Monday Minute: Mid-Ocean Ridges Record Dozens of Ice Ages over Millions of Years

Last Monday, I briefly discussed the origins of oceanic crust and why it contradicts the young-Earth paradigm. Prior to that, I summarized how multiple ice ages have been recorded in Siberian speleothems, which precludes the idea that a single “post-Flood” ice age could explain glacial geology. So in case you are tired of hearing about the ocean floor and ice ages, then this post is not for you!

For this week’s Monday Minute, though, I couldn’t help but to relay the latest report from Science News, because the timing couldn’t have been better. Several papers were recently published (including Crowley et al., 2015), which describe how glacially driven changes in sea level actually impacted the rate of volcanism at mid-ocean ridges. Fortunately, the mechanism is surprisingly elegant and very easy to grasp:

  1. During the peak of ice ages, sea level is estimated to have dropped by ~120 meters worldwide (that missing water was transferred to the land in the form of glacial ice).
  2. Massive hydrostatic pressure at the bottom of the ocean dampens the rate of seafloor volcanism (watch the video in last week’s post to see that volcanism in action). But when sea level dropped, that pressure was relieved, allowing volcanism to proceed at a slightly higher rate.
  3. Therefore, the ocean floor should have been slightly thicker during past ice ages, when sea level was lower. These intervals could be visible on the ocean floor in the form of minor ridges and hills that run perpendicular to the spreading axis.

It sounds like a great hypothesis, but how could we ever test it, since none of us were actually there? That was the goal of researchers, who sought to confirm that regularly spaced ridges and hills near the Australian-Antarctic ridge were indeed associated with Milankovitch-driven ice ages over the past few million years. These features were mapped out at high resolution several years ago in the southern Indian Ocean. Using those data, the researchers were able to plot the thickness of the ocean floor along a transect leading away from the ridge (i.e. back in time).

Now, we have long known the age of oceanic crust gets older as one moves away from spreading ridges, but advances in radiometric dating techniques allowed geologists to attach real numbers to the entire oceanic crust. In case you are eager to object to the accuracy of those dating techniques, I ask that you bear with me. You see, assuming the accuracy of those radiometric dates, the researchers could assign real ages to anomalously thick zones (where volcanism had been more active in the past). Using advanced statistical techniques, they determined that volcanism increased significantly at regular intervals of 23, 41, and 100 thousand years.

Do those numbers sound familiar? If you’ve ever studied Milankovitch theory and its application to paleoclimatology, then they should. These are precisely the intervals at which the Earth’s orbit varies with respect to the sun, and hence they explain the timing of past ice ages. In other words, the new studies confirm not only that Earth has endured dozens of ice ages over the past millions of years, which impacted seafloor volcanism, but also that radiometric dating of oceanic crust is highly accurate. If it were not, then we’d be hard pressed to explain this incredible coincidence.

On a final note, the association of volcanism with sea level elucidates the process by which the ice ages met their end. Volcanism at mid-ocean ridges injects massive amounts of carbon dioxide into the oceans, which eventually makes its way to the surface. The further sea level dropped, therefore, the more CO2 was added to the atmosphere. More CO2 lead to a warmer atmosphere, which melted the glaciers that caused sea level to rise. As sea level rose, rates seafloor volcanism dropped, and the atmosphere stabilized.

What an incredible planet we live on… Wouldn’t you say?

divider Featured image: Bathymetry of the East Pacific Rise via GeoMapApp, for illustrative purposes only.


9 responses to “Monday Minute: Mid-Ocean Ridges Record Dozens of Ice Ages over Millions of Years

    • Yes, Watts shows a bit of novice here when it comes to paleoclimate (though to be fair, he humbly admits that much of this was new to him), and he ends up oversimplifying Milankovitch theory and glacial-interglacial cycles.

      For one, what he calls the “100,000-years problem” really isn’t a problem anymore, because we now understand the role of feedbacks (and which feedbacks are most important to the climate system). These feedbacks amplify effects from solar radiation, which is the reason he sees no obvious match between the Antarctic temperature series and Boreal Summer Insolation.

      Once we factor in how things like ice sheets, sea level, forest coverage, and of course greenhouse gases modify the solar signal, Milankovitch theory perfectly explains why we see a strong 100,000 years cycle.

      Edit: When I say perfectly, I had in mind this paper:
      Kohler et al., 2010, What caused Earth’s temperature variations during the last 800,000 years? Data-based evidence on radiative forcing and constraints on climate sensitivity

      If you’re interested in seeing full text, but don’t have access, just send me an e-mail and I’ll share the PDF.


      • Eschenbach, not Watts.

        You know better than me, but I’m not sure you are not the one over simplifying, and also over estimating our knowledge of amplification factors.

        I have low confidence in the ice, but we do have some reason to believe that rates are too slow to justify my worries. While I admit my information is dated, last I knew (again, some years ago), the foraminifera proxies didn’t agree well with the ice proxies.

        There is always value in looking at the data over the theory. I see little in the data that supports much of current theory on earth’s climate history. Temperature is remarkably stable. Astonishingly stable. 291±8 K
        I also cannot get over the fact that the warmest epic in earth’s history is when the grass-eaters came on the scene, as well as our own first primate ancestors.

        I do agree that we live on a wondrous planet.


      • My mistake! Thanks for the correction.

        I hope that I am not misrepresenting the extent of our current knowledge, but since I’ve been immersed in paleoclimate papers for the last several years, it’s difficult for me not to be overly confident.

        To my knowledge, foraminiferal records line up very well with ice cores, and both are amply supported by cave and coral records. However, this was not initially the case; improvements to age models and proxy calibration have greatly enhanced the correlation.

        May I ask, why do you have “low confidence in the ice”?


      • Speaking of absolute temperatures, I wouldn’t refer to the Scotese plot, by the way. It’s an artifact of his method that temperature only appears to vary by ±8°C, hence it is useful only to consider relatively warm/cool periods (same with Veizer’s classic plot from d18O, which nobody trusts at all to give us absolute temps). I don’t know anyone that would say with confidence that global temperature was that stable over Earth history. Also, it is highly doubtful that the Paleocene/Eocene transition was the warmest epoch in Earth history, let alone the Phanerozoic.

        In any case, a few more thoughts came to mind regarding the first article by Eschenbach. His third paragraph is so hopelessly off the mark, that this particular post ought to be yanked from any serious discussion on paleoclimate. He writes:

        “I hadn’t realized the size of the swings. The cycles are about 21,000 years long and the swings are quite large, up to 100 W/m2 from trough to peak… The current hypothesis is that at equilibrium we should see a swing of ~3°C for each additional 3.7W/m2 of forcing… So according to the current thinking, a swing of an additional 100 W/m2 which is maintained for a thousand years should result in an increased annual temperature swing of about 40°C (73°F) … and we don’t see anything in the geological records even half that size.”

        He is referring, of course, to a plot of summer insolation at 40°N latitude. Yet for some reason he translates this to global insolation and tries to calculate the T change using common sensitivities in climate models. Apparently, what he doesn’t realize is that when summer insolation increases by 100 W/m2 at one latitude on Earth, it decreases elsewhere. Milankovitch cycles don’t increase/decrease the average insolation across the globe, so the net change from glacial-interglacial periods is zero, not ~100 W/m2. Radiative forcing derives solely from various feedbacks: greenhouse gases, ice-sheet and vegetation related albedo changes, etc. But it is the insolation changes at particular latitudinal bands (e.g. where the ice sheets/permafrost zones/sea ice/steppe tundra exist) that induces those feedbacks, which explains the phase relationship between N. Hemispheric summer insolation and glacial cycles.

        It also explains why the 100 kyr peak dominates once global climate became cool enough to sustain large-scale ice sheets: most peaks in summer insolation were not strong enough to overcome the cooling effect of the previous insolation minimum.


  1. I’m a metallurgical and nuclear engineer. I consider it fair to say I’m a solid state physicists, but still an engineer by nature (and by practice since leaving school). Gas does not forever stay trapped. It diffuses. Gibbs free energy must be minimized, and trapped bubbles are not minimum. Still, rate kinetics rule. While I know where the drivers are, I don’t know the kinetics behind it all. Evidence leans toward slow enough kinetics to make the ice bubbles representative. Regardless, they are still not the precise information I think we need for making assumptions for calculating forcings and feedbacks.

    As to foraminifera, my vague understanding is that there has been some success in aligning isotopes, but I thought there were still significant differences when basing on species. There are also the divergent indications of plant stomata versus the ice proxies.


    • That’s a very interesting take; thanks for your insights. I think you’re right that diffusion could be a major concern. On the other hand, even if the absolute measurements are imprecise, we should be able to capture relative changes accurately (i.e. the timing of greenhouse gas fluxes). Do you think?

      Yes, using relative abundance of foraminiferal species, there ought to be something of a mismatch. Species abundance depends not only on temperature, but on stratification in the upper ocean and oceanic current strength, for example. These factors are both a temperature response and a feedback in the climate system. Unfortunately, it’s much more difficult to constrain paleocurrents, but there has been some improvement over the past decade.


  2. Radiometric dating results supporting ages for past glaciations based on Milankovitch cycles (thought to have triggered numerous glaciations during the last 3 million years or so)? Looks like this means the YECs’ version of their being just ‘one’ ‘recent’ ice age glaciation, triggered mainly by ‘additional’ (and not very biblical) volcanism associated cooling just after the sea level was much much higher than normal (during the ‘Genesis Flood’) needs to be BINNED (unless a Young Earth Creationist can REFUTE all this). And by ‘refute’ I don’t mean merely “none of us was there so we must use the Bible as our starting point and not any actual evidence however persuasive”.


  3. Pingback: Monday Minute: Flood Geology of the “Mountains of Ararat” | Age of Rocks·

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s