This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. The word volcano conjures up an image of a peak with a crater at the top. We hear about them when dramatic explosions occur, but actually volcanoes come in many different shapes, and their eruptions may not be explosive at all. Furthermore, volcanism is not confined to the Earth. David Rothery is Professor of Planetary Geosciences at the Open University. He's an expert in the study of volcanic activity using satellite remote sensing, and he investigates volcanism throughout the solar system. David Rothery, welcome to Geology Bites.

David Rothery: Thank you, Oliver. It's very nice to talk with you.

Oliver Strimpel: I said that volcanoes make us think of mountains with sharp peaks and craters, but that's not always the case, is it?

David Rothery: Well, no. That's the picture-book idea of a volcano, and it's true in some cases, but they're not all steep-sided, and they're not all symmetrical. And some of the greatest volumes of volcanic products spread out horizontally without ever building up to form any kind of mountain.

Oliver Strimpel: Can you give us some examples of each of these different kinds of volcano?

David Rothery: OK. Your classic conical volcano is something like Mount Fuji in Japan, or many of the Andes or the Cascades volcanoes in the northwestern United States. But if you go somewhere like Hawaii, where what's erupted is a rock type called basalt, which is rather fluid, the volcanoes don't build up to anything like such a steep cone, although they can be very, very large conical edifices. And then elsewhere, even greater volumes of lava can produce flood basalts, like the Columbia River basalts in Washington State or the Deccan Traps in India, where flows have spread horizontally for hundreds of kilometers. These are not active today, thank goodness, because that would make a hell of a mess of the climate. But lava spreads a long, long way sideways if it's erupted violently enough and in sufficient volumes.

Oliver Strimpel: Is there a basic reason why some lavas are very sticky and make pointy mountains, and others are very runny and just create these very extensive flows?

David Rothery: There are two main factors. One is the stickiness, as you put it, or the viscosity, of the lava. Lava that doesn't flow very freely is richer in silica, or silicon dioxide, than the runnier, lower-viscosity kinds of lava and magma. Magma that comes pretty much directly from the Earth's mantle is a rock type we call basalt, and this is low viscosity and flows relatively freely. If this magma, on the way up, is stored somewhere and cools down slowly enough that some crystals can separate from it, what you've got left is richer in silica. When it oozes out at the surface, it's much more viscous, much more sticky; it doesn't flow so freely, and so it won't travel so far. So it will build up into a bigger, taller pile. But on top of that, as magma undergoes this fractional crystallization, any gases that were within the original magma are concentrated into the remaining part of the melt. And when the gas-rich magma nears the surface, the gas can come out of solution violently enough to drive an explosion. Instead of gooey lava oozing out, you get an explosion sending out a vertical jet of fragmental material, which will then fall back down on the mountain. The combination of explosive eruptions, producing large volumes of ash that settle on the mountain or cascade down in pyroclastic flows, with sticky, high-viscosity lavas is what builds up your classic Mount Fuji-type conical volcano. So it's a complex series of events that gives a picture-book kind of steep conical volcano.

Oliver Strimpel: As I mentioned, volcanism is not confined to the Earth. So where else do we have active volcanoes today?

David Rothery: It depends how you define volcanism. But if you want silicates, if you want rock erupted onto the surface of a body, the only other example for sure is Jupiter's innermost large moon, which is called Io. Io has dozens of volcanoes erupting all the time. You can see them with infrared telescopes from the Earth, but no one suspected it until the space probe Voyager 1 flew past, I think it was in 1979, and found these eruption plumes, because explosive eruptions are the easiest ones to spot. Plumes of fragmental material being thrown 200 kilometers above the surface were witnessed by Voyager, and several spacecraft since then have seen these. So on Io we've got lava flows on the surface and eruptions of an explosive nature into a vacuum. If you're willing to regard eruptions involving molten ice, which is water, but not necessarily just pure water, as an example of volcanism, you can go to Enceladus, a moon of Saturn, a much smaller body than Io, where there are plumes of water-ice crystals being jetted into space. Now, it's a matter of semantics whether you regard that as more akin to geysering than to proper volcanism. But certainly that brings us into the realms of cryovolcanism, because, as we might talk about later, mixtures of ices behave very much like mixtures of silicate minerals. So icy volcanism and silicate volcanism have many, many close parallels. So that's two places in the solar system, Io and Enceladus, where you can see volcanism active today.

Oliver Strimpel: So we have these two kinds of volcanism, silicate volcanism and then cryovolcanism, which involves ice. Are those similar to each other in some respects?

David Rothery: Yes. Let's talk about melting a mixture of silicate minerals. When rock inside the Earth is caused to melt, it's usually because of a change of pressure rather than because it's warmed up. But whatever process causes it to melt, it does not melt all at once, because different minerals melt at different temperatures. So you get this process of partial melting, and the melt that you get out of this mixture of crystals is richer in silicon than what's left behind. So the melt has a different composition from what you began with. And that process of partial melting is exactly what goes on when you melt ice, unless it's pure H2O, pure water ice. If you've got any kind of mixture of H2O and salts, or H2O and some other volatile substance, such as ammonia... The ammonia–water system is a perfect analog to silicate minerals. When you cause it to melt, the melt composition is different from either of the end-members, and it also melts at a lower temperature than either pure end-member would. So you get partial melting in mixtures of ices very, very closely analogous to partial melting in mixtures of silicates.

When we come to explosive eruptions, that's where the Enceladus type of cryovolcanism differs from what we see in an explosive silicate volcano on Earth, or indeed on Io. On Earth or Io, as your magma rises to the surface, there's a gas in solution. It could be water vapor, it could be carbon dioxide, it could be sulfur dioxide, or, probably, a mixture of all three. As the gas nears the surface, and therefore the confining pressure decreases, the gas comes out of solution and forms bubbles. If it forms enough bubbles quickly enough, that drives an explosion, breaks the magma apart into fragments, and jets them out of the volcanic vent. Somewhere like Enceladus, what we think we've got is not a partial melting process going on within its icy crust. We think the ice is floating on an internal ocean of liquid water. There's a global ocean under the ice on Enceladus, and other bodies in the solar system have these global oceans too. So you've already got a complete melt down there, and what happens to cause an explosive eruption is not something in solution in this melt, in this water. What happens is that if the ice cracks and the water is exposed to space, the vacuum of space means there's no confining pressure on the water, and the water begins to boil directly into the vacuum. So the water forms its own gas. It's not a gas dissolved in the water that comes out of solution; the water is turning to gas. So you expose some water to space in a crack in the icy crust, water vapor forms, and water is brought up to the surface by the force of these expanding bubbles. As soon as it gets to the surface, it freezes into ice crystals. So what is jetted into space at Enceladus is jets of ice fragments, propelled by the expansion of water vapor. And it's great, because the Cassini spacecraft was able to fly through those when it was orbiting Saturn, and detect other stuff there, like methane and tiny silica particles, and even atomic hydrogen and various organic molecules as well. So we've got all kinds of measures of what's going on in Enceladus's ocean, which is carried out to space, where we can sample it, by these expanding plumes. But these plumes are not that analogous to volcanic plumes on the Earth, A, because they're into a vacuum, and B, because of how you get the gas in the first place.

Oliver Strimpel: On Earth, the volcanism above subduction zones is a direct result of plate tectonics. That would create volcanoes like Fuji, for example. And the other kind of volcanism, such as in Hawaii, say, is caused by mantle convection. Do we think these processes are actually going on today on Io and Enceladus?

David Rothery: I would leave Enceladus aside. You've got a global ice layer floating on a global internal ocean, a continuous molten layer below. But on Io, it's probably rather analogous to the Hawaii situation, where we've got a mantle below which is warm, and uprising cells in the mantle, mantle plumes, if you like. We have to be careful about the word plume, because we use it both for stuff that's jetted into space and for pipe-like upwellings within a body. OK. So the internal mantle plumes in Io, or below Hawaii on the Earth, are bringing warm mantle upwards, which experiences lower pressure and begins to melt, and that feeds the volcanoes. So the Hawaii situation is rather like what's happening on Io, except Io has it in spades, all over the blooming globe. What Io lacks is subduction, which is where you get volcanoes like those in the Andes and in Japan, where one plate is being drawn down below another. The cause of the melting there is not a decrease in pressure as stuff is drawn upwards. What causes the melting is that the down-going plate has a lot of water: minerals are being hydrated on the ocean floor, and cracks are physically carrying water down. When the water gets driven off, as the plate goes down below the base of the overriding plate, the addition of water causes hydration melting and produces volcanoes above the subduction zone. Those are volcanoes which are going to be richer in silica, and their lava is the less runny kind of lava we spoke about before, of andesitic composition, which tends to be very sticky and more explosive and will build up to become conical mountains. That type of volcano we don't see anywhere else in the solar system, because we don't have subducting slabs carrying water downwards, which is the secret to getting subduction zone volcanism.

Oliver Strimpel: I suppose it's pretty amazing that we actually see volcanism going on somewhere other than the Earth today. But do we also see signs of past volcanism on other bodies in the solar system?

David Rothery: Oh, yeah, pretty much everywhere you look. When you look at the Moon with the naked eye and see the dark patches on it, the lunar maria, those are flood basalts, like the Columbia River basalts or the Deccan Traps we spoke about. Three and a half billion years ago, major impact basins on the Moon were flooded by lavas, and that's probably how the earliest evolved planetary crusts formed everywhere in the solar system. The whole of Mercury, where I'm working, was surfaced by that kind of volcanism, and probably the same happened on Mars as well, and probably also on Venus. But Venus has had volcanism which has gone on into the pretty recent past; it may still be active today. So on Venus, the very early, vast flood basalts have been overprinted by many, many generations of volcanism, and the globe has been resurfaced in patches many times, so the very ancient volcanism isn't there on Venus anymore.

And then we go to the moons of bodies like Jupiter, Saturn, Uranus and Neptune, and all the large ones are icy, except for Io, the innermost large moon of Jupiter, which is tidally heated and has lost any water it used to have, and has just got these hot silicates erupting. But the other large moons of Jupiter, Europa and Ganymede, and possibly Callisto, even further out, have zones of liquid water below the icy crust. Now, Europa's ocean is close enough to the surface that water can get squeezed out as slush to the surface when the ice cracks, and then the cracked ice crushes back together and kind of squeezes it through the cracks. On Ganymede, nothing like that's going on today, because the ocean is too deep inside. And then we've got smaller moons, like Enceladus at Saturn, where we've got the ice exposing the water to space when a crack opens, and it boils away and drives out these plumes. So we've got silicate volcanism in the past on all the terrestrial planets and the Moon, and today on Io. And we've got cryovolcanism, cryo from the Greek word for ice, going on in the past on many of the icy bodies, and on one, possibly two, today: that's Enceladus, and possibly Europa.

Oliver Strimpel: Why did volcanism apparently shut down on the other rocky planets, apart from Earth?

David Rothery: Well, that's a good question. You need heat inside the body to cause the convection. You're not applying heat to make the magma, to melt the rock and cause it to come out; you've got more subtle processes going on, like changes of pressure or changes of hydration. But it still needs to be warm inside and churning around, and the smaller bodies have lost more of their heat, because the smaller you are, the greater the ratio of your surface area to your volume. So the smaller bodies lose heat more quickly. Heat is being produced internally inside rocky bodies by radioactive decay, but if everywhere has the same composition, then the rate of radioactive heat production depends on your volume, not your surface area. So the smaller bodies have cooled down more. The exception is Io. Io is a moon of Jupiter, a little bit bigger than our own Moon, but it is kept warm by a different process. That's tidal heating: you've got a family of moons orbiting Jupiter in resonance, and repeated tidal energy is being stoked into the interiors of these bodies. So that's why Io is an exception. But it's easy to understand why Mars and Mercury have cooled down and are not volcanically active today.

Venus is almost, but not quite, as big and as massive as the Earth. It doesn't have plate tectonics like the Earth, because it's lost its water. That's probably why its rigid outer shell isn't mobile in the same way as it is on the Earth. So there's no subduction, and no subduction zone volcanoes. But we're not sure if Venus has finished its active volcanism; I very much doubt that it has. You have to look at Venus's surface with radar, because it's permanently cloudy, and there's only been one really good mission orbiting Venus, mapping the whole globe with radar. There are impact craters on the surface, but not as many as there would be if the surface were billions of years old. It is a relatively young surface on Venus, and almost everywhere you look is volcanic. There are shield volcanoes like Hawaii; there are some vast lava flows. So Venus has had many, many eruptions. We can't tell if any have been going on in the past century, or the past million years. But if I were to bet on it, there's certainly been active volcanism on Venus in the past million years, and it may be going on today. There are some indications: the amount of sulfur dioxide in Venus's atmosphere has fluctuated, and that's easy to explain if Venus is volcanically active. So it wouldn't surprise me if we've got active volcanism on Venus today, but we've not had a way to spot it happening yet.

Oliver Strimpel: I know you're especially interested in Mercury, and you mentioned briefly that it does also have a volcanic history. Tell us a bit more about that.

David Rothery: OK. Well, I'm involved in the BepiColombo mission, which is a European Space Agency and Japanese space agency mission now on its way to Mercury. It won't get there until December 2025, and it will take a few months before we get into our science orbit and start getting data from it. But there's been a NASA mission in orbit around Mercury, called MESSENGER, and that has convinced pretty much everybody who works on Mercury that it's got a very, very volcanic past. Pretty much the whole of its crust was built up by volcanism. I mentioned earlier the dark patches on the Moon, the lunar maria. Everywhere on Mercury began like the lunar maria. So we've got vast fields of flood basalts, which are a little bit peculiar, because they're very high in magnesium and low in iron, but they're certainly very runny basalts. But more recently, there have been explosive eruptions, which have punched holes through these basalts, so the most recent large-scale volcanism on Mercury has been explosive. We did not expect that on a planet like Mercury; it shouldn't be rich in volatile substances, as it's so close to the Sun. And yet the most recent volcanic evidence is of magmas nearing the surface carrying a big load of volatiles, anything that will turn to gas, and punching these holes in the surface. The stuff gets vented out into space, rather like the eruptions we can see on Io today, and just falls back to the surface on parabolic, ballistic trajectories, giving us these quite bright, red-colored patches of material. The biggest such example on Mercury is a feature called Nathair Facula. Facula is just the International Astronomical Union's descriptive term for a bright spot, and Nathair is the name of this one. It's about 250 kilometers in diameter, and in the center of it is a hole in the ground that's 3.5 kilometers deep and about 30 kilometers across. It's an enormous hole blasted in the surface, probably blasted through the surface by a series of eruptions rather than a single eruption. And there are smaller examples of these faculae all over the globe. It's part of the mystery of Mercury: it's clearly got volatiles, which we didn't expect on a body that close to the Sun. That's not why we're sending BepiColombo there, because BepiColombo was planned before we knew about this, but it's another good thing about BepiColombo. It will, hopefully, tell us what the volatiles are that have been driving these explosive eruptions. But we don't think those eruptions are active today.

Oliver Strimpel: If we assume that there really are volatiles, as it seems we have to, do we have any ideas as to what they're doing there, and how they got there?

David Rothery: No, we don't. But we know there are volatiles, because there are several lines of evidence which are pretty much independent. I've mentioned the explosive volcanic eruptions. We've got some compositional evidence, because we've measured sulfur with the X-ray spectrometer that MESSENGER had. Everywhere you look on Mercury, there's about two to four percent sulfur across the surface, probably not existing as pure sulfur but as sulfur compounds, such as calcium sulfide or magnesium sulfide, but a lot of sulfur nonetheless. There's a lot of potassium, a lot of chlorine and a lot of sodium. And then there are other patches on the surface which are active today, places called hollows, where the top 10 or 20 meters of the surface has just been stripped away. You've got steep-sided, flat-bottomed depressions, which is what are called hollows, and the material must have been volatile, because the stuff has just gone; it's been turned to vapor and lost to space. We don't know how, and we don't know what the volatile substance there is. So we don't know what volatile is lost to form the hollows, and we don't know what volatile is driving the explosive volcanic eruptions, but we do know there's a lot of, as I've said, sulfur, chlorine, sodium and potassium, those volatile elements. So we think there are a lot of volatile elements on Mercury.

Now, how do you do that on a body that's as dense as Mercury? Because Mercury is a dense rocky body; it's got a very large core and a comparatively thin rocky mantle and crust on top of that. So it probably began life with a lot more rock, and the rock was stripped away violently. You'd need a collision to do that, probably. But if you're colliding two bodies together and stripping away a lot of rock, you're going to lose volatiles as well. So that deepens the mystery of Mercury: how do you have a collision stripping the rock away and leaving volatiles behind? Well, perhaps the answer to that is that you start with very volatile-rich rock. So maybe this is now answering your question. Maybe Mercury began to form much further out from the Sun than we now see it. If Mercury began where Mars is, or a little bit further out than Mars, and then came inwards and careened off the early Venus, or one of the bodies which would shortly merge with something else to form the Earth or Venus, then this proto-Mercury could have lost most of its rock in this collision, and what was left carried on and found itself in an inner orbit close to the Sun. And the rock that was left, which was not very much, still had a fair bit of volatiles in it, because it began with lots of volatiles, having formed further from the Sun. Now, that's a very convoluted story, but it's actually the one that best fits the evidence at the moment. We'll have more confidence in it when we've measured more of the volatiles with BepiColombo in orbit around Mercury.

Oliver Strimpel: I just want to go back briefly to these incredible holes that you talked about. You said one of them is three and a half kilometers deep. Does it just have walls around it? I mean, how rigid is this rock?

David Rothery: It doesn't have a rampart. The big thing about these volcanoes on Mercury is that they're not mountains; they're holes in the ground. The internal slopes are about 20 to 30 degrees. If you stood on the brink, it wouldn't be a cliff, but it would still be a long, long way down to the bottom. Yes, there is bedrock exposed there. But if you were walking around on the surface of Mercury, you'd be leaving footprints everywhere. Most people listening to this will have seen pictures of Apollo astronauts walking around on the Moon leaving their boot prints, and what they're leaving their boot prints in is the regolith, the fragmental rocky material that's accumulated as a result of impacts of all sizes breaking the surface apart. This regolith forms on Mercury even faster than it does on the Moon, because Mercury is closer to the Sun and the impacts are faster. So you'd be walking around on quite a powdery surface on Mercury. But it is coherent: dust particles get welded together in a vacuum, and if there are some volatile substances which are somehow mobile through the regolith, they could stick it together as well. So you can form landscapes in this powdery material. Impact craters survive for billions of years, and steep-sided holes in the ground which are not impact craters also survive for up to billions of years.

Oliver Strimpel: Just to explain: you mentioned that the impacts will be stronger because Mercury is closer to the Sun. So I suppose when things fall in, the Sun is accelerating them, and it's deeper down the potential well, so the impacts are just going to be that much more violent. Is that right?

David Rothery: Well, essentially. Another way to look at it is that Mercury orbits the Sun with a faster orbital velocity than the Earth does, so it hits anything in its path at a faster speed.

Oliver Strimpel: Is there a particular puzzle that you're especially looking forward to answering when BepiColombo finally switches its instruments on?

David Rothery: Yeah. We're mapping the whole of Mercury at the moment with the images from MESSENGER, and the thing that's most perplexing to me is the hollows that we see there. I've mentioned these already, the steep-sided, flat-bottomed depressions, and we don't know what the heck is going on there. How are you turning stuff to vapor and letting it escape to space, and then turning it off? Because the hollows grow, and then they appear to stop, leaving a big cluster of hollows, and then next door there are no hollows. So what's going on? And if you're removing 10 or 20 meters of material, was it all volatile? Did you volatilize all of it? If not, what proportion is volatile? If it's half volatile and half just ordinary rock, and you've got a 20-meter-deep depression, the floor of it must now be another 20 meters of pure rock. So you've volatilized 40 meters of material, and the 20 meters at the bottom is pure residue, pure lag deposit. But how do you remove volatiles through several meters of lag? I mean, we don't know what is going on. Is it the heat from the Sun? Is it solar photons? Is it micrometeorites? Is it charged particles from the solar wind hitting the surface and breaking chemical bonds? We don't know. So we want to get compositional measurements inside a hollow and just outside an active hollow, and see what the difference is. And we'll have other instruments in space which might be able to sniff out what elements are coming off the hollows as well. So the hollow-forming process, for me, is the biggest mystery on Mercury, and that's not volcanic. I've veered away from volcanism and said it's hollows that we most need to understand.

Oliver Strimpel: For my last question, I want to ask you: if I were to put you in charge of selecting the next science mission for the world's space agencies, what would it be?

David Rothery: What I'd like to do is something which nobody else is going to do. We need a mission to Venus, and that's going to happen; I wouldn't change that. But if I could have an extra one, so I could find out something in my lifetime which I otherwise wouldn't, it would be to go to Miranda, which is a small moon of Uranus that Voyager 2 flew past in the 1980s. It saw less than half of Miranda. Miranda has got, oozing out of it in just a few places, very viscous, what look like cryovolcanic lava flows. We think it's probably what you get when you partially melt a mixture of H2O and ammonia: you get these viscous flows just oozing out. So how do these flows get out to the surface? Water ice is less dense than liquid water, so you wouldn't get a water flow rising to the surface of an icy body. But if you've got the right density contrast between water ice and ammonia ice and the melt that you get from them, the melt might be buoyant enough to get to the surface, especially if you've got some rock to weigh down the ice. Well, what's going on on the half of the globe of Miranda that we've never seen, because it was in darkness when Voyager flew by? It's a frustrating globe. It's got wonderful textures all over it. We should go back there to study the cryovolcanism on Miranda and understand how this whole weird globe is put together. Nobody's going to do that for the next 20, 30, 40 years, so I'm not going to see Miranda's other half. So if you could give me a mission I could snap my fingers and make happen, that's what I'd do.

Oliver Strimpel: David Rothery, thank you very much.

David Rothery: It was great talking to you, Oliver.

Oliver Strimpel: For more about Geology Bites, as well as pictures and diagrams that illustrate this podcast, you can go to geologybites.com.