Guest essay by Eric Worrall
Scientists have just discovered a 1.8 million square kilometre lake semi-molten carbonate (CO2) compounds under our feet – but still think the contribution of volcanoes to the annual CO2 emission budget is insignificant compared to human emissions.
Scientists uncover huge reservoir of melting carbon under Western United States
February 13, 2017
New research published in Earth and Planetary Science Letters describes how scientists have used the world’s largest array of seismic sensors to map a deep-Earth area of melting carbon covering 1.8 million square kilometres. Situated under the Western US, 350km beneath the Earth’s surface, the discovered melting region challenges accepted understanding of how much carbon the Earth contains – much more than previously understood.
“It would be impossible for us to drill far enough down to physically ‘see’ the Earth’s mantle, so using this massive group of sensors we have to paint a picture of it using mathematical equations to interpret what is beneath us,” said Dr Sash Hier-Majumder of Royal Holloway.
He continued, “Under the western US is a huge underground partially-molten reservoir of liquid carbonate. It is a result of one of the tectonic plates of the Pacific Ocean forced underneath the western USA, undergoing partial melting thanks to gasses like CO2 and H2O contained in the minerals dissolved in it.”
As a result of this study, scientists now understand the amount of CO2 in the Earth’s upper mantle may be up to 100 trillion metric tons. In comparison, the US Environmental Protection Agency estimates the global carbon emission in 2011 was nearly 10 billion metric tons – a tiny amount in comparison. The deep carbon reservoir discovered by Dr. Hier-Majumder will eventually make its way to the surface through volcanic eruptions, and contribute to climate change albeit very slowly.
The abstract of the study;
Pervasive upper mantle melting beneath the western US
We report from converted seismic waves, a pervasive seismically anomalous layer above the transition zone beneath the western US. The layer, characterized by an average shear wave speed reduction of 1.6%, spans over an area of ∼1.8×106 km2 with thicknesses varying between 25 and 70 km. The location of the layer correlates with the present location of a segment of the Farallon plate. This spatial correlation and the sharp seismic signal atop of the layer indicate that the layer is caused by compositional heterogeneity. Analysis of the seismic signature reveals that the compositional heterogeneity can be ascribed to a small volume of partial melt (0.5 ± 0.2 vol% on average). This article presents the first high resolution map of the melt present within the layer. Despite spatial variations in temperature, the calculated melt volume fraction correlates strongly with the amplitude of P–S conversion throughout the region. Comparing the values of temperature calculated from the seismic signal with available petrological constraints, we infer that melting in the layer is caused by release of volatiles from the subducted Farallon slab. This partially molten zone beneath the western US can sequester at least 1.2×1017 kg of volatiles, and can act as a large regional reservoir of volatile species such as H or C.
What do scientists have to say about volcanic degassing? Most scientists don’t seem to think geological emissions are large, compared to anthropogenic emissions, but they tend to express a lot of uncertainty about estimates of geological emissions.
For example, Mörner and Etiope (2002);
Carbon degassing from the lithosphere
Nils-Axel Mörner, Giuseppe Etiopeb,
So far, the role of present-day Earth degassing in global C budget and climate effects has been focused to volcanic emissions. The non-volcanic escape of CO2–CH4 from the upper mantle, from carbonate bearing rocks in the crust, from hydrocarbon accumulations and from surface deposits and processes is here discussed in detail. An inventory of recent available data is presented. For the first time, a so large quantity of data is considered altogether showing clearly that the geological flux of carbon was previously significantly underestimated. Several lines of evidence show that non-volcanic C fluxes in «colder» environments are much greater than generally assumed. Local and regional data suggest that metamorphic decarbonation, hydrocarbon leakage and mud volcanoes could be significant CO2–CH4 sources at global scale. Moreover, extensive surface gas-geochemical observations, including soil–atmosphere flux investigations, open the possibility that ecosystems controlled by biogenic activity (soil, permafrost, seawater) can host important components of endogenous C gas (geogas), even in the absence of surface gas manifestations. This would imply the existence of a geological diffuse, background emission over large areas of our planet. New theories concerning the occurrence of pervasive geogas and lithospheric processes of C-gas production («lithospheric loss in rigidity») can be taken as novel reference and rationale for re-evaluating geological sources of CO2 and CH4, and an important endeavour and work prospect for the years to come.
Our survey shows that it is still very hard to arrive at a meaningful estimate of the lithospheric non-volcanic degassing into the atmosphere. Orders of 102–103 Mt CO2/year can be provisionally considered. Assuming as lower limit for a global subaerial volcanic degassing 300 Mt/year, the lithosphere may emit directly into the atmosphere at least 600 Mt CO2/year (about 10% of the C source due to deforestation and land-use exchange), an estimate we still consider conservative. It is likely that temporal variations of lithosphere degassing, at Quaternary and secular scale, may influence the atmospheric C budget. The present-day lithosphere degassing would seem higher than the value considered to balance at Ma time-scale the CO2 uptake due to silicate weathering.
There is a huge disparity between the amount of CO2 which is believed to be trapped a few hundred miles under our feet (1.2×1017 kg), vs the amount of CO2 scientists believe is emitted by volcanoes (300-600×109 kg / year).
Any free CO2 under the kind of pressure found in the Earth’s crust is likely to form a supercritical fluid – an strange, highly mobile form of CO2 which has characteristics of both a liquid and a gas.
I’m not saying Mörner and Etiope (2002) estimate is wrong, but it could be. Geologist Ian Plimer believes volcanic CO2 emissions are grossly underestimated, because vast numbers of submarine volcanoes are overlooked by studies which estimate emissions.
Volcanoes and CO2
Plimer has said that volcanic eruptions release more carbon dioxide (CO2) than human activity; in particular that submarine volcanoes emit large amounts of CO2 and that the influence of the gases from these volcanoes on the Earth’s climate is under-represented in climate models. The United States Geological Survey has calculated that human emissions of CO2 are about 130 times larger than volcanic emissions, including submarine emissions. The United States Environmental Protection Agency (EPA) stated that Plimer’s claim “has no factual basis.” This was confirmed in a 2011 survey published in the Eos journal of the American Geophysical Union, which found that anthropogenic emissions of CO2 are 135 times larger than those from all volcanoes on Earth. A 2015 study from The Earth Institute at Columbia University published in Geophysical Research Letters says activity from undersea volcanoes varies with tide, with greater activity at neap tide, and with more activity in ice ages with their lower sea levels. Dr. Maya Tolstoy, who conducted the study, says this might explain abrupt ends to ice ages.
I accept that humans contribute significantly to global atmospheric CO2 levels. But given the vast disparity between the estimated reservoir and estimated emissions, it is not difficult to see how the volcanic emission rate could fluctuate significantly over time, even in the absence of major eruptions, or how the “conservative” estimate of annual global geological CO2 emissions might be revised substantially upwards in the future.
Update (EW): h/t ATheoK Fixed a copy / paste error – 1.08×106 square kilometres, not 1.08×106 square kilometres