Historically, the earth has had five "Ice Ages" Each of them lasted millions, tens of millions, or hundreds of millions of years.

The last ice age reportedly ended perhaps 10,000 years ago. That seems to represent the beginning of recorded history.

Except that a period 10,000 years is a "rounding error" in the span of millions of years. If it was true that "we haven't been in an ice age for at least 1 million years, then I would believe that we were "clear" of the last one. But what is to say that the last 10,000 years isn't just an usually warm "spell" in what would otherwise be a cold period of say, a million years?

  • 10
    $\begingroup$ What makes you say we are not in an ice age? We have ice on/near both polar caps, which is pretty unusual historically speaking. I believe that by some accounts, we actually are in an ice age. To put it differently: it's really just a matter of definition. $\endgroup$
    – gerrit
    Commented Apr 15, 2014 at 22:44
  • 1
    $\begingroup$ Indeed, we currently are in an "ice age". $\endgroup$
    – DrewP84
    Commented Apr 15, 2014 at 22:50
  • $\begingroup$ A geologist would say we are in an Ice Age :-) $\endgroup$
    – winwaed
    Commented Apr 15, 2014 at 22:58
  • 2
    $\begingroup$ Don't forget that we know massively more about the last few tens of thousands of years (Holocene and last glacial period) than the deeper time beyond that. For the holocene, we have ice core records, tree rings, peat bog pollen data, etc. We know precise values of atmospheric gas concentrations, sea levels, ice volumes, etc. back to the last interglacial. For tens of thousands of years, we can use radiocarbon dating, but it becomes difficult past around 40Ka ago. So, your "rounding error" idea is incorrect because precision increases dramatically in the last 40Ka. $\endgroup$ Commented Apr 16, 2014 at 0:26
  • $\begingroup$ An Ice age is just defined as a period where a significant portion of the earths landmasses are covered in ice sheets, as long as greenland and/or antarctica have ice sheets we are in an ice age. $\endgroup$
    – John
    Commented Apr 13, 2019 at 2:25

3 Answers 3


I think you're slightly confused by some of the terminology. (Caveat: I'm a geophysist, take anything I say with a grain of salt!)

We're currently in an interglacial during a prolonged period of icehouse climate (most of the Cenozoic).

During most of the Earth's history, the overall climate tends to be much warmer and more stable on the million-year scale. These periods are referred to as "greenhouse" climates. During these times, there's relatively little evidence for extensive ice sheets (either on continents or over the ocean) at the poles.

However, during times of "icehouse" climate, the Earth becomes cooler. Ice masses develop at the poles and due to regular preturbations in Earth's orbit (Milankovich cycles), continental ice sheets tend to advance and retreat. (Assuming there are continents near the poles.) This leads to regular glacial and interglacial periods.

Presumably, regular fluctuations in climate due to Milankovich cycles would occur during periods of greenhouse climate. However, ice sheets cause positive feedback that amplifies the effect of the Milankovich cycles. This is largely due to ice and snow's albedo: they reflect more sunlight back into space instead of adsorbing it. Therefore greenhouse Milankovich cycles were probably smaller. (There's probably a lot more to the story. I'm not an ideal person to ask.)

  • $\begingroup$ Your second sentence reminds me of something I read in an earth science books for kids as a child: "We are in an ice age, but a warm period of that ice age". Of course, this was confusing for the younger me, growing up in a semi-arid climate! $\endgroup$
    – Gimelist
    Commented Nov 1, 2014 at 14:13

Essentially, we can't. At a meeting some years ago, different specialists tried to define the onset of glaciation and found that each (sub-) discpline had their own definition. An oceanographer boldly stated the glaciation started 2000 years ago since a change in the Greenland ocean currents occurred at that time which was observed to also change before the latest glaciation. But, since we are unable to predict natural variations with any certainty (at least useful for any human lifetime) such observations remain speculation and ideas worthy of more research. Milankovich theory predicts an ice age but the onset will likely be very weak for us to detect in the general noise and human interference with the system.

In addition, we must remember that and ice age is not a sudden event (as seen in the movies) and is likely to start with a slow general drop in temperature with superimposed oscillations (both colder and warmer), at least if our understanding of past ice ages is anything to go by. Hence, we will not know, other than in hindsight, and by some definition, when such a time started.

So, our understanding says we should be heading towards a colder period but how we enter into it, for example, given our influence of the climate system makes it difficult to predict.

  • 1
    $\begingroup$ Such is the problems with most geoscience; none of us can live thousands to millions of years. $\endgroup$
    – Neo
    Commented Apr 16, 2014 at 6:11

I will answer mostly from archaeological perspective; I don't know much about the paleoclimatology before the first people started to produce stone tools. I know just some basics of other disciplines where they affect human (pre)history, but it might help.

First, even though longer and older cold periods are also refered as "ice ages", most people don't understand "ice age" as this, but as the time when our ancestors hunted mammoths. This is "archaeological" perspective. If you are interrested in the older geological history, Joe Kington's answer explains the "geological" perspective better than I could. From this perspective, no ice age ended 10000 BC - the climate during current "big Ice Age" just went little milder.

I've been told several times during paleolithic and paleoclimatology courses that according to current state of knowledge, there were 52 cold and the same number of warm periods during the quaternary. I didn't really try to find some more sources now, and I'm not sure whether the cold periods can all be classified as "ice ages". I'm not sure, but I think that Bølling-Allerød interstadial is considered to be one of the warm period and Younger Dryas might be an example of a cold period. If we divide the duration of quaternary by 52, we get around 50ky per cold + warm period; so either the 52 + 52 periods were longer and roughly regular, or Bølling-Allerød/ Younger Dryas drag the average length down, while ice ages lasting tens or hundreds thousands years drag the average up.
Anyway, even if there were just four big Ice Ages during quaternary, as scientists expected few decades ago, they wouldn't last millions of years, and the rounding error would be much smaller than 10ky.

Bølling-Allerød and Younger Dryas are especially relevant to your question, since they show that the last Ice Age (with maximum of glaciation around 20ky BP) didn't end suddenly, but that it ended for some 2000 years, then returned for some 1300 years and than ended for good (we usually mark this moment, around 11500 BP, as the beginning of Holocene). The prevailing theory is that the reason of the sudden cold shift is melting of a huge amount of ice, which decreased salinity of ocean water, slowing the Gulf stream down. So there is some "rounding error", but we are out of its scope - I don't think there is enough ice to slow down the Gulf stream enough to trigger something that could be called "ice age" in the "archaeological" sense.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.