In a recent article in Nature Geoscience, DCO Reservoirs and Fluxes Community Members Kyle Anderson and Michael Poland, both of the U.S. Geological Survey, described the results of a new way to estimate the amount of carbon deep in Earth’s mantle . They combined measurements of carbon dioxide emitted from Kīlauea, Hawaii’s most active volcano, with indirect measurements of how much magma flows into the volcano. Through modeling, they calculate that the carbon content of the mantle beneath Hawaii is about 40% higher than previous estimates. If verified, these findings could have important implications for the global carbon budget, the size of deep carbon fluxes, and historical reconstructions of Earth’s climate.
Peter Barry (pictured), a noble gas geochemist at the University of Oxford, wrote a News and Views commentary to accompany the paper in Nature Geoscience . Barry is a member of DCO’s Deep Energy Community and the Reservoirs and Fluxes Community who has been active in organizing Early Career Scientist workshops and was the lead principal investigator on the Biology Meets Subduction expedition in February 2017.
He talked with DCO science writer Patricia Waldron over Skype about the new paper and its significance for deep carbon research.
Let’s start with a basic question. Carbon scientists study volcanoes because the materials that erupt out of them give clues to the rocks below in the mantle. But these studies have arrived at a large range of mantle carbon concentrations. Why is it so hard to estimate quantities of carbon in the mantle?
That’s the heart of the whole issue, because when you have partial melting of the mantle to form molten rock, called magma, the carbon wants to leave instantaneously, because it is so incompatible with the melt. So any time you collect a rock sample at a volcano, it’s largely degassed, meaning that it doesn’t have very much of its initial carbon left; it escaped to the atmosphere.
If we’re trying to make a direct measurement of carbon, we’ll go to places where there’s confining pressure that contains the carbon dioxide gas. At a mid-ocean ridge, for example, you have the whole weight of the ocean sitting on that volcano as it’s erupting, so magma there is less degassed than magma erupting at the surface.
We also go to places that are subglacial, like Iceland, and look at rocks that are erupted underneath the confining pressure of a glacier. That keeps the gas content relatively high in the rocks.
Anderson and Poland are studying hotspot volcanoes. These are places like Hawaii, or Reunion Island, which are subaerial volcanoes, meaning the magma there is erupted at the surface.
What is a hotspot volcano?
The theory behind a hotspot volcano is that there are thermal anomalies in the mantle and the extreme heat causes a deep-seated plume to emanate from perhaps as deep as the core-mantle boundary, towards the surface. Hotspots represent a strange and intriguing type of volcanism.
How much carbon did they find?
Anderson and Poland calculate a value of around 1% by weight carbon in the undegassed magma and they also make an estimate of 263 ppm (parts per million) in the mantle source region. If there really is 40% more carbon in the mantle source, compared with what was previously estimated, then that has broad implications for all kinds of things relating to the evolution of Earth’s interior and even long-term climate dynamics.
What did you think of their approach?
They’ve taken a unique approach to solving a problem that’s been around for a long time and it’s completely different from what I would do as a geochemist. They are computer modelers and they’re utilizing a great deal of available information and used it as inputs for their models. But the questions that they’re asking are fundamentally the same as what geochemists like me ask.
What do you find to be most interesting about this work?
It’s interesting to have a fundamental understanding of how much carbon is in the various parts of the mantle, because it governs the overall carbon budget of the planet.
How would these findings impact climate studies?
On human timescales, certainly in modern times, anthropogenic carbon emissions are huge relative to volcanic emissions. We burn 135 times more carbon from fossil fuels than what comes out of volcanoes every year, which is a staggering statistic.
However, before we released any anthropogenic emissions, the carbon in Earth’s atmosphere over the last hundreds of millions or billions of years was likely governed in part by volcanic emissions. So if this is a robust estimate and there was and is significantly more carbon coming from these deep sources, that has really important implications for how we interpret past climate dynamics.
What’s the next step?
So far Anderson and Poland have only made these models for Hawaiian volcanoes, so I think the test for this will be applying the same technique to other hotspot volcanoes around the world. If they can do that, and reach similar conclusions, then we will really start to shift our understanding of the carbon content of the mantle. The estimates they’re making suggest that there’s something like an order of magnitude more carbon in the deep mantle source than there is in the upper mantle.
Is there anything else you would like to add?
To see research that’s focused on volcanology and mantle reservoirs featured in Nature Geoscience is fantastic. Understanding the reservoirs and fluxes of carbon is exactly what many scientists in the DCO community think about, so I think, in that sense, this is extremely exciting.