Weighing Volcanoes with Gravity

Chris -   The key to understanding the behaviour of a volcano must lie, partly at least, in understanding the plumbing system that brings magma from deep inside the earth up to the surface so that it can erupt.  One way to do this is to observe the changes in the gravity field that occur over a volcano.  And to explain more, we're joined by Dr. Hazel Rymer who is from the Open University.  Hazel, welcome.

Hazel -   Hello.

Chris -   Thank you for coming to join us on the Naked Scientists.  First of all, tell us how on earth do you use gravity to work out the plumbing of a volcano.

Hazel -   Well, good question.  It's not the most obvious thing to do, is it?  Well, we use a very expensive piece of kit called a gravity meter.  It looks a bit like an old car battery.  It's a box with knobs on.

Chris -   Sounds technical.

(c) NASA' alt='Gravity anomaly map' >Hazel -   Very technical and very unimpressive, but they cost several tens of thousands of pounds actually.  And what they are, basically, is a mass on the end of a spring inside them.  So, it's a bit like the scales that you might see in any market, where you stick two or three apples into the scale pan and then the spring will extend, and the dial will move, telling you how far the spring has extended, that would tell you the weight of the stuff in the pan.  What you're doing there is you're changing the mass, as in you're changing the number of apples in your pan, but you're keeping g, the acceleration due to gravity, constant or you're assuming it's constant.

What we do with the gravity meter is we keep our number of apples in our pan constant.  In fact, in this particular case, it's a tiny, tiny little mass, about a gram or so.  And what we're doing is, in a way, looking for how much the spring extends or doesn't extend in different places as we move our gravity meter around the volcano.  And the variable in this case is not mass as in number of apples, but it's little g, the acceleration due to gravity.  So, what we're doing, effectively, is seeing what the weight of our little scale pan inside our gravity meter is, as we move around the volcano.  All you have to do is put it into your backpack and wonder around across your volcano and set up a series of stations - it could be anything between 10 and 100 points on the volcano - and you measure gravity at them.

Chris -   Presumably, you use GPS so you know you're using the same or measuring the same spot each time...

Hazel -   Yes.  Well there are several ways of doing gravity actually and yes, you always do need to use GPS to locate yourself, but one way is simply to end up with a map of gravity.  So, just as you end up with a contour map, that would just tell you the different heights in a particular area, so the lines would represent equal heights and you know, if you have lots of them very close together, then the terrain is very steep.  That's a topographic map.  You can have a gravity map that tells you how gravity varies with space.  So for example, at the top of a volcano, you might find that there was what we call a gravity anomaly where gravity was particularly high or particularly low as the case may be, and you would see that by very close together isogals which are the contours related to gravity.  But the other thing is to see how gravity changes through time.

Chris -   That's what I was going to say because why should the gravity of a volcano change?  Presumably, it's gaining mass from somewhere and that presumably is magma rising into or exiting the volcano, and this will affect how much it weighs - to use a horrible term in a physics sense, but it will affect its gravitational attraction for your box.

Hazel -   Yes, exactly right.  There are a lot of other things that you need to check and of course, that's why you use GPS and other methods because if the volcano inflates or deflates -
Chris [Kilburn] was just talking about Campi Flegrei which is - well, sometimes it sinks down and sometimes it rises because it's a caldera in a state of unrest, but for example, as a caldera deflates as it goes down, you on the surface of the earth actually get closer to the centre of the earth and that means that gravity will increase.  So you have to be able to correct for that effect.  Having done that, then any net change in gravity that you measure must be due to subsurface mass changes, and in the case of an active volcano, you would probably interpret that in terms of magma movements.

Chris -   But getting back to the point you made which is the time one, many observations that are made of volcanoes are not made longitudinally.  People don't go back to the same spot year on year, on year, and then get an accurate record of how that thing is evolving.  Obviously, not on a geological timescale, but on a human time scale.  Is that something you've been able to do then because you've been doing this for - I was looking at your publications - it goes back 20 years!

Hazel -   I know.  I was.  It's just that I'm old!

Chris -   I wasn't going to say that!...

Hazel -   It is unusual to do it, yes.  Not very many people have the opportunity and the privilege to keep revisiting several volcanoes for a long period of time, and actually, you do have to look at several volcanoes because most of them don't do anything most of the time.

Chris -   Geologists must need to live longer.  You need some of those genes that make you live to be 100...

Hazel -   Absolutely right.  Yes, that would be useful, wouldn't it?  You still wouldn't live as long as a volcano unfortunately.

Chris -   Ideally not.  So you've been working in Iceland?  This thing, this blessed thing that I'm not going to try and say the name of it but perhaps you can for me, but what have you found there?

Hazel -   Okay.  Well, I've not been working on Eyjafjallajökull itself.  I've been working on a volcano slightly further north than that and it has the advantage of not being covered up by an ice cap at the moment, so it's much easier to measure.  And what I've been doing, first of all, I set up a gravity network on the volcano to look at the general structure of the volcano, so I had a gravity map as I described earlier.  So started with that and from that, and of course the work of lots and lots of other people, we were able to see what the general structure underneath the volcano is and where the magma chambers are.  So, as Chris was talking about Campi Flegrei, we think there are several pods of magma underneath the Campi Flegrei.

At Askja volcano -  which is the one I've been looking at, more or less in the centre of the country - we think that there's a chamber or a storage reservoir of magma at about 3 kilometres depth, and another one, a little bit further down.  So we work that out with among other things, the gravity surveying.  And then going back to each of those stations year after, after year, as you say, and making corrections for the amount the ground has been deforming, we've been able to see that over the years - well, for the first 17 years of the study, there was a net mass loss of the volcano.  In fact, what was going on was the ground was going down.  It was deflating since the last eruption in 1961 and it looked as though magma was draining away from the volcano and that seems to have been what was going on until about 2007.

After that, the seismic activity changed a little bit and there was a lot of seismic activity, quite shallow, to the north of Askja volcano, and it was unclear from seismic activity. You can tell that magma is moving, you can't necessarily tell which way it's going.  Our gravity measurements suggested that over 2008 and 2009 we've actually got an influx of magma coming back into the volcano.  So, the mass underneath it has increased which is quite exciting.

Chris -   Do you think that the activity we've seen this year in its partner, the one I can't say, do you think that's relevant then to this?  Do you think that one could be a symptom of the other, if you like?

Hazel -   Well never say never, but I don't think there has to be a connection.  The volcanoes are several 10s of kilometres apart from each other and of course, the plumbing system is linked at huge depth, but they have separate feeder systems, so I don't think that there's likely to be a connection.

Chris -   And just to finish this off, Hazel.  Does the data that you've generated here inform our understanding of how volcanoes behave in general?  In other words, does this tell us a bit more about what predictions we can make about their likely evolution over both human and geological timescales?

Hazel -   Well I hope so.  Otherwise, I don't why I've been doing this.

Chris -   You may have you've got negative results.  I don't know, you obviously haven't, but...

Hazel -   Well yes, I think it does, but again, as Chris was saying earlier, there are lots of different types of volcanoes and he talked about the sort of pointy peak volcanoes and the flat ones, and the ones that are basically a whole in the ground.  There's a huge variety of volcanoes and they all work in slightly different ways.  There are a lot of variables to do with the thickness of the lithosphere, the crust that their coming up through, the chemistry of the magma.  Sometimes it's very thick, sticky stuff, so it's harder for it to move around.  They're are a huge number of variables and you need to understand all of those variables to be able to put together a sensible picture of how an individual volcano is working.

Chris -   But you made a start and that's the most important thing.

Hazel -   I've had a go.

Chris -   Brilliant.  Thank you very much, Hazel Rymer from the Open University.