The World in 2050: Four Forces Shaping Civilization's Northern Future

temperature, and are meaningful many decades into the future. They do this by taking account of certain things— like deep ocean circulation and increasing greenhouse gas concentrations—that simply don’t matter for short-term weather. It’s not possible to know what the exact temperature will be in Chicago next August 14 or January 2 at three o’clock in the afternoon, but it’s very possible to know what the average August or January temperatures will be. One is weather, the other is climate.

Climate models are also amazing tools for figuring out how our complex world actually works. Suppose that it is an observed fact that summer rainfall is declining in Georgia, but this phenomenon simply won’t show up in a climate model’s simulations no matter how many times it is run. Puzzled, its programmers realize that something is missing and wonder what it might be. Into the model goes a hypothesis—say, loss of forest (trees pump enormous volumes of water vapor back to the atmosphere), because many trees have been removed to build Atlanta suburbs. Does the model now correctly simulate the measured rainfall decline? If so, congratulations—new scientific understanding has been won about how rainfall works in Georgia, and the climate model has been made more realistic. If not, on to test the next hypothesis down the list. Eventually the missing bit of physics is discovered, the model is improved, and its creators move on to ponder its next little failure.

At their core, climate modelers seek to understand how the atmosphere functions, and how it responds to changing drivers. By studying when and where the models break down, we improve scientific understanding of how the real world works, and our models become more accurate. After more than fifty years of trial and error, they have now evolved far beyond their primitive ancestors of the 1960s. We’ve learned a great deal about how Earth’s climate system actually operates. In today’s generation of models, complicated things like El Nino and the Hadley Circulation emerge organically without programmers having to “add” them at all. That is very encouraging, because it tells us the models’ assumptions and physics⁴⁸⁶ are realistic and working correctly.

The big push now is to hone down climate model spatial resolutions (i.e., the “pixel size” of their simulations) from hundreds of kilometers, useful for broad-scale projections like the ones presented in this book, to kilometers, which is what local planners need. But even at the coarser spatial scale of today’s generation of models, many important conclusions about our future are now well vetted and uncontroversial. All of the megatrends discussed so far—rising global average temperature, the amplified warming in the Arctic, rising winter precipitation around the northern high latitudes—fall within this uncontroversial category.

More troublesome are the short-sellers and inside traders of natural climatic variability. Volcanoes, wildfires, and sunspot cycles are just a few of many phenomena imprinting their own natural variations over the underlying greenhouse gas signal. But now these volatile (and fairly common) phenomena, too, are being added to climate models and tested.

Where climate models suffer most is in capturing rare events lying totally outside of our modern experience. Most weather stations are less than a century old; the satellite data era began only in the 1960s and ’70s. These records are far too short to illuminate the full range of our Earth’s twitchy behavior. Shifting oceans and ice sheets are key drivers of climate yet contain toggles and circuits with longer patience than our short instrumental records. They add boosts, buffers, and dips to the overall greenhouse effect, so we must understand them as well.

Unfortunately, a naturally twitchy climate makes the steady, predictable push from anthropogenic greenhouse gases more dangerous, not less. From the geological past we know the Earth’s climate has not always been so quiet as it is now. Therefore, through greenhouse loading we are applying a persistent pressure to a system prone to sudden jumps in ways we don’t fully understand. Imagine a wildcat quietly sleeping on your porch—it looks peaceful but is by nature an ill-tempered, unpredictable beast that might spring into a flurry of teeth and claws in an instant. Greenhouse gases are your knuckles pressing inexorably into its soft slumbering belly; the global ecosystem is your exposed hand and arm.

Rare or threshold behaviors—like a permanent reorganization of rainfall patterns, accelerated sea-level rise, or a giant burp of greenhouse gas from the ground—all pose legitimate threats to the world. We know they are plausible but, unlike greenhouse gas forcing, don’t know yet how probable. But their behaviors, too, must be added to climate models somehow. Just because something seems unlikely doesn’t mean it won’t happen, or that its impacts are not potentially enormous if it does. These are the climate genies, and we are just beginning to discern the outline of their various sleeping forms. To find them at all, we must turn to the prehistoric past.

The Flickering Switch

One of my personal heroes in science is Richard B. Alley, an outstandingly accomplished glaciologist and professor of geosciences at Penn State University. Not only has he cranked out one landmark idea after another, published nearly forty times in Science and Nature, been elected to the National Academy of Sciences, and written a wonderful popular book explaining it all for the rest of us,⁴⁸⁷ he is also about the nicest and most enthusiastic guy one could ever hope to meet.

In 1994, Alley came to deliver a guest lecture at Cornell University, where I was a lowly second-year graduate student. Everyone was abuzz that Richard Alley was coming, because he had just published a pair of back-to-back articles in Nature that had stunned the climate-science community.⁴⁸⁸ Even my thesis advisor—who was pretty famous himself, having written the paper putting together the theory of plate tectonics⁴⁸⁹—was talking about them. But a great thing about academia is that it is on open, democratic affair even when it comes to its pop icons. Visiting celebrities will hang out for a day or two happily chatting with whomever, even lowly second-year graduate students. Landing a meeting with one is largely a matter of getting to the sign-up sheet first, which of course I did.

When my time slot arrived I went to meet Alley, armed with a list of questions about his Nature papers so I could hear more from the great man himself. That lasted about forty- five seconds, before he insisted on hearing all about my work. I couldn’t believe it. It was a dumb little side project of my research, but Alley’s enthusiasm was totally contagious. We relocated to my lab hole, where he huddled alongside me, giving all manner of helpful advice and inspiration. By the time he ran off late to his next appointment, I was so excited about my project I barely remembered I’d forgotten to ask more about his. That’s just the kind of guy he is.⁴⁹⁰

What had everyone gabbling was what Alley and his colleagues had dug out of the Greenland Ice Sheet. The U.S. National Science Foundation had funded construction of a drilling and laboratory camp on top of it to extract a two-mile-long ice core called GISP2, an enormous task taking about four years.⁴⁹¹ Preserved in the upper sections of ice cores are annual layers, like the rings of a tree. Each one contains the compressed equivalent of a full year’s worth of snow accumulation falling on the ice sheet surface (cores are drilled from deep ice sheet interiors where it never melts). By counting the layers down-core and measuring their thickness and chemistry, a very long reconstruction of past climate variations is obtained. We even get tiny samples of the ancient atmosphere, by cracking into air bubbles trapped in the ice. From these high-resolution annual measurements in Greenland, Alley and his colleagues had discovered that around twelve thousand years ago, just when we were pulling out of the last ice age, the climate began shuddering wildly.

The shudders happened faster than anyone had dreamed possible. Our climatic emergence from the last ice age, it seems, was neither gradual nor smooth. Instead it underwent rapid flip-flops, seesawing back and forth between glacial and interglacial (warm) temperatures several times before finally settling down into a warmer state. These large temperature swings happened in less than a decade and as quickly as three years. Precipitation doubled in as little as a single year. Around Greenland, at least, there was no gradual, smooth transition from a cold ice age to the balmy interglacial period of today. Alley’s team had shown that climate could sometimes teeter as well, like a “flickering switch,” between two very different states. Furthermore, it had happened other times in earlier millennia, so this was not a totally isolated event. The extreme rapidity of these changes, concluded Alley, implied “some kind of threshold or trigger in the North Atlantic climate system.”⁴⁹²

Thus was born a brand-new subfield of climate science known today as “abrupt climate change.” Twenty years ago anyone who hypothesized a sudden, showstopping event—a century-long drought, a rapid temperature climb, or the fast die-off of forests—would have been laughed off. But today a growing body of evidence from ice cores, tree rings, ocean sediments, and other natural archives tells that such things have happened in the past. We’ve long known the Earth’s climate has experienced big changes before but assumed they only occurred slowly over geological time, like the gradual turning of a dial. Now we know they can sometimes happen abruptly as well,