The Continuing Mysteries of the Ice Ages

May 20, 2009

Stuart Gaffin is a climate researcher at Columbia University and a regular contributor with his blog “Exhausted Earth”. Thomson Reuters is not responsible for the content – the views are the author’s alone.

Understanding the ice age cycles that have occurred on the Earth during the past million years is — without question — one of the great scientific puzzles of all time.

By ‘great’ I mean not only the importance for many current environmental issues, like climate change and the massive greenhouse gas increases, but great in the sense that solving the mysteries of their occurrence requires breakthroughs from so many different fields of science.

I did post-graduate research on ice ages with a climatologist, Barry Saltzman, who was a co-discoverer of chaos (with Ed Lorenz) and who spent much of his later career on ice age science. I remember marveling with him over the almost ridiculous number of disciplines ice-age science involves: geology, glaciology, climatology, atmospheric physics and chemistry, oceanography, astronomy, geochemistry, biology, geomagnetism, meteorology, nonlinear mathematics and probably other fields I’m not listing!

It’s like we are riding on the back of some kind of Moby Dick of science, learning incredible things about the Earth along the way. Not only that, but the ice ages obviously have shaped us as a species as well since we evolved during the waxing and waning of the enormous ice sheets, influencing our harnessing of technologies (tools, clothing, shelter) and minds. Children are intrinsically fascinated by the ice ages and scientists have been attacking the problem since 1800.

Against this venerable background, I am humbly publishing a paper that tries to paint a new picture of the ice ages. I am linking it here (and publisher here) for those of you want to see the details of my case.

There is a more-or-less standard model of the recent ice ages that says they have resulted from a gradual long-term cooling of the Earth since the time of the dinosaurs, when no glaciers are known, towards the present. At some point, around 34 million years ago, the Earth became cool enough for large scale glaciers to form on Antarctica and later elsewhere, culminating in the great 100-thousand year cycles that have dominated recent times.

My paper does not question this model — indeed it is essential for my case that it is largely correct. Rather I focus on the 100-thousand year cycle mystery and what caused that. The essence of my paper is that this mystery of the cycles may be related to another cycle in geology that is also unexplained – sedimentary cycles on the margins of continents during warm geologic epochs.

These coastal cycles were discovered by oil geologists (most notably at Exxon) during the 1960s and 1970s.  In my paper I try to show that there are a mind-bending number of parallels between these two cycles that cannot be accidental. The one common link that explains these parallels is that both ice sheets and continental margin systems are sedimentary systems.

You can read my paper if interested but among other things I conclude that continental margin sediments may oscillate almost automatically. Another conclusion, relevant to current global warming, is that my paper bolsters the case that Antarctica did not support any glaciation during very high greenhouse gas periods of geologic history. I think we are just at the ‘tip of the iceberg’ in understanding the mysteries of ice ages and their relation to other systems on Earth.

The leviathan lives on!

(Credits: top right: an image of Cro Magnon Man, Photo by Roderick Mickens, copyright by American Museum of Natural History. Map centre: Ice sheet extent during ice age: Hannes Grobe/AWI)


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

A breakthrough in the ice age mysteries will undoubtedly be interesting. Lots of unknown facts still remain hidden in Antarctica and the ice just got a lot more interesting. Recent research shows that frozen influenza strains are becoming active as a result of the global warming. Now who would have known that? ozen-killers-in-antartic.html

Posted by Sunny G | Report as abusive

Stuart, have you considered the involvement of the methane cycle? In order for methane to convert from a solid to liquid and then gas, it must extract heat from the surroundings. Methane is gas at most temperatures. But under high pressure and in cold climates, there is a solid form of it underwater that is tremendously abundant.

Since the earth has many natural mechanisms intended to maintain equillibrium conditions, it makes sense for a cooling system to kick in if the system overheats. Heat would be taken from the surface and carried up where it would radiate into space leaving a frosty surface on the planet.

During a chill-out period, there would be widespread loss of life. But among the first to recover during a warming would be bacterial populations which would have time to break down the residual matter. So all of the nutrients including trace elements would be returned to the soil for the next age of plant and animal life. Then the who thing repeats. That is my guess anyways.

Posted by Don | Report as abusive


Methane (CH4) and Carbon Dioxide (CO2) have long been known to be intimately involved in the ice age cycles. Indeed these ice core greenhouse gas (ghg) change discoveries are among the biggest ‘breakthroughs’ from other disciplines (atmospheric chemistry, glaciology, etc) that I was referring to.

The causes of these ice age ghg changes are quite important, to say the least, for our current CO2 and CH4 situation. Many different hypotheses have been put forward including, for CH4, coastal and continental shelf source changes you mention. But a consensus on the dominant mechanisms, if there is one in each case, does not exist yet, is my understanding.

It’s safe to say these ghg changes acted as key positive feedback process for ice age cycles through their effect on atmospheric thermal radiation, like today.

I don’t discuss these ghg cycles directly in my paper because we don’t have data on how CO2 and CH4 may have changed during the Cretaceous 1-3 million year sediment cyles I discuss. However I do believe that such changes probably took place because coastal processes have such an important influence on ocean chemistry among other things. As proxies for paleo-atmospheric chemistry improve over time, hopefully we will know more about this.

Posted by Stuart | Report as abusive

I’m assuming that the oceanic methane and solid methane being referred to are methane hydrates (MH), which is gaseous methane wrapped in ice. As recent reports from Norway and other stations show, there has been a significant increase of methane in atmospheric concentrations, especially in the northern latitudes. As Katey Walter’s work, and that of Russian scientists, has detailed, methane hydrates in both permafrost and oceanic deposits in the Artic have been releasing methane in large amounts for the past few years. Whether or not this is the beginning of feedback loop initiation is unknown but suspected.

Also of concern is the fact that the oil and gas industry has kept a dirty little secret from the public eye for many years. They have been drilling into and through these MH beds with terrible results, including over 500 gas blowouts worldwide in the last 25 years. But because these blowouts occur from shallow gas hydrates, before any oil pocket has been penetrated, these purely gas blowouts aren’t considered a pollution event and don’t have to be reported as such, since they evaporate and leave no tell tale sign.

The truth of the matter is that they have released massive amounts of methane from the MH reservoir, and none of this, including methane release from the Artic, is included in the modeling used to estimate global warming scenarios, which is probably why methane concentrations have been increasing once again.

I have put together a report entitled “Fire in Ice” which details all of this. If interested let me know and Ill get a copy to you.

Posted by Dan Zimmerman | Report as abusive

Dan, thank you for illuminating truth. This may well be a factor in climate models getting time frames wrong on computer generated forecasts.

Posted by Anubis | Report as abusive