Ice Ages Revisited: A Story of Climate

June, 1990. Our bush pilot maneuvered his Cessna 185 through the jagged peaks of the Brooks Range, bringing it and us down on the tundra beside Alaska's Jago River, above the Arctic Circle. We set up camp and had supper.

Later that "night" I was still awake, and crawled out of the tent for a walk. It occurred to me to snap a photo of my wristwatch, to memorialize the moment. Almost midnight. Sun still going strong. So close to solstice, it would remain above the horizon all night. No wonder I couldn't sleep.

Even hundreds of miles south, in Fairbanks, the continuous light had been disorienting. Locals spoke of high-energy summers with nonstop frenetic activity, like late-night baseball games. The long, dark, torpid winter, in which the sun barely makes an appearance, is just the opposite. In the far north, that's when you do your sleeping.

As every schoolchild knows, the earth rotates about its axis: an imaginary line that runs through the planet, connecting the north and south poles. The spin axis is not vertical with respect to the sun. Rather, the earth tilts seasonally toward or away (depending on your perspective relative to the equator) from the sun, exposing the polar regions in particular to extreme annual variations in the duration of sunlight.

(Click on the diagram for a better look)

During the northern summers, the Arctic is tilted toward the sun, bathing the region in near-continuous sunlight for a few months. The exact conditions vary with latitude and day of the year.

Simultaneously, the south pole is tilted away from the sun, so the Antarctic experiences the cold of winter and near-continuous darkness.

Half a year later, with the earth moved in its orbit to the opposite side of the sun, the seasons are reversed. Now the Arctic, tilted away from the sun, is cold and dark. The Antarctic enjoys summer and sunlight because it tilts toward the sun.

Importantly, then, the abundance or dearth of sunlight at high latitudes during summer or winter is a feature of the planet's tilt. By examining the diagram, you can see that if the spin axis were more vertical, the Arctic summers would receive somewhat less sunlight, because the northern hemisphere would be tilted somewhat less toward the sun. At the same time, the southern hemisphere would be tilted less away from the sun, so the Antarctic would receive more winter sunlight.

Indeed, the poles are just extreme cases of seasonal variation that the entire planet experiences. We have seasons precisely because of axis tilt.

Moreover, planetary tilt explains not just seasons, but also glacial epochs over fairly long time scales. Which is to say, "ice ages."

That's because the degree of axis tilt varies slightly over a 40,000-year period. Sometimes the earth straightens up a bit, and sometimes it leans over. The variation is not large, but it is significant and predictable.

Consider the diagram. You can readily see that larger amounts of tilt favors warmer polar summers and colder polar winters, since more tilt points the poles more directly toward the sun during the summer, and away from the sun during the winter.

Conversely, a more vertical axis favors colder polar summers and warmer polar winters. The effect is crucial to ice age formation, so be sure you understand it.

An important feature of cold polar winters is that they are comparatively dry—by which I mean there is relatively little precipitation, which occurs mainly in the form of snow. Warm winters are comparatively wet, because a warmer atmosphere can hold more moisture than a colder one. Warm winters tend to have greater amounts of snowfall.

We now have sufficient basic understanding to explore how the planet descends into glaciated periods where ice accumulates at high latitudes and spreads to lower ones—particularly in the northern hemisphere, which has more land cover than the southern hemisphere. We will introduce additional details as we go.

As noted, the degree of the earth's axis tilt varies over a 40,000-year period. A more vertical axis favors ice accumulation, because the relatively warm winters receive more snowfall, as described above. Snow is compacted into ice to form glaciers, which are also called ice sheets.

A more vertical axis favors ice accumulation another way, too. The relatively cooler summers allow more of the winter snowfall to persist across seasons. That is, less snow and ice melts during the cooler summer, so it builds up over time.

So the first requirement for an ice age is reduced axis tilt. Once the earth straightens up sufficiently, things get started. Since the axis tilt period is 40,000 years, you might have to wait a while before conditions are right. (Period means the length of time for an entire cycle to complete—in this case, the transition from minimum to maximum tilt, and then back to minimum. So minimum tilt occurs every 40,000 years.)

Even then, the process is slow. It takes thousands of years to enter a full-blown ice age, and several crucial feedbacks are necessary to achieve it. Here's how things proceed.

Ice buildup begins with the warm winters and cool summers that come with reduced axis tilt. Once enough ice accumulates in the early going, the first important feedback kicks in. That feedback involves what scientists call albedo, which refers to the reflectivity of surfaces and bodies, especially the earth. Because snow and ice are quite reflective, much of the sunlight falling upon them is reflected back into space without causing much warming.

By contrast, bare land and open water are comparatively dark, so they absorb much more incoming solar radiation than do snow and ice, and thus promote warming.

As an aside, the albedo feedback is as important to today's unnatural anthropogenic warming as it is to the natural progression of ice ages. It's just that in a warming climate the albedo feedback is positive: melting snow and ice promote more absorption of sunlight and so more warming. In a cooling climate the albedo feedback is negative: increased ice formation causes less sunlight absorption and more cooling. Finally, albedo also explains why, in a warming climate, high latitudes warm much faster than lower latitudes: It's because they're losing their ice cover.

As an ice age gets going, the formation of ice feeds back on itself through abedo effects, and ice coverage increases. Initially this occurs at high latitudes, which means closer to the poles. But the cooling caused by increased reflectivity causes the ice to slowly spread over a larger geographic area. As you'd expect, global average temperature decreases too, thanks to less absorbed sunlight warming the earth overall.

With temperature decrease comes a second crucial feedback. Colder temperature causes atmospheric carbon dioxide concentrations to likewise decrease. (One reason is that colder oceans can hold more dissolved CO2 than warmer ones. The oceans are a huge CO2 surface reservoir, containing far more of the gas than the atmosphere.) Reductions in carbon dioxide seem to lag temperature changes by a few hundred years. In some circumstances the lag appears to be much less.

This may strike you as exactly backward. After all, according to our popular understanding of global warming, CO2 is supposed to drive climate, not respond to it. It actually does both, which is what you'd expect of a feedback. In many natural situations changes in temperature precede and cause changes in CO2. But because CO2 is a greenhouse gas, its concentration is also a cause of further temperature change. Altering CO2 concentration directly and unnaturally, such as by burning fossil fuels, affects climate just as potently as alterations that occur through natural feedback mechanisms. Ultimately, CO2 concentration is the single most important determinant of climate.

So the CO2 feedback can go in both directions, promoting both warming and cooling. When the earth descends into an ice age, colder temperatures cause less atmospheric CO2, which in turn causes even colder temperatures. Other somewhat less important greenhouse gases participate in a similar fashion during the progression into an ice age.

Because the crucial albedo and greenhouse gas feedbacks are so slow, an ice age takes thousands of years to develop. Much later, ice ages unwind as increasing axis tilt promotes the melting of ice. The feedbacks involved are the same, especially albedo and greenhouse gases, but they turn positive.

I suppose I should tell you that ice age development is a little more complicated than I've let on so far. Up to this point I've avoided mentioning another factor in the onset, intensity, and duration of ice ages, but we really ought to take a quick look at it for completeness. We will get to that now.

It turns out that axis tilt is not the only instigator of ice ages. Another weaker but important factor (factors, actually) involves the shape of the earth's orbit around the sun.

The earth traces an orbit that's ever so slightly elliptical, not perfectly circular. That means there is some small variation over time, including in a single year, in Earth's distance from the sun. The variable distance affects how much solar radiation the earth receives at any particular time. The day of the year when the earth is closest to the sun varies through the entire calendar over a period of about 20,000 years. Since warm winters and cool summers favor ice accumulation, the time of year of closest proximity to the sun can be important in nudging conditions toward or away from more ice, and reinforcing or weakening the primary effect of axis tilt.

Furthermore, the eccentricity (the amount of deviation from circular) of the earth's orbit itself varies from nearly zero (almost circular) to nearly six percent (a bit more elliptical), and does so with a non-uniform periodicity. In case you're wondering, the variation in the shape of the earth's orbit is caused by the gravitational tug of other planets in the solar system.

So the effect and timing of proximity to the sun is variable, and can amplify or diminish the effect of axis tilt depending on whether and how much all the factors are in or out of phase with each other.

Axis tilt considered alone would now be pushing the northern hemisphere toward a buildup of ice sheets (as always, over thousands of years), but in fact things are moving in the opposite direction. Despite a more vertical axis that ought to be initiating a new ice age, ice is actually melting at a furious rate, particularly in the Arctic. It's crucial that we understand why, and what it means for the future of humanity.

First, let's get a sense of duration. The most recent ice age began roughly 100,000 years ago, peaked around 20,000 years ago, and was gone by 10,000 years ago. The warm period (called an "interglacial") beginning approximately 10,000 years ago or a little earlier is named the Holocene. It's the geological epoch in which the entire history of human civilization has occurred.

What's remarkable about ice ages is how incredibly weak are the factors that cause them compared to other important drivers of climate, including (especially!) anthropogenic (human caused) drivers. That's not at all what most of us would expect. We imagine ice ages to be imponderably immense phenomena, consistent with the amazing realization that 20,000 years ago there were ice sheets 2 miles thick covering Canada. But our common intuition about them is wrong, and that misguided intuition could be leading us astray on other climate understandings as well.

How do we know this? It's important to recognize that the factors that determine the energy balance of the earth—and thus its temperature—are readily quantifiable. The physics is quite straightforward—not that we'll be doing any of it here. Later I'll direct you to where you can go for somewhat deeper looks that are still accessible to the layman, should you so desire.

Scientists quantify climate forcings—which are physical phenomena that alter the planet's equilibrium temperature to make it warmer or cooler—in units of watts per square meter of the earth's surface. To get a sense of scale, you might be interested to know that the earth averages 240 watts of incoming sunlight per square meter of surface. When the earth is in energy equilibrium, it radiates an equivalent amount out into space as heat, maintaining a constant planetary temperature. Unfortunately, the earth isn't currently in energy equilibrium.

(Fun fact: The total surface area of the earth is 510,072,000,000,000 square meters, which at 240 watts per square meter means that at any moment the earth is receiving 122,417,280,000,000,000 watts of solar radiation. That's 122 million gigawatts striking the earth each and every instant. The earth thus receives around 1.1 billion billion kilowatt-hours of solar radiation annually, which is around 38,000 times humanity's annual electricity production. None of which is directly relevant to our discussion here, so back to forcings.)

Forcings—physical phenomena that alter the planet's equilibrium temperature—can be positive or negative. Positive forcings tend to make the climate warmer, and negative forcings make it cooler. Forcings can be both natural and human-caused, and we combine their values using simple summation to yield an overall climate forcing quantity that determines (subject to "sensitivity," which we won't deal with here) the earth's equilibrium temperature.

I reproduced a list of the most important forcings, and explained them in more detail, in a previous essay entitled Of Ice Ages and Men. The strongest recent natural forcing is a slight increase in solar irradiance (brightness of the sun), which can vary a bit up or down over intermediate time scales. Natural forcings operating since the beginning of the Industrial Revolution sum to a scant 0.35 watts per square meter. This small positive value means that, over the past 250 years or so, the earth would be expected to warm just slightly due to natural causes alone.

Human-caused forcings (some positive, some negative) over that 250 years sum to 1.45. Greenhouse gases are the largest human-caused forcings. Total climate forcings are thus 1.8 watts per square meter.

This means the earth's energy balance—the balance between incoming (solar) and outgoing (heat) radiation—is currently 1.8 watts per square meter greater than what would maintain the planet at its current temperature. Simple physics requires the planet to warm until incoming and outgoing radiation are back in balance at a higher equilibrium temperature. I explain this in more detail in my previous essay.

The point I want to drive home is that the climate forcings that instigate ice ages, involving the axis tilt and orbit shape factors we've discussed here, amount to a tiny fraction of one watt per square meter—too small to be listed with the most important climate forcings that are now operating. How can that be? How can such tiny forcings result in thick ice sheets covering a large part of the planet?

The answer is that those small negative forcings, operating with and against other small natural forcings, do their work over many thousands of years, with the help of slow but relentless feedbacks that kick in along the way and amplify the initial effect. It takes shockingly little forcing, but a whole lot of time, to build an ice age.

But now those small negative natural forcings that might otherwise be presently instigating another cycle of glaciation are completely and utterly swamped by anthropogenic forcings, rendering them entirely impotent. There will be no more ice ages.

I must underscore that statement, unambiguously. The arithmetic is trivial. Ice age inducing climate forcings are a small (negative) fraction of a watt. Human caused forcings sum to 1.45 watts. Total climate forcing is presently 1.8 watts. Ice ages are no match for humanity.

A good thing, perhaps? I mean, who wants an ice age?

The problem is we are going far beyond avoiding an ice age. Humanity is on track to heat the planet to a degree not seen in millions of years, with devastating implications for human civilization and even the composition of life on Earth.

Certainly the human species has never experienced anything like what's in store in the next century or two. After all, Homo sapiens has only existed for a few hundred thousand years. Precursor species of the genus Homo only existed for the past couple of million years. The very recent, stable, and accommodating climate of the Holocene is what allowed human civilization to take hold in the first place. The climate will soon become unstable and unaccommodating.

And unlike temperature changes entering and exiting ice ages, the heating of the planet will happen fast, catastrophically so. Adaptation for human civilization and for all of life will be exceedingly difficult. Extinctions will dominate the biosphere. The lived experience of the human population will be decidedly unpleasant.

Even if you want to head off a developing ice age—and who wouldn't?—what we are now doing to the climate is extraordinary overkill. As renowned climate scientist James Hansen put it, "a single chlorofluorocarbon factory would be more than sufficient to overcome any natural tendency toward an ice age." Which underscores how scant is the climate forcing that causes them, and by implication how horrifyingly large is the anthropogenic forcing currently underway. "Ice sheets will not descend over North America and Europe again as long as we are around to stop them," Hansen says.

It turns out that with respect to climate, nature is no match for humanity. Ice ages certainly aren't. Humans now have control of the climate. What will we do with it?

---

Postscript - This essay is the result of a recent brief conversation with a climate change skeptic that lasted just a few minutes. "Ice ages" were mentioned (volcanoes, too: a subject I hope to address in the future), as often happens in these brief, vapid interactions in which each side imagines the other to be naïve and misinformed, but with polite smiles all around. The implication of the hand-wave at ice ages is that natural climactic phenomena are too immense to yield to a trifling humanity. Our effect on the climate must be negligible by comparison. Actually explaining ice ages seems a good way to refute that misguided implication.

So I went back and re-read my previous essay, Of Ice Ages and Men. I wasn't entirely happy with it. It takes too long to get to the point, with too much diversion, and too much snark. The paragraphs are too long for an online essay, although they'd be fine in a book. There were other problems too. So this latest effort is an attempt to take another crack at the subject, in a way that might actually influence my recent discussant were she to read it. I have no idea whether I've achieved the improvement I hoped for.

The original essay is still worth a read, because it includes a good bit of additional detail that I've omitted or skimmed over here. The present treatment was intended to be a little shorter and a little simpler, while still developing the ultimate crucial understanding about the weakness of natural climate forcings and the strength of anthropogenic ones. But on that score you can and should go much deeper than anything I've written.

The best book I've found on climate science is Storms of My Grandchildren, by renowned climate scientist James Hansen. Much of what I've conveyed here comes from that book.

Hansen, who might well be the most important climate scientist in existence, and one of the most prominent prophets of impending calamity, has impeccable credentials. He was director of NASA's Goddard Institute for Space Studies for three decades, and was one of the earliest scientists to offer testimony to the U.S. Congress about how humanity is altering the climate. Hansen cut his climate teeth at NASA in the 1970s studying the runaway climate of Venus (whose atmosphere is dense enough to crush you at the planet's surface, where it's hot enough to melt lead), and then moved on to studying Earth. Finally retired from NASA, he now hangs out at Columbia University.

I've read Hansen's book multiple times. Every time I return to it I'm even more impressed than before by how well and comprehensively he explains the topic. There's a lot here that's suitable for almost any reader who is willing to engage. My appreciation of the book increases every time I refer to it.

If you want to truly understand climate science, this is your book. I recommend it highly and without reservation. Get your own copy. Read it. Underline large sections and mull their meaning. Go back later to refresh your understanding. This is important and fascinating stuff, and you will find no better guide.

Copyright (C) 2018 James Michael Brennan, All Rights Reserved

The latest from Does It Hurt To Think? is here.