This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. The first thing that comes to mind when we think of volcanic eruptions is lava and perhaps plumes of ash. But volcanoes also emit gas. In fact, some emit a lot more gas than anything else. Where does this gas come from? What can it tell us about the Earth’s interior? What effect does it have on the Earth’s atmosphere? Marie Edmonds is Professor of Volcanology and Petrology at the University of Cambridge. She observes volcanic gas in the field and develops frameworks to understand the physical processes that generate the gas and subsequently recycle it into the Earth. Marie Edmonds, welcome to Geology Bites.

Marie Edmonds: Thank you, Oliver. I’m delighted to be here.

Oliver Strimpel: Do all volcanoes emit gas when they erupt?

Marie Edmonds: Yes, they do. All volcanoes emit gases when they erupt, and actually many volcanoes emit gases when they’re not erupting as well. Volcanic gases are made up of water and carbon dioxide, but also sulfur dioxide, chlorine and fluorine, trace metals, and other minor nitrogen- and carbon-bearing species. Fortunately, all volcanoes emit gas.

Oliver Strimpel: It seems as if you have a great deal of choice then as to where you go to make your field observations. Which sites have you concentrated on in your work?

Marie Edmonds: Yes, I’ve worked on a number of different volcanoes in the past. In particular, I’ve spent a great deal of time working in the Caribbean, on the Soufrière Hills eruption in Montserrat, and also working in Hawaii on Kīlauea Volcano. And most recently, I’ve been working on volcanic systems in Ethiopia along the East African Rift, in Papua New Guinea, Yasur volcano in Vanuatu, and also Etna in Italy, among others.

Oliver Strimpel: Gosh, that’s quite a wide range, but why did you pick those particular sites?

Marie Edmonds: Well, some of these were opportunistic. I had the opportunity to do a lot of fieldwork in Montserrat in the Caribbean during the 1995 to 2011 eruption. I was incredibly lucky. These volcanoes represent a very common type of volcano present in global island arcs, that’s subduction zones, where one plate subducts beneath another plate. And these are really hazardous volcanoes, because they’re water-rich and they erupt highly viscous magmas, and that makes them prone to explosive activity. So these are an important class of volcanoes to understand.

And then I worked with the US Geological Survey for a few years in Hawaii, and this experience introduced me to Hawaiian volcanoes. And these are, of course, entirely different. Hawaiian lavas are voluminous, low viscosity. They can form lava lakes, with very high gas fluxes. Other field areas we chose because they have potential for answering specific questions. So we’ve recently been working, for example, on trying to understand Strombolian volcanic activity. This is where you have small explosions caused by an accumulation of gas beneath a viscous plug in the conduit. So we chose Yasur volcano in Vanuatu as the type example of this sort of eruption. And the Ethiopian volcanoes were part of a much larger consortium I was part of, to understand the links between tectonics (that’s the movement of the plates) and the magmatism and the volcanism during continental rifting.

Oliver Strimpel: Can you tell us a bit about the volcanism and the accompanying gas emissions of continental rifting volcanism? It’s much less well known, really, than the explosive subduction zone type volcanism and the mid-ocean island, Hawaiian-type volcanism.

Marie Edmonds: Well, what we found is that there’s a vast amount of carbon dioxide outgassing from continental rifts through faults and fractures. And this carbon dioxide is derived from magmas that are stored deep in the crust. So magmas are rising up from the mantle as the plates are rifted apart, and then these magmas just pond in the crust. And we made detailed studies of the chemistry of these magmas, and in particular the volatiles that they carry. They’re carrying large amounts of carbon, and other volatiles as well, like water and chlorine and fluorine. And these volatiles exsolve from the magma, form a gas, and then this gas migrates up to the surface, where it can outgas to the atmosphere. Or sometimes, in particular in the East African Rift, it gets dissolved in these alkaline soda lakes along the rift, making them extremely carbonate-rich. And it’s thought now, actually, that the continental rift system is one of the most important volcanic environments for CO2 release to the atmosphere. And this probably was particularly important in the geological past, during times when the supercontinents rifted apart.

Oliver Strimpel: Where is that happening today, apart from East Africa?

Marie Edmonds: Major rifts exist in eastern China. There are continental rifts in New Zealand and in Antarctica, as well as the very large East African Rift system.

Oliver Strimpel: When you go to these active volcanic sites, you can often smell the pungent sulfur dioxide and the rotten eggs from hydrogen sulfide, but you and your students go into the field and actually make proper measurements of the composition and volume of gas emitted when volcanoes erupt. How do you do that?

Marie Edmonds: Well, we have a large range of different sorts of instruments to measure volcanic gases. A big one is spectrometry, so we have spectrometers of various different kinds. For example, we use a spectrometer that works in the ultraviolet region to measure sulfur dioxide. Now, sulfur dioxide is one we measure routinely across all volcanoes, because it has a very strong absorption signature in the UV, and it’s largely absent in the background atmosphere. So we point the spectrometer at the blue sky to use that as a background, and then we traverse a two-dimensional section of the gas plume downwind. We can quantify the mass of SO2 and then multiply that by the plume speed to work out how much SO2 is being emitted by the volcano, in tons per day. And a standard sort of SO2 flux might be several hundred to several thousand tons per day.

We also use infrared spectrometers to measure the concentrations of the other gas species. For some of these, we actually use a pump to suck the gas into chambers inside the instrument, and we use lasers to generate absorption spectra. And some of these instruments we use from the ground, or we use them in helicopters or fixed-wing aircraft, or even mounted on drones; that’s something we’re doing much more of now. Another aspect we’re interested in is metal degassing from volcanoes. So metals like copper, zinc and lead actually exist as high-temperature gases emitted from volcanoes, and they form aerosols when the plume mixes with the air. So again, we use a pump to pass the plume through filter papers of different sizes, and this traps the particles, and then we analyze them back in the laboratory.

Oliver Strimpel: Do these gases exist deep down in the mantle, or do they come out of the magma as the magma makes its way to the surface and the pressure drops?

Marie Edmonds: The latter. So magmas contain lots of dissolved volatile species. That means species that are actually bound in with the liquid rock when they’re generated in the deep mantle of the Earth, that’s more than 30 to 70 kilometers beneath the Earth’s surface. And as magmas make their way up to the Earth’s surface, these volatiles outgas from the magma, just like the bubbles that come out of champagne when you pop the cork. Essentially, their solubility decreases with pressure, so the decreasing pressure as the magmas ascend to the surface forces them out of the liquid and into the bubbles. And magma can hold a few weight percent of water and carbon dioxide, and that actually generates a huge amount of gas once the magma gets to the Earth’s surface. And don’t forget, gases also expand as pressure decreases; that’s Boyle’s law. So magmas foam and vesiculate as they approach the surface, and that’s actually what drives volcanic eruptions. Conservation of mass means that as the magma density goes down due to the vesiculation, the magma accelerates up towards the surface, and so the gases are quite literally driving the eruptions.

Oliver Strimpel: How do we know the depth of the origin of these various gases that we measure?

Marie Edmonds: Well, we can reconstruct the behavior of these gas species principally by conducting experiments on natural magmas under conditions of high temperature and pressure. Now, these aren’t easy experiments to do, but we find that different gases have different solubilities, and that’s essentially how much the magma can hold for a particular pressure and magma composition. So we know, for instance, that carbon is far less soluble in magmas than sulfur, so it comes out of the liquid phase and into the gas phase, i.e., it exsolves, much deeper in the crust than sulfur. So one thing we do is measure the ratio between these two species in the gases at the surface. As the magmas ascend, the gases change from being rich in carbon and relatively low in sulfur to very low in carbon and relatively rich in sulfur. So this turns out to be a very useful volcano monitoring tool, and it can help us in forecasting eruptions.

Oliver Strimpel: Is that because we can track the ascent of the magma as it approaches the surface?

Marie Edmonds: Yes, that’s right. A rising magma will outgas progressively more sulfur-rich gas. So periods between eruptions may involve either no gas emissions, or perhaps CO2 emissions from deeply stored magmas. But periods of unrest, however, which may precede eruptions, may be recognized by a changing composition of the gas, so the gas becomes more SO2-rich as magma approaches the surface. So what we think is happening is that the gas is escaping the magma and reaching the surface before the magma. So recognizing these chemical signals at the surface can provide an early warning of an eruption, and that may be weeks or months or days.

Oliver Strimpel: You mentioned that we can also find metals in the volcanic gas. I presume they’re not in pure metallic form, but what can we learn by measuring the abundance of metals in volcanic gas?

Marie Edmonds: Yeah, this is an interesting area, one that we’re spending a lot of time thinking about at the moment in my research group at Cambridge. So volcanic gases are actually a snapshot of the sorts of water-bearing fluids that ultimately form metal deposits beneath volcanoes, and you may know these as ore deposits. So in particular, we know that copper, for example, is transported as a chloride in fluids generated by the degassing of magma. And somewhat surprisingly, we found that volcanoes like Etna in Italy, which is erupting now, are actually outgassing many tons of copper into the atmosphere every day, which probably means an ore deposit is not forming beneath this volcano. But understanding the nature of these fluids brings us one step closer to understanding how in the future we might find better ways to source copper for the switch to electricity-focused green energy sources, for example.

Oliver Strimpel: You said that because Etna is outgassing many tons of copper a day, it’s probably not forming an ore. Is that because all the copper is coming out and there’s none left to form the ore?

Marie Edmonds: Well, we think hydrothermal ore deposits form beneath volcanoes when the metals, like copper and gold and so on, are degassed from magmas, and then they become sequestered away in a heavy brine, that’s a salty fluid that accumulates at depth, and then the ore-forming minerals are formed from this brine directly through precipitation. Now, in active volcanoes, this fluid, which contains the metals, is instead brought straight up to the surface by the magma and then outgassed to the atmosphere instead of accumulating at depth. So there is a school of thought that suggests that volcanism and ore deposit formation are mutually exclusive. The idea is that eruptions blow out all of the copper and gold to the atmosphere instead of allowing these fluids to cool and precipitate out the metals in the subsurface.

Oliver Strimpel: Apart from measuring the ratios of different gas concentrations to figure out the depth at which the gas was exsolved, are there other measurements that can reveal where the material forming that gas actually came from in the first place?

Marie Edmonds: Yes, we can also measure the isotopic composition of the gases. So for example, the two main isotopes of carbon in gases are carbon-12 and carbon-13. Mantle carbon has a particular ratio between the two. Now, we express isotopic composition as a ratio of carbon-13 to carbon-12, and we compare it to a standard, and we express it in parts per thousand, or per mil. So mantle carbon has a composition of minus 7 per mil, and volcanoes like Hawaii and Iceland outgas carbon with an isotopic signature very close to that. However, volcanoes in subduction zones, where one plate is sliding beneath another, outgas carbon that’s much heavier, i.e., has a composition much richer in carbon-13. In fact, it’s almost identical to the carbon isotopic composition of the shells of marine creatures that form limestone on the sea floor. And that tells us that the carbon that volcanoes are outgassing in subduction zones is actually recycled carbon. It’s ultimately derived from the breakdown of these limestones once they’re plunged down into the mantle and heated up. The carbon is then incorporated into magma and comes straight back up towards the surface.

Now, we do see some variation. So for example, subduction zone volcanoes at high latitudes don’t tend to show this heavy carbon signature, and that’s because not much limestone is being formed in the colder parts of the ocean at those high latitudes. Instead, the carbon there is mainly in the form of organic carbon, and that’s formed by photosynthesizing organisms in the oceans, and therefore that carbon is very light, i.e., poor in carbon-13. And that’s reflected in the carbon isotopic composition of the gases.

Oliver Strimpel: Okay, so there are three different isotopic signatures for carbon. There’s the mantle signature, and then two organically derived ones, one from shells and one from photosynthesis and microorganisms. Can you just explain what the difference is in the fractionation processes that go on in the latter two?

Marie Edmonds: The organisms that take in carbon to form their shells are really just taking in the carbon from seawater, and so the isotopic composition of their shells reflects the isotopic composition of the seawater. Whereas with organic matter, carbon that’s taken out of the atmosphere by plants and living things through photosynthesis, this is an entirely different process, and that really fractionates the carbon isotopes very strongly. So plants, for example, take up light carbon very strongly, and so you get a very light isotopic composition in organic carbon. And both of these carbon reservoirs are present on the ocean floor in piles of sediments. Now, the carbon isotopic composition of the mantle is largely derived from a combination of surface material being subducted back into the mantle, but also the primordial carbon that’s essentially derived from chondritic material, meteoritic material, or planetesimals that came together to form the planet. That’s where the mantle carbon isotopic signature comes from.

Oliver Strimpel: Can you say a little bit more about the mechanism of carbon recycling?

Marie Edmonds: Yes. So the volcanoes are outgassing carbon into the atmosphere, and they do that at quite high rates, although the carbon flux from volcanoes is still around a hundredth of the flux of carbon from the burning of fossil fuels, from human-related activities, from land use changes and so on. So compared to the anthropogenic flux, the volcanic carbon flux is still very small. But in geological terms, volcanoes are the main source of carbon into the atmosphere. And then carbon is removed from the surface environment by acid rain weathering of mountains. Eventually that carbon ends up in the ocean. Limestones are formed from the shells of creatures that take up that carbon in the ocean, forming thick sequences of limestone on the sea floor. And then eventually that plate is subducted. The ultimate fate of all of the sea floor, of course, is subduction and return to the deep mantle. And that’s really what makes our planet unique: we have this recycling mechanism, so the atmosphere can’t get overloaded with carbon. On very long time scales, it gets returned into the mantle. The slab gets heated up, and the carbon gets removed from the slab. Some of it gets returned to the deep mantle, and some of it gets recycled back up through volcanoes.

Oliver Strimpel: Can you combine your observations of volcanic gas with that of the accompanying magma to learn more about what’s going on in the mantle?

Marie Edmonds: Yes. We can study the compositions of melt inclusions, and we do this a lot in my group at Cambridge. These melt inclusions are tiny increments of melt trapped inside crystals. And how this works is that the crystals act as pressure vessels, keeping the melts at high pressure and the volatiles in solution. And so we can measure the concentration of volatiles in the melt before they degas, before they experience these low pressures. And so by measuring the amount of carbon in rare, entirely undegassed samples (we find these on the sea floor and in various deposits from Iceland), we can work out how much carbon might be stored in the mantle. And it turns out that the amount of carbon in the atmosphere and the oceans is much, much less than 1% of the total carbon in our planet. In fact, almost 10% lies in the solid mantle, and a whopping 90% might in fact lie in the core of the Earth, which is effectively isolated as a reservoir.

Oliver Strimpel: Gosh, that’s surprising. I had no idea that you could get carbon existing inside molten iron, I suppose it is.

Marie Edmonds: Yeah. Well, carbon in fact may be just one of a few light elements that exist in the core. The solubility of carbon and sulfur and oxygen is actually remarkably high in these iron alloys that we have in the core of the planet.

Oliver Strimpel: Coming back to one of your field locations, there’s a lot of seismic and even eruptive activity in Iceland at the moment. Are you performing measurements there, and what are you hoping to learn from them?

Marie Edmonds: Yes, that’s right. There’s an ongoing eruption on the Reykjanes Peninsula, at Fagradalsfjall, which began on the 21st of March. The fissure was originally 500 to 700 meters long, producing basaltic lava. It has now built up a row of cones that are feeding an ever-growing lava flow field, and it’s actually southwest Iceland’s first eruption in 800 years. It’s looking like a fairly small eruption so far, but it was preceded by six months or so of small earthquakes before it began, and that allowed the Icelandic scientists to forecast the eruption remarkably well and accurately. My colleague Evgenia Ilyinskaya of the University of Leeds is doing fieldwork there at the moment, measuring gas flux and collecting samples to measure trace metals. And we’re hoping to find out whether this eruption is producing gases of a similar composition to Icelandic eruptions that have happened in the last 10 years, for example the Holuhraun eruption, which we studied a few years ago. Now, what seems to be quite interesting about the current eruption is that the magmas are very primitive. That means they’re very rich in magnesium, so they haven’t undergone much processing in the crust before they’re erupted, and so that might have an implication for the metal emissions. And we’re also interested in how toxic the plume may be, and how it may impact humans, livestock and the environment.

Oliver Strimpel: The other active volcano you mentioned is Mount Etna, which I suppose is very different, because that’s a subduction zone related volcano, isn’t it?

Marie Edmonds: Yeah, that’s right. So those are much more water-rich magmas, and the water in the magma means that you have a slightly different eruptive style at Etna as well. You have a lot more ash-rich eruptions, more explosive types of eruptions. This one in Iceland is very water-poor. It’s not erupting underneath the ice cap, which can cause explosive activity, so it’s a very effusive eruption. And in fact the metal signatures for Etna and Iceland are quite different, related to that lack of a subduction component in the Iceland scenario.

Oliver Strimpel: For my last question, I want to ask you: if you had access to unlimited funds for your research, how would you use them?

Marie Edmonds: Well, that’s a tough question. There are many ways I’d like to use those kinds of funds, but I think, fundamentally, there’s a great need to improve monitoring and hazard awareness in poorly resourced areas, developing countries for example, to mitigate risk and gain understanding of volcanic hazards. Of course, many of the most explosive and dangerous volcanoes on our planet occur in subduction zones, and many of those occur in regions that are very difficult to access. They don’t have established volcano observatories with the funds to establish monitoring networks on the volcanoes, and that puts local communities at risk. The hazards are often concentrated where the populations are, just like for earthquakes, so urbanization is an increasing problem in volcanic areas. And we’re also becoming much more aware of how volcanoes impact our environment, for example air quality, water resources and so on.

Oliver Strimpel: Marie Edmonds, thank you very much.

Marie Edmonds: Thank you, Oliver.

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