Twenty-eight DCO members came together from 29 April  –  4 May, 2018 at the Carnegie Institution for Science in Washington, DC to calculate a new estimate of global carbon dioxide (CO2) degassing from large volcanic emitters, small volcanic sources and diffuse degassing from volcanic regions. The synthesis of massive amounts of data was successfully tackled through a hands-on approach. Science talks were interspersed with breakout sessions, followed by more of the same.  “It was the most productive workshop I have ever attended,” said Terry Plank, DCO Executive Committee member and Reservoirs and Fluxes Science Community member (Columbia University, USA), “and should serve as a model for others to come.”

The DECADE synthesis workshop group attendees came prepared with a wealth of available volcanic emissions data they used to create a new global estimate.  Their work was exhaustive, with some of the highlights provided below.

The attendees evaluated emissions from subaerial volcanoes with active gas plumes to produce an updated and improved estimate of global SO2 flux. This quantity was then combined with their best present knowledge of C/S ratios in the plumes of those volcanoes to derive a corresponding emission of volcanogenic carbon.

They accounted for different types of emitters, including passively degassing volcanoes, explosive eruptions, and effusive eruptions and distinguished between arc and non-arc volcanic sources. They compiled data covering 11 years from 2005 to 2015, and used information from long-term monitoring from space (mostly OMI satellite) and ground (mostly NOVAC network), as well as short-term campaign data and reports from the literature. The group also identified the need for further comparisons between satellite- and ground-based flux observations and the lack of C/S data, in particular for large eruptions.

To improve estimates from small volcanic sources, they assembled a new compilation of worldwide data from more than 40 volcanoes that emit small CO2 plumes and carefully selected appropriate volcanoes to include in the extrapolation.  The group then individually reviewed more than 750 volcanoes from across the globe that could potentially host small plumes and categorized their emissions as ‘magmatic’, ‘hydrothermal’, or ‘none’.  Said Tobias Fischer (University of New Mexico, USA), one of the initial workshop organizers, “When complete, this analysis will be the most rigorous and transparent estimate of global CO2 emissions from small volcanic sources yet determined.”

Attendees also delivered the first estimate of global carbon dioxide (CO2) degassing from diffuse degassing sources of volcanoes based on the published data reported in MaGa, a recent catalogue of diffuse gas emissions around the world, while also addressing uncertainties of the data.  Other attendees analyzed subduction data that were providing insights into volatile cycling on both regional and global scales, while others considered what could be learned from the rock record.

Overall, the results from this workshop will provide new and more rigorously constrained global deep carbon emission estimates, new insights into the fate of subducted carbon and new methods for estimating volcanic CO2 fluxes through time using petrologic parameters. The workshop also highlighted the need for continued multi-disciplinary efforts in the area of volcanic and tectonic degassing to advance understanding of the transfer of volatiles between Earth’s reservoirs.

A joint study between Carnegie and the Woods Hole Oceanographic Institution has determined that the average temperature of Earth’s mantle beneath ocean basins was about 110 degrees Fahrenheit (60 Celsius) higher than previously thought, due to water present in deep minerals. The results are published in Science.

Earth’s mantle, the layer just beneath the crust, was the source of most of the magma that erupts at volcanoes. Minerals that make up the mantle contain small amounts of water, not as a liquid, but as individual molecules in the mineral’s atomic structure.  Mid-ocean ridges, volcanic undersea mountain ranges, are formed when these mantle minerals exceed their melting point, become partially molten, and produce magma that ascends to the surface. As the magmas cool, they form basalt, the most-common rock on Earth and the basis of oceanic crust. In these oceanic ridges, basalt can be three to four miles thick.

Studying these undersea ranges can teach scientists about what was happening in the mantle, and about the Earth’s subsurface geochemistry.

One longstanding question has been a measurement of what’s called the mantle’s potential temperature. Potential temperature was a quantification of the average temperature of a dynamic system if every part of it were theoretically brought to the same pressure. Determining the potential temperature of a mantle system allows scientists better to understand flow pathways and conductivity beneath the Earth’s crust. The potential temperature of an area of the mantle can be more closely estimated by knowing the melting point of the mantle rocks that eventually erupt as magma and then cool to form the oceanic crust.

In damp conditions, the melting point of peridotite, which melts to form the bulk of mid-ocean ridge basalts, was dramatically lower than in dry conditions, regardless of pressure. This means that the depth at which the mantle rocks start to melt and well up to the surface will be different if the peridotite contains water, and beneath the oceanic crust, the upper mantle was thought to contain small amounts of water—between 50 and 200 parts per million in the minerals of mantle rock.

So lead author Emily Sarafian of Woods Hole, Carnegie’s Erik Hauri, and their team set out to use lab experiments in order to determine the melting point of peridotite under mantle-like pressures in the presence of known amounts of water.

“Small amounts of water have a big effect on melting temperature, and this was the first time experiments have ever been conducted to determine precisely how the mantle’s melting temperature depends on such small amounts of water,” Hauri said.

They found that the potential temperature of the mantle beneath the oceanic crust was hotter than had previously been estimated.

“These results may change our understanding of the mantle’s viscosity and how it influences some tectonic plate movements,” Sarafian added.

The study’s other co-authors are Glenn Gaetani and Adam Sarafian, also of Woods Hole.

This research was funded by the National Science Foundation and the Woods Hole Oceanographic Institution’s Deep Ocean Exploration Institute.

The atmosphere that allows our planet to sustain life formed from gases emitted by volcanoes early in Earth’s history. These volatile elements are constantly recycled back into the deep Earth at subduction zones, where tectonic plates sink into the mantle. During this process the sinking plate was subjected to increasing heat and pressure, and releases volatiles. These volatiles, once added to the mantle, induce melting and fuel volcanic explosions, completing the cycle. While this depiction of the earth’s giant recycling factory was well established conceptually, they do not know how efficient it is. They can estimate how much goes in, but have little idea what proportion was released back to the atmosphere, and what proportion remains trapped at depth. This question was crucial if they want to understand how our atmosphere formed and our planet became able to sustain life. In the present-day context, characterizing how much gas comes out of the giant recycling factory was also key to understanding volcanic effects on climate, volcanic emissions being significant – but poorly constrained – parameters in current climate models..

Their team of early career volcanologists was conducting expeditions to the South American Andes. Their objective was to provide the first accurate and large-scale estimate of the flux of volatile species (H2O, H2, CO2, CO, SO2, H2S, HCl, HF, and more) emitted by volcanoes of the Nazca subduction zone. The journey was taking us across half a continent, from the giant volcanoes of Ecuador through the altiplanoes of Peru and to the Southern tip of Chile, traveling on some of the Earth’s highest roads, and climbing some of the Earth’s tallest volcanoes.

Even though carbon was one of the most abundant elements on Earth, it was actually very difficult to determine how much of it exists below the surface in Earth’s interior.

Research by Deep Carbon Observatory scientists Marion Le Voyer, Erik Hauri (Carnegie Institution for Science, USA), Katherine Kelley (University of Rhode Island, USA) and Elizabeth Cottrell (Smithsonian Institution, USA) has doubled the world’s known finds of mantle carbon. Their findings, based on analyses of crystals containing mantle magma samples, are published in Nature Communications.

Overall, there was a lot about carbon chemistry that takes place below Earth’s crust that scientists still don’t understand. In particular, the amount of carbon in the Earth’s mantle has been the subject of hot debate for decades. This topic was of interest because the amount of carbon present in the mantle underpins our planet’s geological processes, including triggering volcanic activity and sustaining the biosphere. It also affects our atmosphere when carbon dioxide gas was released by eruptions; volcanic eruptions played a large role in pre-historic climate variations.

But it’s difficult to measure the amount of carbon that exists below the Earth’s surface. Scientists can study the igneous rocks that formed when mantle melts, called magma, rose to the surface, erupted as lava, and hardened again to create a rock that was called basalt. However, the process of ascent and eruption releases almost all the magma’s carbon as carbon dioxide gas, which makes the erupted basaltic rocks poor indicators of the amount of carbon that was in the magmas from which they formed.

“This is how explosive eruptions happen,” Hauri explained. “The sudden catastrophic loss of gas that, before the eruption, was dissolved into the magma at high pressure, but during eruption has nowhere else to go, leaving no post-eruption trace in the hardened basalt of the amount carbon once present.”

But Le Voyer, Hauri, and their team analyzed some basalt samples from the equatorial mid-Atlantic ridge that contained previously unstudied tiny magmatic inclusions, small pockets of pure magma that were completely trapped inside solid crystals that protected them from degassing during magma ascent and eruption. Analysis showed that these inclusions had trapped their original carbon content before being erupted on the seafloor.

“This is only the second time that samples of magma containing their original carbon content have ever been found and analyzed, doubling our knowledge of the region’s carbon chemistry,” Hauri said.

The very first samples containing their original carbon were also revealed at Carnegie, by Hauri and Brown University professor Alberto Saal, in 2002. Those samples came from the Pacific seafloor. Comparison of the data for these two samples revealed that the mantle’s carbon content was much less uniform than scientists had previously predicted, varying by as much as two orders of magnitude in different parts of the mantle.

“Our discovery that mantle carbon has a more complex distribution than previously thought has many implications for how mantle processes may vary by location,” added Le Voyer, who conducted this research as a postdoc at Carnegie and was now at the University of Maryland.

When volcanoes erupt, they spew lava, ash, and gas into the atmosphere and over the surrounding landscape. The impacts of volcanic eruptions in populated areas are well documented, since scientists can monitor gas emissions and collect physical samples with relative ease. However, a significant fraction of Earth’s volcanoes are remote, making direct observation challenging.

Some researchers have therefore turned to space-based techniques, collecting data from satellites. In a new paper published in the journal Nature Communications, DCO collaborators Brendan McCormick Kilbride, Marie Edmonds (both at the University of Cambridge, UK), and Juliet Biggs (University of Bristol, UK), analyzed several sets of satellite data in order to reconcile, for the first time, geophysical and geochemical models of magma composition [1].

“We’ve known for some time that when many volcanoes erupt we see a change in elevation, like the volcano has deflated, and we can measure this change using satellite observations,” said Edmonds, co-Chair of DCO’s Reservoirs and Fluxes Community. “We also know that volcanic eruptions produce gas plumes rich in sulfur dioxide, which we can also measure from space. However, what’s been a bit of a mystery is how these two datasets fit together. We couldn’t spot a pattern.”

By measuring how much a volcano deforms throughout an eruption, scientists can make an estimate of the volume of material previously contained within the magma chamber. These measurements, however, were at odds with the mass of sulfur dioxide measured in the erupted gas plume.

“We were seeing way more sulfur dioxide than we would expect for the volume of materials expelled during an eruption,” said first author McCormick.

The team therefore began modeling what was going on beneath the surface. They found that if more gases like sulfur dioxide and carbon dioxide are present in magma, they change its physical properties. Magmas with high concentrations of these volatiles are not homogenous. Instead, the gases clump together, creating bubbles. These bubbly, spongy magmas are a lot more compressible than magmas without bubbles, partially explaining why many large eruptions rich in sulfur dioxide show only limited changes in ground deformation.

The authors also suggest that in eruptions with huge gas emissions, a fraction of pure gas could collect at the top of the magma chamber prior to eruption making the ratio of erupted gas to eruption volume even larger.

“What we’ve done here is build a model of volcanic eruptions that could explain several aspects of volatile rich magmas,” added McCormick. “We’ve shown that our model is consistent with several datasets. If we can confirm this relationship between sulfur dioxide output and eruption volume, we come closer to being able to use satellite data alone to investigate some of Earth’s most remote volcanoes.”

“These data also tell us a lot about the Earth system,” added Edmonds. “As we refine our model, and integrate data from DCO’s DECADE volcano monitoring stations, we’ll generate a much more accurate picture of deep volatile cycles, including the deep carbon cycle.”

Turrialba volcano had deposited ash on the capital city of Costa Rica and its 3 million inhabitants numerous times since 2014.

In a new article in the Journal of Geophysical Research and an online Earthchem database, a DCO-DECADE team led by Maarten de Moor (National University, Costa Rica) and Alessandro Aiuppa (Palermo University, Italy) tracked changes in gas composition and flux from 2014 to present [1,2]. The near-continuous and high-frequency gas monitoring time series (Multi-GAS and scanning DOAS stations) reveal a volcano in a state of extreme turmoil, posing an increasing threat to local lives and livelihoods.

During the monitoring time period the deployed instruments recorded large changes in gas composition and flux. Carbon dioxide to sulfur ratios show significant variations with notable peaks that occur weeks to days before eruptive episodes. The precursory spikes in carbon dioxide are the result of pulses of deep magma injected into the volcanic system. These rising magma bodies are directly responsible for the timing and magnitude of Turrialba’s eruptive periods. The hydrogen sulfide to sulfur dioxide ratio also displays remarkable changes over two orders of magnitude during the monitoring period, from constituting a major component of the bulk gas early in the time series to being undetectable by the Multi-GAS station. The disappearance of detectable H₂S in the gas emissions indicates the progressive boiling off of what must be an enormous hydrothermal system as magma intrusions invade the volcanic edifice, and a transition to purely magmatic gas compositions.

Perhaps most interesting for understanding the deep carbon cycle, the authors’ estimations of CO₂ flux from Turrialba suggest the most voluminous release of carbon dioxide did not occur at the same time as magma intrusion. Rather, they saw the release of huge amounts of CO₂ well after the first magma intrusion event, an observation also associated with high H₂S emissions. This suggests that a major proportion of CO₂ released over the three-year period was actually hydrothermal in origin, with deeply derived CO₂ stored in the hydrothermal system for an indeterminate amount of time and then released into the atmosphere during energetic explosions.

Decompression degassing modelling and analysis of the CO₂-SO₂-H₂S-H₂O system allowed the authors to estimate magma depth. The long time series of SO₂ emission rate allowed them to calculate the total volume of magma intruded since the beginning of Turrialba’s unrest. Importantly, the large variations in both gas composition and CO₂ flux highlight the need for continuous gas monitoring to define the carbon budget of convergent margins. These results once again show the potential power of high-frequency gas monitoring for forecasting eruptions.