Deep Future

by Curt Stager

  Forams are especially useful to paleo-oceanographers because they contain oxygen-18, a heavy, stable isotopic form of oxygen atom. The ratio of that isotope to the more abundant oxygen-16 atoms in a foram’s shell reflects the temperature of the waters in which the little creature lived. In that sense, oxygen isotopes can serve as paleothermometers, though their accuracy is blunted when ice sheets selectively trap normal oxygen-16 and hold tons of it away from the oceans. Fortunately for our purposes, the early Cenozoic was largely ice-free, so the oxygen thermometer reads more clearly than it does for samples deposited during recent ice ages; temperature estimates that predate the buildup of ice in Antarctica roughly 34 million years ago are therefore more reliable than the ones that follow it.

  Foram data from Pacific cores suggest that tropical sea-surface temperatures jumped as much as 5°F (3°C) above their already-warm state during the opening centuries of the PETM; other temperature indicators in those same cores and from Indian Ocean sediments double that jump. But even more dramatic are the results gleaned from fossil forams of the polar seas. They show that the surface waters of the Arctic and Antarctic warmed by 14 to 18°F (8 to 10°C) within a few thousand years. One estimate puts sea-surface temperatures at the North Pole in the 70s Fahrenheit (23 to 24°C) or even higher. In other words, a time-traveling visitor to the early Eocene could swim in a perpetually ice-free Arctic Ocean with barely a shiver.

  Deep-sea oxygen isotopes and temperatures throughout the Cenozoic Era, showing early warm conditions followed by long-term global cooling. After Zachos et al., 2008

  Sediment cores lifted from the sea floor near the North Pole by Yale geoscientist Mark Pagani and colleagues also contain substances, called “tetraether lipids,” that were once embedded in the oily cellular membranes of floating plankton. Like melted butter after you put it in the fridge, membrane oils can stiffen if the surrounding habitat cools off and they can turn runny if it warms too much. Certain kinds of tetraether lipids help cells to resist those harmful changes by keeping membranes at optimal fluidity as local temperatures vary. By analyzing the lipid contents of polar ocean cores, Pagani’s team found that Arctic sea-surface temperatures ranged from 65 to 75 degrees (18 to 24°C) during the PETM, thus independently supporting the foram data.

  We don’t have a dense, worldwide array of ancient weather stations to tell us exactly what global mean temperatures were during that thermal maximum; as in the case of the Eemian, we have to rely on a relatively small number of reference points. We do know that the tropical seas warmed quite a bit along with the polar regions; according to geochemical data from cores drilled off the coast of Tanzania, temperatures there were as high as 95 to 104°F (35 to 40°C). As best we can tell from widely scattered geological study sites, the PETM heated the planet as a whole by something close to 10°F (5 to 6°C) within several thousand years, and it apparently displayed much the same latitudinal asymmetry that we’re experiencing now, with the poles warming more than the tropics. Today, we attribute this to an extra thermal kick as reflective snow and ice give way to darker, heat-absorbing surfaces, but we have no direct evidence of extensive polar ice back then. So why the enhanced polar warming? Perhaps it means that some snow and sea ice still formed in winter. If it kept some locations unusually cool by reflecting sunlight away, then losing those last reflective surfaces when the PETM began might explain the extra jolt of polar warming.

  With no gas bubbles from ice cores to guide us that far back in time, climate scientists face major challenges as they try to figure out what produced such warming in a human-free world. But all investigators agree that carbon-based gases, that is, CO2 and methane, were central players, just as they are today. The evidence for this rings strong and clear in PETM-age sedimentary deposits: a worldwide dip in the abundance of carbon-13, a rare version of carbon that’s found in the bodies and waste products of all living things. Only a colossal intrusion of CO2, methane, or some combination of the two into the atmosphere and oceans could have caused it.

  Not only do geoscientists use this distinctive “carbon isotope excursion” to infer the causes of the PETM; they also use it to identify the warm interval itself in rocks and sediments and even in the teeth of extinct animals and the leafy remains of early flowering plants. The carbon-13 dip is a remarkably informative chemical label that allows you to match the ages of samples from one continent to another more precisely than radiometric dating methods would allow, thereby confirming that extreme environmental changes occurred synchronously all over the planet.

  Here’s how it works. Every living thing is full of carbon, a tiny fraction of which is carbon-13 (also written as 13C). That’s because plants inhale 13C-contaminated carbon dioxide from the air around them, although they avoid doing so as much as possible. Because that selectivity is imperfect, plants do contain traces of unwanted 13C, but not nearly as much as you’d find in a comparable mass of molecules drifting about in the open air. The same situation holds true for animals, ourselves included, because grazers eat plants and thereby recycle their carbon through the world’s food webs. Because most living bodies are thus well scrubbed of 13C, an unusually low concentration in a geological deposit means that it probably contains the remains of long-dead organisms.

  One of the first scientists to recognize the PETM’s distinctive 13C signature in a sediment record was Lowell Stott, now a paleo-oceanographer at the University of Southern California. As a graduate student back in the late 1980s, Stott was studying South Atlantic cores and wondering why the 13C values in his ancient forams were coming out so low.

  “I got these bizarre numbers that didn’t make sense to me,” he told me recently. “I thought I might have made a mistake, so I reran the samples, but I still got the same result. Then I tried another species of foram and got the same result again.” His advisor, James Kennett, didn’t know what to make of it, either. It was both the dread and the dream of any graduate student—a finding that’s difficult to explain but too important to ignore. “I was naive enough to be excited by it, though,” Stott chuckled. “Nobody had ever seen this kind of thing before.” Three years later, Kennett and Stott published the discovery in Nature.

  In follow-up studies by others, similar drops in 13C values were found in PETM-age deposits from all the major ocean basins as well as on land. And the most reasonable explanation for that global 13C depletion was a massive release of organic fumes, either CO2 or methane, from decaying peat or other carbon-rich deposits. But which gas was it, where was it hiding before the PETM, and why did it emerge so suddenly?

  If CO2 was the culprit, then it should also have acidified the oceans because it produces carbonic acid when it dissolves in water. Marine sediment cores are well suited to resolve such a question, in addition to preserving the 13C excursion as a convenient time marker. A typical deep-sea core is a cylinder of moist gray to brown mud that can sometimes be so full of chalky carbonate particles that it fizzes on contact with strong vinegar. But the stuff deposited during the PETM forms a strikingly obvious reddish band ranging from a few inches to a foot or more thick because the pale carbonates have been eaten away, leaving behind a rusty claylike residue.

  The transition across the base of the red zone is abrupt, indicating a sudden acidification of the deep sea, but the return to normal carbonate deposition was gradual, lasting between 50,000 and 200,000 years. This pattern would be consistent with a dramatic CO2 rise followed by a long slow drawdown, much like the extreme 5,000-Gton carbon emissions scenario that we’re hoping to avoid now. In the context of our current situation, it presents a stern warning; this sort of thing really can happen to our future Earth because, quite clearly, it has already happened before.

  One reasonable explanation for the initial CO2 release stems from seemingly unrelated work on the geological history of the North Atlantic. Great crustal cracks along the midline of that widening ocean basin have been pushing North America and Europe farther and farther apart throughout the Cenozoic era, but during
the late Paleocene an unusually active spreading zone opened up between Greenland and what would later become Scandinavia. For hundreds of thousands of years, huge floods of glowing lava burst from the submarine cracks and seared their way into carbon-rich sedimentary deposits. If highly organic or limy materials on the deep-sea floor were burned in this manner, then they would have released CO2 just as our fossil fuel combustion does, and because they contained the remains of dead marine organisms the resultant vapors would also have been depleted in 13C. If enough such gases were released into the atmosphere and oceans, then they could have caused the greenhouse warming as well as the global decline of 13C concentrations.

  But what if the PETM’s main pollutant gas were methane instead of CO2? Methane rapidly oxidizes into CO2 in the atmosphere, so a methane burst might indirectly cause carbonic acid pollution of the oceans as well. The presumed source? Microbially generated methane ice that accumulates in certain kinds of wet sediments, especially on or near marine continental shelves. Also called “clathrates,” these odd substances form when sediment-dwelling bacteria release methane as metabolic waste. Under the right combination of cool temperature and high pressure, bacterial methane can become trapped in tiny cages made of loosely linked water molecules, forming delicate crystal lattices that resemble an unstable version of dry ice. Haul a mud-caked chunk of white methane ice up to the sea surface and put a match to it, and it burns like a candle.

  Methane ices are so unstable that any number of triggers, from sea-level changes to climatic warming or volcanism, could presumably have launched a carbon-rich gas assault on the PETM atmosphere. If one or two thousand gigatons of methane emerged over a short time period, 13C concentrations around the world would drop more steeply than if an equal volume of CO2 escaped. Bacterial methane contains even less 13C than most other biological substances do, so it would take less of it to produce the abrupt 13C excursion than if CO2 were the only operator.

  James Kennett has used this hypothetical mechanism, sometimes called a “clathrate gun,” to explain not only the PETM greenhouse but also many of the noteworthy climatic swings of the last hundred million years. The idea is fairly simple. Clathrates build up in oceanic muds, peatlands, and permafrost like a growing arsenal of air rifles loaded with methane charges. Occasionally, the gases burst from their barrels to trigger a global warming until natural processes eventually consume or rebury them.

  The clathrate-gun hypothesis has gained favor among those who expect our own greenhouse to cause a similar blowout in the future. But there are problems with it, too. A 1,000 to 2,000-Gton methane pulse could explain the global 13C dilution, but to some investigators the magnitudes of the warming and 13C excursion together seem to be more consistent with a larger, 5,000-Gton slug of CO2. In addition, David Archer suggests that most of the methane ices in place today are too diffuse and well insulated by thick blankets of sediment to respond suddenly to most environmental changes. If he’s right, then the clathrate gun has a safety catch on it. Archer envisions only gradual releases spread over thousands of years, more of a squirt gun dribble than a blast.

  Whatever really caused that initial gas spike, it almost certainly set off a cascade of follow-up releases; this could very well happen during a 5,000-Gton emission of our own making, too. This kind of additive, self-amplifying process is referred to by specialists as a positive feedback loop.

  Biomechanics expert Steve Vogel, one of my mentors at Duke University, used to describe such feedback loops in a way that caught the attention of all but the most inattentive of his undergraduate nonmajors. “Imagine lying in bed next to the partner of your dreams,” he’d say, savoring the sudden lifting of drowsy heads. “You’ve each got an electric blanket to cover you in the cold, unheated room, and you each have a temperature control knob in your hand. Unfortunately, however, in the darkness and distractions of the moment, each of you has grabbed the other’s control knob.”

  Once the expected chuckles had subsided, he would continue. “Now what happens as the night wears on and the room grows cooler? You feel chilly so you crank the knob up a bit. But that only warms your partner’s blanket, and when they feel too warm they turn their own knob down a bit more. That makes you feel even colder, so you turn your knob to an even higher setting, and around and around it goes. Pretty soon, you’re dying of cold while your partner is dying of heat. That’s what we call a positive feedback loop, not because it’s positive in a good way but because its additive effects grow and grow in a self-amplifying cycle.”

  Applying that memorable concept to the planet, we see that pushing temperatures up too far can unleash positive feedbacks that keep cranking the global thermostat higher and higher. From that point on, the warmer it gets, the more it stimulates processes that release even more heat-trapping gases.

  In the case of global warming, there is a long list of things that could become participants in such feedbacks. Warmer water holds less gas in solution than cold water does, so marine heating can squeeze more CO2 out of the oceans and into the air. A hotter atmosphere also becomes more humid because it sucks moisture from the soils, lakes, and seas beneath it, and water vapor is a major greenhouse gas in its own right. Heating speeds the gas-producing decay of organic matter in wetlands and thawing permafrost, and it can also destabilize frozen methane. By some estimates, the carbon content of today’s clathrate methane inventory may approach that of all other fossil fuels combined, so the PETM example makes a strong case against pushing future temperatures any higher than we have to.

  And what was the world like in that superhothouse? For one thing, the PETM finished off whatever land-based ice may have persisted near the poles through the already-warm Paleocene epoch. With sea temperatures in the far north in the 70s Fahrenheit, the Arctic Ocean became a tepid, brackish lake. This was a world without ice caps or extensive glaciers; if any snow fell at all, perhaps on the highest circumpolar mountains in the long darkness of midwinter, most or all of it would probably have melted during the warmer, sunnier months of the year.

  To tease those insights from ancient marine oozes, paleo-oceanographers have focused less on forams than on the remains of tiny planktonic algae. Unlike forams, those free-floating microbes were more like miniature single-celled plants. Being photosynthetic they needed lots of sunlight in order to grow, more than a thick cap of sea ice could have transmitted, so their abundance in sediment core samples from that time period suggests open-water conditions. Furthermore, modern versions of those algae now live in mildly brackish or freshwater habitats, so they also suggest that the polar ocean was diluted by unfrozen, freely running rivers. The additional confining presence of a land bridge across the Bering Strait would have surrounded that low-salinity ocean by land on all but the North Atlantic side of the basin, thereby allowing even more runoff from the encircling continents to dilute it further. Together, these findings evoke an open-water body that resembled the nearly landlocked Baltic Sea of today: saltiest near the Atlantic outlet and least saline at the far inland end.

  The South Pole was a very different place back then, as well. There’s no firm evidence of large ice sheets anywhere on Earth during the PETM, and no telltale pebbles were sprinkled by drifting, dirt-caked icebergs into the marine muds off the Antarctic coast. Instead, distinctive clay minerals in those muds tell of deep weathering and erosion of soils by warm, wet climates. Kaolinite clay, for example, doesn’t often form near the poles today, but it’s common in hot, rainy places such as the tropical Niger Delta, and it evidently clouded the southernmost seas 55 million years ago. If ice did exist on Antarctica at all, then it was probably restricted to the highest peaks of the interior.

  The deepest layers of the oceans, now among the world’s coldest waters, warmed by 7°F or more (4°C) and killed many of the cold-loving creatures that had been living on the bottom. Some investigators attribute this to heating at one or both of the poles because icy temperatures there normally make dense, oxygen-rich seawater sink and spread across the
floors of the ocean basins, and they suspect that high PETM temperatures might have prevented that flow. Others believe that heavy rains capped the circumpolar seas with buoyant, low-salinity layers that slowed or stopped the sinking of cool, well-oxygenated surface water.

  Whatever its cause, a choking off of life-giving bottom currents apparently smothered deep-sea communities with warm, stagnant, oxygen-poor water. Up to half of all bottom-dwelling foram species vanished from the fossil record, as did other organisms that lived in similarly deep habitats. In contrast, most surface dwellers survived the PETM with no apparent difficulty. But that makes sense, because floaters near the sea surface were probably used to warmer temperatures anyway, and because oxygen could have reached them from the overlying air and from photosynthetic algae floating among them.

  As if sweltering and suffocating weren’t enough of a burden for deep-ocean life, the water also became corrosive enough to burn that aforementioned red clay band into the sediment record. As we await the life-or-death sentences of many marine species that face carbonic acidification in today’s oceans, the PETM example may be painfully instructive. If we go down a similar path during the Anthropocene, will everything from clams to corals vanish with a fizz?

  The good news is that many species somehow came through the ordeal intact, especially if they lived in shallow habitats. Some forams died out or developed thinner shells, but others thrived. Oysters and many other mollusks survived, as did quite a few corals. Some chalky planktonic algae even managed to gain shell weight rather than lose it to dissolution. Overall, extinctions in the upper layers of the oceans were largely balanced by the appearance of new species, often with little indication that acidity was the cause of the die-offs. We don’t know how so many shell-bearers resisted the acid bath, but it’s clear that they did. The bad news is that few, if any, of our modern species existed in their present form 55 million years ago, so we can’t say that our marine neighbors have ever survived such severe environmental changes before.