How do CO2 levels relate to ice ages and sea-level?

In Paleoclimate by Rob Monroe7 Comments

Core of ancient Antarctic ice

An ice core collected by the DISC (Deep Ice Sheet Coring) drill from the West Antarctic Ice Sheet

In a recent comment, a reader posted a graphic in which CO2 and sea-level rise appear to be correlated throughout the last 700,000 years. While these two factors do appear to rise and fall together, what drives these two factors over the period of 700,000 years is very different from those over the last 50 – 100 years. As a result, this correlation is not a reliable guide to the cause and effect relationship between CO2 and sea-level rise in recent years.

Until the Industrial Revolution, melting of snow and ice at the end of cyclical glacial periods was driven by changes in the Earth’s orbit. As snow and ice melted on the planet, the albedo, or reflectivity, of the earth declined, with land and ocean absorbing more heat from the sun (think the reflectivity of cars in a parking lot versus the hot blacktop of the parking lot itself). As the planet slowly warmed, a warmer and better-mixed ocean released CO2 into the atmosphere, which amplified the warming that was already in progress. In all, temperature typically increased by 6°C (11° F) over thousands of years during these interglacial periods, and one third of this increase was a result of the CO2 that outgassed from the ocean once warming began. Therefore, CO2 was not the initial cause of melting ice on the planet. It merely amplified a signal that was already in progress.

By contrast, as humans burn fossil fuels, we are creating a new driver of snow and ice melt and accompanying sea-level rise. This time, additional CO2 entering the atmosphere is coming from sources of carbon that had previously been sequestered in fuels such as oil and gas. We know from basic physics that CO2 traps heat. The phenomenon is perhaps best demonstrated by Earth’s neighbor Venus. If we lived on Venus, the computer you’re reading this post on now would be melting. That’s because Venus has 100,000 times the amount of CO2 in its atmosphere as Earth does and its average temperature of 462° C (863° F) is high enough to melt lead.

Thus, while in the past sea-level rise as a result of melting glaciers and thermal expansion was driven by changes in Earth’s orbit and amplified by oceanic CO2 released after initial warming, today anthropogenic CO2 is directly driving this melting and sea-level rise.

– Emily Kelly is a 6th year student in the laboratory of Scripps marine biologist Jennifer Smith


  1. Matthew Murrey

    I’m curious. If the early warming periods induced CO2 release from the oceans, will the current warming do the same, or since we are in an interglacial period anyway, have the oceans already released expected amounts of CO2? Hope my question makes sense. Thanks

    1. R. Gates

      Great question. Indeed, during normal interglacials we see a series of feedback mechanisms come into play that have the net effect of allowing the oceans and biosphere to release more carbon into the atmosphere, which eventually balances out with ocean temperatures, atmospheric temperatures, and atmospheric carbon dioxide levels reaching a general balance for the bulk of the interglacial before slowly declining as the next glacial advance occurs. This is all part of the general carbon cycle and the balance is created through a balance of a series of negative and positive feedbacks that keep the CO2 levels in check. During the early part of this particular interglacial, which of course goes by the common name of the Holocene, the same general pattern held. However, the big difference is of course the actions of humans (hence the term “Anthropocene” being adopted by some). Through many processes, but primarily the burning of fossil fuels, the carbon cycle has become unbalanced. Natural feedbacks are being overwhelmed by the large amount of carbon that humans are transferring from lithosphere to atmosphere and ocean. The ocean has become a net carbon sink, with it slowly becoming more acidic as a result. Even as the Earth gets warmer, the oceans will continue to absorb more carbon, however, the rate of absorption might indeed slow down as saturation is reached, whereby, the atmospheric levels would start to increase even faster if humans continue to burn fossil fuels.

      More troubling or at least of concern to some in the climate science community is not slowdown of the rate of absorption of CO2 by the ocean (which is troubling enough), but the potential for the warming oceans to begin to destabilize methane hydrates which exist at various depths in the ocean around the world. These hydrates, if they destabilize and release the more potent greenhouse gas methane and that makes its way to the atmosphere, could rapidly accelerate global warming– cresting a powerful new positive feedback process. Exactly this one-two punch, CO2 and then methane, may have in fact caused the PETM event some 55 million years ago.–Eocene_Thermal_Maximum

      Methane is already at its highest levels in millions of years and rising. So while the slowdown of the absorption of CO2 by the oceans is worth watching and of concern, the potential destabilization of methane hydrates and other sources of methane might be of a bigger and immediate concern.

      1. Jeff Severinghaus

        To Mr. Gates: The methane hydrate scenario is indeed worrisome, but some recent research suggest that most of the methane that will be released in coming years will be consumed by microbes in the ocean or in soils before it ever makes it out to the atmosphere. My students and I measured the carbon-14 content of atmospheric methane from air trapped in Greenland ice, before and after a major abrupt warming event that occurred 12,000 years ago. Atmospheric methane concentrations increased 50% during this event, so a natural hypothesis was that the warming cause methane hydrate to destabilize and outgas methane to the atmosphere. Carbon-14 allows one to distinguish between hydrate methane and other sources such as wetlands, because the hydrates have no carbon-14. What we found was that the carbon-14 content of atmospheric methane did NOT drop as would be expected if methane hydrate had caused the methane concentration increase. Instead, what likely happened was that methane-eating bacteria gobbled up the methane before it could enter the atmosphere. The Deepwater Horizon accident provided another piece of evidence – more than 99% of the methane released into the Gulf of Mexico in that accident never made it to the atmosphere because it was consumed by microbes in the water thousands of feet below the surface. If you would like to read more, see

        1. R. Gates


          Thanks for that information– very informative, and perhaps a bit comforting. I also note with interest though that some indications are that methane producing organisms could increase in activity with the melting and warming going on in Arctic tundra. Specifically, I would cite this research:

          So the picture seems a bit more complicated. As a percentage increase, methane has been outpacing CO2 a bit, and of course, regardless of source, these kinds of trends for either gas cannot continue without potentially significant effects on the climate in the long-term.

          1. R. Gates

            Just a quote from the article referenced above:

            “”Methanoflorens stordalenmirensis seems to be a indicator species for melting permafrost. It is rarely found where there is permafrost, but where the peat is warmer and the permafrost is melting we can see that it just grows and grows. It is possible that we can use it to measure the health of mires and their permafrost. The recently documented global distribution also shows, on a much larger scale, that this microbe spreads to new permafrost areas in time with them thawing out. This is not good news for a stable climate”, says Rhiannon Mondav

            Read more at:

    2. Jeff Severinghaus

      Yes, it will. The oceans will “release” a small amount of CO2 to the atmosphere in response to the human-caused warming, just as they did during the ice ages. The reason I put “release” in quotation marks is that the huge amount of human CO2 so overwhelms the natural fluxes of CO2 out of the ocean, such that in fact the ocean will simply take up a little less CO2 than it would have if the temperature had been held constant, but effectively it is a release. But keep in mind the relative magnitudes here – the whole ice age change was only 80 ppm CO2, whereas humans are on track to increase CO2 by many hundreds of ppm in the coming centuries. So it is a modest amplifying feedback, perhaps only 10-20%.

  2. robert Wilson

    So increased methane availability increases primary productivity. Can we develop this potential as a technology? Would it work with green algae? What is the state of research on photosynthetic panels? Agriculture, like no till farming, that conserves carbon. Most of the focus is on reducing source carbon. How can we increase carbon sink?

Leave a Comment