled with layers of gravel, crushed limestone and recycled asphalt. Explosive charges were used to simulate eruptions, but the number and timing of the charges varied with each experiment.

For the first explosion, a roughly one-pound charge was buried at about half a meter into a pit with just three layers of material (mentioned above), and the charge was set off in one go – trying to reproduce the major-explosion dominated model. (The following photos and videos appear courtesy of Sarah Ogburn – please don’t re-use them anywhere else!)

 

The first explosion pit (photo courtesy of Sarah Ogburn)

Taking measurements at the first test pit (photo courtesy of Sarah Ogburn)

The second experiment was actually conducted in three parts, with each explosion taking about a third of the charge of the initial experiment. The idea here was to try and reproduce the incremental growth model, so a new charge was set into the pit following the first explosion, and again after the second. The charge was set at the same depth each time (which in real life would be equivalent to a water source that replenishes itself, or doesn’t experience drawdown for some other reason).

After the first of three explosions (photo courtesy of Sarah Ogburn)

The second of three (photo courtesy of Sarah Ogburn)

After the final explosion (photo courtesy of Sarah Ogburn)

The third experiment was similar to the second, only with four layers of material, and an explosive set at a deeper level each time (simulating a water source that’s being depleted by the magma-water interaction). During and between each experiment, of course, there was a lot of data collection going on – measurements were made of things like tephra dispersal and crater dimensions after the explosions, and high-speed and thermal cameras were running while the ‘eruptions’ were taking place.

Taking GPS measurements of the first explosion crater (photo courtesy of Sarah Ogburn)

The bunker at the test site, with digital and thermal cameras and the GPS total station (photo courtesy of Sarah Ogburn)

Collecting tephra from the first explosion (photo courtesy of Sarah Ogburn)

The experiments were preliminary, but as Dr. Valentine – the head of the Center For Geohazards Studies at UB – mentioned in the press release, they were very promising for future tests. (If you read through the whole release, you may have seen the mention about diamonds. The connection is a little tenuous as it relates to the experiments, but diatremes, such as the ones formed in maar eruptions, are where we find kimberlites. Kimberlites are a mantle-derived volcanic rock that sometimes contain diamonds, and their surface expression is thought to be similar to that of a maar – so by studying the way diatremes are formed in maar eruption, we may be able to derive information that makes it easier to locate kimberlite pipes.)

My contribution to these experiments was small (in fact, I only helped fill in the pits – I missed the explosions because I was doing outreach back on campus), but it’s exciting to see that UB volcanologists are teaming up with scientists from outside the department, and around the world, to work on these kinds of large-scale investigations!

For further discussions of maar formation (and the Lunar Crater maar), see Valentine et al. 2011, Models of maar volcanoes, Lunar Crater (Nevada, USA), Bulletin of Volcanology, v. 73, p. 753-765. doi:10.1007/s00445-011-0451-6

Update – There’s been some additional coverage of the experiments in other media outlets – check out the following links (we’ll forgive the trouble they’re having with the plural form of maar):

‘Maars’ Volcanoes: Research Seeks to Understand Strange Eruptions (LiveScience)

How to understand a maars volcano? Blow something up (MSNBC)