THE SHORT DROP-OFF, PALAU, JULY 1983
A pair of divers slowly rose up the side of the blue wall, into the zone of living coral and red colors now, into schools of fish. The two men—I was one of them—floated upward no faster than their slowest bubbles, short inhales and long exhales from the rubber mouthpieces clutched by experienced mouths. Each of us had one arm stretched upward, making us look like two eager schoolboys trying to provide a well-known answer to some question, but in this case we each also had a hand gripped on the flexible hose extending from our buoyancy compensators, for our ascent from well below 130 feet demanded a bleeding of the buoyancy compensators, the gas in our lungs, gas in our tanks, gas in our blood, and, most dangerous, gas in our flotation devices expanded from the lowering pressure dictated by Boyle’s law.
At 40 feet we came to the top of the sheer coral wall, confronted by the enormous expanse of the reef front and reef top itself, a warm, sunny, multicolored universe of unbelievable diversity, of unbelievable abundance, the marine equivalent of a rain forest, but a delicate rain forest, one balanced on narrow ranges of heat and oxygen. Strong leg kicks powered us over the edge of the wall into this vast summery habitat, into ever shallower crystalline water, the anchor line of our boat now brightly visible, a yellow line pointing up toward the world of air we would soon return to. Soon, but not yet. Surfacing now would be tempting fate, physics, and physiology, which demanded that two decompression stops be made, for a couple of minutes at 20 feet, and for a good 10 minutes at 10 feet.
Our dive had been so deep to supposedly find deeper water fish for the American continental aquarium that my diving partner, Michael Weekley, worked for; the reality was that this was a pleasure dive, at a beautiful reef wall somewhat egregiously misnamed the Short Drop-Off. There was nothing short about this reef’s drop, with depths exceeding 2,000 feet only a few hundred yards from the top reef’s breaker zone. This wall was also the closest to our month-long base and sometime home: the Palau Mariculture Demonstration Center in Koror, the jewel of Micronesia and site of some of the most luxuriant and pristine coral reefs in the entire Indo-Pacific region, the vast swath of the tropical Pacific Ocean that is the diversity center of coral, mollusks, and tropical fish on our planet.
Our trip as a whole was about more than collecting fish for aquariums. We had come to Palau to further unravel the mystery of cephalopod biology and paleobiology, including that most interesting question concerning the relative fates of the two great stocks of externally shelled cephalopods, the ammonites and nautiluses. The ammonites had survived many mass extinctions, even if wounded. Through the end of the Permian period, end of the Triassic period, and through the several other extinctions during the Mesozoic era, they survived to flourish in ever-greater numbers following the events. Not so during the Cretaceous period. The ammonites died, quickly, while the nautiloids had survived, and lived still, in large numbers in that water. But the most curious aspect of it all was that the nautiloids seemed to show a very different pattern of survival compared with the ammonites at each of the mass extinctions—at the end of the Permian and Triassic, more nautiloids died out than did ammonites, but at the end of the Cretaceous it was the ammonites that went bust—and we scientists thought we might be discovering why. Somehow the surface waters were as lethal at the end of the Cretaceous period as the deeper waters had been during the other mass extinctions.
Lethal? Well, not exactly these waters, not today, I mused, holding on to the rope with clenched knees, keeping my depth at the 20-foot mark on my depth gauge, watching the seconds tick by on my watch, hoping all the nitrogen molecules that had been forced to dissolve into my blood by the deep dive were now making their way out of solution and back into gas in my lungs, and not into my bloodstream, liver, brain, or spinal column. That gas could cause death, hideous death, simply by changing depth was a curious thought, and although there is nothing more boring than hanging on a line, there was no going up to that nearby boat—we were unable to surface even if the Loch Ness Monster were hungrily circling (or, more to the point, if a large tiger shark were circling, which had happened to me while decompressing after a deep night dive photographing wild nautiluses in New Caledonia the year before).
Two minutes were up, and we moved even shallower, into the blood-warm 10-foot-deep water, the boat bottom quite close now, things looking closer in water than they were. Ammonites and nautiloids—one stock lived and one died—and I was having vague ideas about why this curious pattern had occurred. At the time the Alvarez impact hypothesis was the hottest science on the planet, and a major aspect of the new research dealt with finding out which organisms survived and which perished, and why. My trip to Zumaya the year before had put me into the middle of this controversy. Some of the paleontological silverbacks of the time espoused simple chance—it was not bad genes, simply bad luck. There had to be a reason that the nautiloids made it, to survive right up to now, as evidenced by the huge populations of them here in Palau and elsewhere in the South Pacific, and I suspected that I knew.
Over the course of the past four weeks we had pulled off a wonderful scientific feat, or so we congratulated ourselves in the local bar afterward, drinking the only beer then available, Old Milwaukee. Four of us had come to Palau armed with miniature transmitters that could relay information on the motion and depth of any animal tagged with them—if one could stay in range of the battery-powered transmitter, that is, and that range was less than half a mile in the best of circumstances. We had caught a large cage full of nautiluses the first week of our stay, and we had tagged four of them by permanently attaching the small transmitters to their shells.
After that came a nightmare time—tracking meant sitting over the deep-living nautiluses in a small boat, and the only boat that was safe enough to move around the treacherous reefs at night was an open Boston Whaler with a big outboard engine. Even marking the dangerous reefs each sunset with glowing Cyalume-filled markers only partially alleviated the chance of running onto the wave-swept reef tops in the dark of the tropical nights.
The tags on each of the four nautiluses gave out continuous information, with a strain gauge causing their sonic beeps to increase in duration with ever-greater depth. But there was no automated recording of this data; instead, every 15 to 30 minutes, a hydrophone was dangled over the side of the boat, a measurement of the depth was taken, and a precise location was noted on the nautical chart of the region.
We took turns sleeping in the bottom of the boat, huddled together under a tarp during the frequent rain squalls coming out of the endless Pacific Ocean, took turns running the even smaller lifeboat to shore to pick up more sandwiches, took turns climbing into the sea to urinate and defecate, took turns slathering endless quantities of sunscreen onto cracking tanned skin, took turns telling our life stories or dissecting our relationships with girlfriends and wives yet again.
Seven days and seven nights, we followed one particular tagged nautilus, getting seven days and seven nights of data for that one and at least three days’ and nights’ data for the other three. From these four nautiluses we found that each day these cephalopods dived away from the sun’s light and spent the day in slow motion in the darker depths of up to 1,500 feet. But each sunset, as night fell with tropical swiftness, they swam inward, up the contour toward the reef, never into water shallow enough for a human but far shallower than their daytime habitats, moving and feeding at 400 to 500 feet, always on the bottom. It was clear that these animals were part of the deepwater fauna, not members of the sunlit world.
Depth. Was that the reason for the survivability of the nautilus during the Cretaceous crisis? It was already known that the nautiluses lay large but few eggs, and that these eggs take a year to hatch. Through the use of oxygen isotopes it had been found that the hatching depths were more than 700 feet. Ammonites, on the other hand, seemed to have been dwellers of far shallower water. If the ancient nautiloids were similar in habit to their still-living counterparts, the puzzle at least had all of its pieces, if not in their right places. At the end of the Permian and Triassic periods, the deepwater animals fared worse, while the shallower forms did better. At the end of the Cretaceous period, however, it seemed that just the reverse held. The shallow-water fauna was almost exterminated, plankton as well as animals, but the deepwater forms—the diatoms and the nautiloids—came through unscathed. Those researchers studying the effects of asteroid impact, work catalyzed by the Alvarez hypothesis, came to the conclusion that the surface of the sea down to 100 feet would have been lethal to most of its inhabitants, owing to a combination of high acidity and toxins falling from the sky after the titanic impact. The ammonites lived up there, bred there, and at the end of the Cretaceous, died up there. Yet in the other mass extinctions, it was as if just the opposite held true: In those, such as the Permian and Triassic events, it was the deep that was more lethal than the shallows.
Five minutes left on the line at 10 feet, which meant only five minutes more to visit this unbelievably beautiful place, for this was the last dive, the last day, and after it was over, it would be time to break camp and load up for the long trip home scheduled for the next day. I reflected on the day before, another kind of dive, one in a place very different from this one. Our team had visited a large, baking, and stinking freshwater lake in the interior of the island, one famous for the untold numbers of jellyfish that floated in the crystal-clear surface waters of the lake. But we found that the water filled with jellyfish was but a thin stratum atop a very different water mass. Below was a place with no animals, for it had no oxygen. What it did have was a deep purple color, and rising from this deeper layer of far more primitive life were small bubbles of a toxic gas: hydrogen sulfide. The bacteria were of two kinds, and both used sulfur in their system. Both needed sunlight as well, but they could not live in oxygenated waters. One was purple in color, and it was this species that lent the highly distinctive purple color to the lake’s bottom water. Amid these were green bacteria, and these too were metabolizing sulfur.
But a third kind of bacteria was here as well, made up perhaps of several species, invisible to the naked eye. In their cells they produced hydrogen sulfide as a waste product of their metabolism. Only the thin layer of oxygen-laden water kept them from coming to the surface, where, if they could but get there, they would receive more light, grow faster, and release poison directly into the atmosphere.
We did not know it at the time, but in visiting this lake we had visited what would be recognized as the best modern analog of a hypothesized ancient ocean state that would be named a Canfield ocean, after geologist Don Canfield, who, with his mentor Robert Berner of Yale University, discovered evidence that Earth’s oceans, long before the rise of animals, were chemically and biologically different from the oceans of today and were highly toxic, saturated by hydrogen sulfide. Our ocean, saturated with oxygen from top to bottom, is chemically far different, and far more benign, certainly to us animals, and even to most microbes. I had no idea at the time that those strange lakes would help answer that nagging question about the different fates of the ammonites and nautiloids, and certainly none that they would radically alter our understanding of mass extinctions. That understanding was still nearly three decades in the future. Until then, it was impacts all the way down.
Time was up for us two divers hanging on the anchor line. I remember taking one last look at the glorious reef around me. It was good, at this sublimely happy and peaceful moment, that I could not see into the future as well as I could see into the distant corners of this reef in such clear water, or even into the far reaches of time encapsulated in the sedimentary rocks I also studied.
I gave the thumbs-up to Mike, a man fated to drown almost exactly a year to the day after this dive, on a fine July morning in New Caledonia, and then have his lungs and heart popped by the remaining and expanding gases locked in his chest as I pulled him up from deep to shallow water in a rescue attempt. It would turn me away from studying the modern, and away from the sea, toward the landward study of darker things, the study of the mass extinctions themselves, for what better way to understand unexpected, unexplained death than to take its measure in its most sepulchral form?
And it was not just we that were doomed, each in our own way; even the ancient and vigorous Palauan reef around us was in its last years of life: In the early 1990s a large mass of warm, low-oxygen water would rise from the depths and kill all the corals of the Short Drop-Off, even those in the shallowest water. The lethal deepwater was very warm, that warmth having been generated by Earth’s global warming. Today, like so many reefs around the world, the once thriving reef community at Palau’s Short Drop-Off is a cemetery ultimately caused by anthropogenic carbon dioxide, a victim of what came to be known as coral bleaching, thanks to the washed-out colors it and other reefs would develop as they succumbed to water too warm. It would be one of the first shots of an oncoming greenhouse extinction, if my colleagues and I have correctly interpreted the clues from the past. The time for studying the nautiluses came and went, another decade passed, and with increasing heat the reefs began to die. Something Wicked This Way Comes, to steal a phrase from Ray Bradbury.
BY 2005, IT WAS PRETTY CLEAR THAT THE GEOLOGICAL AND BIOLOGICAL detectives knew what did not cause the Paleocene, Triassic, and most important, the immense Permian extinctions: asteroids from space. But eliminating impact as an extinction’s cause (and at the same time, ripping the heart out of the now well-entrenched paradigm that impact had been the cause of most, if not all, mass extinctions) led to the very unsatisfactory state of not having the culprit in hand. If not impact, what? No one was going back to the twentieth-century saw about “slow climate change.” How could slowly changing climate kill so many species? Likewise for flood basalts like the Palisades—even if they seemed the only viable alternative to impact, no one knew how they could kill anything. Although it was clear that the great flood basalts would have made Earth’s air rich in carbon dioxide and thus would have led to rapid global warming, no one had been able to reconcile the effect—vast numbers of species killed—with the purported cause. Everyone assumed that if it became warmer, species would adapt by simply migrating pole-ward, for the increase in atmospheric heat from the volcanically produced carbon dioxide would have been on millennial or greater timescales, and such slow change—even if it was caused by enormous volcanoes—was just not a reasonable cause for a 90 percent death rate, such as Earth had suffered in the Permian extinction.
Some new ideas were needed. Happily, the time between the fall of the impact paradigm and the rise of its successor was not long.
Microbiologists studying the bacteria found in the jellyfish lakes of Palau and other similar kinds of anoxic lakes soon made a surprising discovery: that the varieties of bacteria in the waters left records of their presence in sediment. Microbiologists discovered that other organisms, and not just the peculiar microbes living in the Palauan lakes, left distinctive evidence of their presence too. Green plants using photosynthesis leave behind several distinct kinds of compounds, as do various kinds of microbes from other kinds of environments. A new kind of fossil was discovered.
Rather than looking for body fossils, microbiologists studying these strange, low-oxygen sites began to extract organic residues from the strata at the bottoms of their sampling sites, or even in the water itself, in search of chemical fossils, which are known as biomarkers. Other microbiologists, by studying modern organisms, figured out which biomarker came from which microbe. These biomarkers can serve as evidence of long-dead life forms that usually did not leave any skeletal fossils. Various kinds of microbes, for example, leave behind traces of the distinctive lipids, or fatty molecules, present in their cell membranes.
This biomarker research was first conducted on rocks predating the history of animals and plants, in part to determine when and under what conditions life first emerged on Earth. But within the past few years, scientists began sampling the mass-extinction boundaries. Using new kinds of mass spectrographs known as gas chromatography mass spectrometers, with skill and luck, investigators can tease out and identify what was there. Of greatest interest to the extinction detectives were the microbes living in water that was high in light, low in oxygen, and, to their surprise, high in hydrogen sulfide.
One such organism is a tiny species known as a photosynthetic purple bacterium. Today we can find such microbes in the Black Sea as well as lakes such as that in Palau. For energy they take up sulfur compounds—particularly hydrogen sulfide—and oxidize it. These microbes would be found only if other, more noxious characters were present as well—the bacteria that produce the hydrogen sulfide. Anyone who has taken freshman chemistry labs before the gas was banned from such teaching activities will remember how nasty and toxic the stuff is. Because of this extreme toxicity, most life avoids it. However, one large group of microbes is the exception to this. First near Australia, and then from numerous latest Permian-age strata from around the globe, it was confirmed that in case after case there was biomarker evidence of two kinds of microbes that inhabit water that must be low in oxygen but high in light and hydrogen sulfide. The light connection indicates that these were shallow waters, not the deep sea. It leads to a horrifying new view of the deep past, and to the tent pole to hold aloft a new paradigm for mass extinctions.
A team from Pennsylvania State University put the various pieces together. Lee Kump, one of the world’s foremost experts on the chemistry of the oceans and especially its carbon cycle, along with his longtime colleague Mike Arthur (also of Penn State) and Alexander Pavlov (of the University of Colorado), published a bombshell paper in mid- 2005 suggesting not only that there were great numbers of the nasty sulfur bacteria near the end of the Permian but also that the hydrogen sulfide that they produced was involved in the extinctions both on land and in the sea.
Only under unusual circumstances, such as those that exist in the Black Sea, do anoxic conditions below the surface permit a wide variety of oxygen-hating organisms to thrive in the water column. Those deep-dwelling anaerobic microbes churn out copious amounts of hydrogen sulfide, which dissolves into the seawater. As its concentration builds, the gas diffuses upward, where it encounters oxygen diffusing downward. So long as their balance remains undisturbed, the oxygenated and hydrogen sulfide–saturated waters remain separate, with a stable interface known as the chemocline. Typically the purple and green sulfur bacteria live in that chemocline, enjoying the supply of hydrogen sulfide from below and sunlight from above. Yet if oxygen levels drop in the oceans, conditions begin to favor the deep-sea anaerobic bacteria, which proliferate and produce greater quantities of hydrogen sulfide. In Kump and Arthur’s models, if the deepwater hydrogen sulfide concentrations were to increase beyond some critical threshold, perhaps 200 parts per million, during such an interval of oceanic anoxia, then the chemocline separating the hydrogen sulfide–rich deepwater from oxygenated surface water could have moved up to the top abruptly.
So: If deepwater hydrogen sulfide concentrations increased beyond a critical threshold during oceanic anoxic intervals (times when the ocean bottom, and perhaps even its surface regions, lose oxygen), then the chemocline (such as those in the modern Black Sea) separating sulfur-rich deep waters from oxygenated surface waters could have risen abruptly to the ocean surface. The horrific result would be great bubbles of highly poisonous hydrogen sulfide gas rising into the atmosphere. This new entry into planetary killing can be referred to as the Kump hypothesis.
The proposal is relevant to more than just the end of the Permian; the same process may have occurred at other times in Earth’s history and thus might have been the dominant cause of mass extinctions. Kump and his team did some rough calculations and were astounded to conclude that the amount of hydrogen sulfide gas entering the late Permian atmosphere would be more than 2,000 times greater than the small amount emitted by volcanoes today. Most likely, enough would have entered the atmosphere to be toxic. Moreover, the ozone shield, a layer that protects life from dangerous levels of ultraviolet rays, also would have been destroyed. Indeed, there is evidence that this happened at the end of the Permian period, for fossil spores from the extinction interval in Greenland sediments show evidence of being damaged by ultraviolet light, just the kind of damage expected from the loss of the ozone layer. Today we see various holes in the atmosphere, and under them, especially in the Antarctic, the biomass of phytoplankton rapidly decreases. (In fact, in late 2006, the hole over Antarctica was the largest ever observed.) If the base of the food chain is destroyed, it is not long until the organisms higher up suffer as well. (The complete loss of our ozone layer has even been invoked as a way to have caused a major mass extinction if Earth had been hit by particles from a nearby supernova, which also would have destroyed the ozone layer.)
Finally, the emergence of hydrogen sulfide from the seas would have coincided with an abrupt increase in both carbon dioxide and methane concentrations coming from the bottoms of the ocean that would have significantly amplified greenhouse warming from carbon dioxide pouring out of the eruptions—one of the largest in the history of the planet—that built the Siberian Traps. Hydrogen sulfide becomes more lethal as temperature rises, demonstrated in hideous lab experiments by physiologists long before Kump and his crew zeroed in on this poison as an extinction mechanism in which various animals and plants were exposed to hydrogen sulfide in closed chambers under conditions of ever-increasing temperature.
Kump’s group undertook the difficult job of looking at the potential distribution of hydrogen sulfide emission around the globe. For this they used something called a global circulation model, or GCM. These models were originally developed to understand modern weather and climate patterns, but because the positions of the continents, as well as temperature, oxygen, and carbon dioxide levels in the atmosphere and oceans, are known for the critical period at the end of the Permian period and into the Triassic period, the method could be applied to the Permian. Lastly, Kump and his team looked for areas that would have seen high erosion rates for phosphorus-bearing minerals. Phosphorus is a prime component of fertilizer, and the sulfur microbes would have thrived if there had been an abundance of it; if oceanic phosphorus levels were observed to rapidly rise at the end of the Permian, the amount of hydrogen sulfide in the oceans and atmosphere would have jumped too. Because the level of the sea dropped at the end of the Permian, there would be vast regions with trapped phosphorus that had been underwater but that now eroded under rainfall and wind into the oceans, fertilizing them. Identifying them was tantamount to identifying the sources of hydrogen sulfide.
WASHINGTON, D.C., MARCH 2006
By 2006 the Kump hypothesis was enjoying ever-widening support as evidence in its favor kept coming in. Most important, geochemist Roger Summons of MIT found evidence for the presence of the hydrogen sulfide–producing microbes at the P-T boundary in nine places around the world. The toxic bloom was essentially global in extent. Questions remained, however, including whether there would have been enough hydrogen sulfide to actually kill things. Further, there was the question of whether all of this could be connected to the most salient evidence from the Paleocene thermal event, that the conveyer current system of the time had shifted to produce a warm, anoxic ocean bottom, or the main evidence from the Triassic mass extinction, that there was a series of mass extinctions, not just one, as evinced by the isotope record. The possibilities were exciting—it looked as if the evidence from the Paleocene, Permian, and Triassic extinctions could be forged into a new paradigm for mass extinction.
The Kump group presented new findings that added to their initial 2005 model at a large, NASA-sponsored astrobiology meeting held in Washington, D.C., in March 2006. It was the year’s largest gathering of astrobiologists. Although much of the meeting dealt with more mainstream astrobiological topics, such as the new data from Mars and Titan showing that liquid of one kind or another had once been present on both bodies, or on the limits of extremophilic microbes on Earth, one afternoon was set aside for mass extinctions, for they were increasingly viewed as viable topics of astrobiology.
In a packed room, Kump and I presented back-to-back papers. In his, Kump reexamined the validity of his 2005 suggestion that it was hydrogen sulfide that actually killed things when the ocean states changed.
He showed a series of slides that in movie fashion showed that ancient Permian world. It was the oceans that were the critical element, and we all watched, fascinated, as the oceans became ever redder—the red chosen as the means to illustrate rising hydrogen sulfide levels. As they turned from pale pink to dark red, all the oceans were shown to be accomplices in the poisoning of the world. And perhaps most interesting of all, the overabundance of hydrogen sulfide did not happen only once but occurred over and over, as a succession of burps clustered around the time that the P-T boundary strata were being deposited around the world. Kump finished with the most ominous note. Not only did the model show where the hydrogen sulfide would emerge from the sea into the air, but he also showed new calculations that corroborated his earlier 2005 estimates of how much hydrogen sulfide would have eventually gone into the atmosphere. The results: There would have been more than enough to kill off most land life as the nasty stuff came out as bubbles. There would also be high levels of it dissolved into shallow seawater, where it would have been lethal in shallow marine settings as well, especially among shallow-water organisms that secreted calcium carbonate skeletons, such as corals, clams, brachiopods, and bryozoans, all invertebrate victims of the greatest extinction. Those organisms had already been teetering on the edge of extinction by the highly acidic seawater of the time, a product of the great volumes of carbon dioxide that entered seawater from the atmosphere. (Alarmingly this is occurring in our world in the Arctic Ocean, now so acidic that one group of mollusks, the pteropods, which are important in the food chain, are going extinct as their shells dissolve off their backs, as described by John Raven of the University of Dundee in Scotland and his colleagues in 2005.)
By the end of this session it was clear that the oceans were the key. But why would they change state?
A SHORT FERRY RIDE FROM SEATTLE LIES AN UPSCALE COMMUTER HAVEN called Bainbridge Island. Each morning thousands of suits take the 30-minute ride from suburbia to downtown offices and then rush back again at the end of the day. The tax base on the island, with its numerous waterfront and water-view houses is enormous, and why not—the view is sublime from the east side of the island looking at the magnificent cityscape of the downtown Seattle waterfront looming upward across two miles of Puget Sound. Even back in the Great Depression the wealthy valued this island, which became the site of the novel Snow Falling on Cedars, but not everyone who rides the boat to Bainbridge is a lawyer or plays golf. Some of the riders are students of various paleontology classes, for the southern tip of the island, on a closed country club, to be exact, is made up of 30-million-year-old sedimentary rocks that had been deposited on a fairly deep ocean bottom. Such outcrops are rare in the Seattle area, since the ice ages managed to dump untold tons of sand and gravel on the entire region repeatedly, covering the most useful teaching tools of a paleontologist, the rocks containing fossil life. The reason these outcrops are exposed is itself plenty ominous; the entire southern end of the island was thrown upward during the last mega earthquake that the region is prone to every 200 years or so. (The last was 200 years ago, and the tsunami wave generated by this monster quake crossed the ocean to devastate Japan, where its visitation was recorded in much art. Modern Seattle is a doomed city, each of its residents betting that a giant quake will not happen in our lifetimes. But what is life if not a gamble?)
The fossil-bearing rocks carry a salient message about the nature of the ocean back then: It was much like the ocean now—oxygenated from top to bottom—and we presume that this was maintained in some way by a conveyer current system analogous to that of today. By the Oligocene epoch of 30 million years ago, the world had cooled to something like its present state, after having been much warmer during the previous epoch, the Eocene. The ocean was warmer (as evidenced by a larger percentage of tropical snails and clams), but oxygen levels were the same at the bottom as at the top, and that has remained the case since. The abundance of life on the fossil bottom confirms that the oceans were animal-friendly from top to bottom, as does the fact that almost no bedding is visible in these rocks: On that ancient sea bottom a host of invertebrates managed to munch through the surface sediment to the extent that the original bedding was destroyed. We see this on the Bainbridge outcrops. No sedimentary bedding at all, just thick piles of well-sorted sediment rich in the shells of clams, snails, and other invertebrates. To a professional fossil finder, this kind of bottom is completely unlike the sea bottoms turned outcrops of older times.
The main drivers that created this mixed ocean were the extreme temperature differences that existed, and still exist, between the cool polar regions and the tropics. When there are warm surface areas and cold surface areas of the ocean, cold water spontaneously flows toward the warm, and vice versa. But more than surface currents accomplish this. Cold seawater is denser than warm water of the same chemistry and thus sinks. Saline water is denser than less saline water of the same temperature and also sinks. In the heat of the tropical sun, water rapidly evaporates, making the surface saltier and thus denser. In the Arctic, the melting of ice adds water to the sea, making it fresher. All of these factors create seawater bodies of different temperature and salinity that want to mix with others of different values, and in so doing produce conveyer currents throughout the world’s oceans.
But this kind of ocean is a relatively new one. We have to go back only a slight way further in time to find a very different kind of ocean, one where the bottoms had very little oxygen. A longer ferry ride takes one north to the outermost island fringes of America, tucked into a larger archipelago of Canadian territory. There, too, rocky outcrops bear fossils, but both the rocks and fossils are very, very different from their younger Bainbridge counterparts. Here the black sea bottom beds show fine lamination and almost no fossils. The only remains are of surface-dwelling creatures of the time, fish, and chambered cephalopods such as nautiloids and ammonites. The layering was caused by the same sedimentary processes that were found on the younger Bainbridge Island beds, but the difference comes from the fact that unlike the Bainbridge beds, which were deposited on a well-oxygenated sea bottom, here there was an ocean bottom devoid of oxygen. The clues are numerous. Not only are there well-laminated beds but also numerous blebs of pyrite, or fool’s gold, a sulfur-rich mineral that forms in the absence of oxygen. Sometime in the interval between the older fossil beds and the younger, the ocean radically changed.
The rock records and fossil records offer abundant testimony that this unmixed, or stratified, ocean, not the current mixed ocean, was far more common over most of geological time. The stratification involved temperature and salinity, and, for life, two far more important factors: dissolved oxygen and organic (reduced) carbon. They were characterized by an oxygenated surface layer, overlaying a much thicker water stratum with little or no oxygen. Encountering no oxygen at the sea bottom, the sediments accumulating on them became filled with black minerals colored by the abundance of sulfur within them; these sediments formed in a fashion similar to that responsible for the black layers found today on any beach when a clam digger gets below oxygenated sand and enters the thick black layer with its rotten-egg smell.
These black shales can be found all the way back to the dawn of life on Earth, at least 3.5 billion years ago. Paleontologists often love a good fossilized anoxic ocean bottom. Not for what lived there—there was little, and even less with shells available to fossilize—but because animals from the upper and still oxygenated layers fall to the bottom to be preserved, often in spectacular fashion. There are untold examples, the best being the life captured in the exquisite Burgess Shale, invertebrates and plants that fell onto a deep Cambrian anoxic bottom that preserved even their soft parts as well as the more commonly fossilized skeletons. Nice! And how about Archaeopteryx and much else from the Solnhofen limestone of Jurassic Germany, those early birds whose bodies fell onto an anoxic bottom, a place bereft of the scavengers that usually feast on such fowl. No scavengers ruffled those first feathers spread out on the bottom sediment in sprawling death. But for other kinds of life, like us animals, the low-oxygen conditions are highly inimical.
The stratified oceans can themselves be subdivided into two kinds. When oxygen levels at the bottom are just low, there may still be a few animals here and there, or maybe not. But the most common organisms by far on these bottoms are microbes. Even a tiny bit of oxygen, too little to support animals, is enough to maintain one kind of microbe, although it plays no part in the affairs of us animals. But when oxygen levels really reach bottom, a very different kind of microbial kingdom takes over, one dominated by bacteria that use sulfur as foodstuff. These are the nasty forms that make the poisonous hydrogen sulfide. At times when they have been present in abundance—and this can only be ascertained by finding their characteristic biomarkers in the organic fraction of the rocks making up these ancient sea bottoms—we say that the ocean containing them was a Canfield ocean.
So toxic were Canfield oceans that they might have reduced animal life, or even inhibited its first evolution for millions of years during the long-age Precambrian era, which includes the time from life’s origin to less than 600 million years ago. There seem to be two reasons for this. First is the obvious toxicity of the hydrogen sulfide, but just as important may have been the microbes’ inhibition of nitrogen formation in compounds useful for plant life. While many kinds of microbes can “fix” biologically useless nitrogen, an essential element for life, into compounds that are biologically useful, the eukaryotes—plants, animals, fungi, and a variety of other groups—cannot do this trick and so depend on microbes to do the job for them. Enter the Canfield ocean’s gang of sulfur bacteria, and little nitrogen becomes available, because this kind of bacteria couldn’t care less about nitrogen, and also inhibits other microbes from supplying it. A nitrogen-poor ocean would have been an ocean literally in need of fertilizer and not getting it. It would have been just like a soil from which all the nitrogen has been leached—only a small amount of plant life will grow. Nasty place, that Canfield ocean. Perhaps if a mixed ocean turned into a Canfield ocean, a great mass extinction would soon follow. Is there any evidence that these Canfield oceans, which we know existed in the time before animals, made their destructive returns, like a bad plague, in the time of animals as well?
Yes. The most important of the mass extinctions was clearly at the P-T extinction, and it was indeed a time of a Canfield ocean, an identification made in 2005 when a team led by biogeochemist Kliti Grice of the Curtin University of Technology in Perth, Australia, published a seminal paper in Science on research that demonstrated that the oceans at the end of the Permian period showed biomarkers of the microbes that would be expected in a Canfield ocean. The second, the T-J extinction, is just now being examined, but already beds from the Alps have shown the presence of isorenieratane, the biomarker characteristic of the purple and green sulfur photosynthesizing bacteria, forms that can live only in seas shallow enough for light to penetrate that are also low in oxygen and high in hydrogen sulfide concentrations. We do know that the last few million years of the Triassic period and the first few million years of the Jurassic period were characterized by a series of isotopic perturbations that coincided with pulses of anoxia in the sea, and both of these strongly suggest that a series of short-lived Canfield oceans led to the sequential series of mass extinctions that, combined, we call the T-J mass extinction.
Three different ocean states—the mixed ocean, and two kinds of unmixed ocean, the anoxic and Canfield oceans. How and when does one become one of the others? Here is where the conveyer current systems come in. They seem to be the gatekeepers for determining which ocean type will be present. And it may not be the presence of any of these oceans that causes distress to life but the change from one state to another.
THE SOURCE OF THE MASS EXTINCTIONS WAS A CHANGE IN THE LOCATION at which bottom waters are formed. Near the end of the Paleocene epoch, the source of our Earth’s deepwater shifted from the high latitudes to lower latitudes, and the kind of water making it to the ocean bottoms was different as well: It changed from cold, oxygenated water to warm water containing less oxygen. The result of this was the extinction of deepwater organisms that Jim Kennett and Lowell Stott were investigating. The cause of the Paleocene event is thus linked to a changeover of the conveyer belt system. What about the biggest of all extinctions, the Permian? It turns out that for it, too, a changed conveyer current holds the smoking gun.
In 2005, climatologists Jeffrey T. Kiehl and Christine A. Shields of the Climate Change Research Section at the National Center for Atmospheric Research used a global circulation climate model to look at the Permian world. Kiehl and Shields wanted to know if Permian ocean circulation patterns were disrupted at the time of the extinction. When they plugged in the known positions of the continents and inputted a warmed world as well, their modeled Permian world showed a shift in the positions of its conveyer belt currents. They proposed that sudden global warming caused a change in ocean state.
Oceanic currents play a huge role in current climate and global temperature. Today, whether the conveyer current system in the north Atlantic Ocean runs seems to be controlled by the amount of ice cover on Earth, and in a complicated fashion (no weather is ever simple, alas) by the nature of tropical warming or cooling. By the end of the Permian period, Earth appears to have had no ice—the ice caps had all melted away from their early Permian maximum. (The early Permian period of 300 million to about 270 million years ago was so globally cold that there were vast continental glaciers resembling those of our own recent Ice Age. By the late Permian period, some 260 million to 250 million years ago, however, they were either gone or going fast, according to the geological evidence from these time intervals.) The conveyer current does not shut down in the absence of ice. Rather, it shifts the positions of its starting and ending points (where water either comes up from depth or dives down to ocean bottoms). That shift may have been crucial in the mass death that followed.
Because the continents were in such different positions at that time, models we use today to understand ocean current systems are still crude for the Permian oceans, and they have much less precision than those we can make for the modern world. Nevertheless, it seems fairly clear that by the end of the Permian period, ocean circulation had changed so that the deep ocean bottoms filled with great volumes of warm, virtually oxygen-free seawater. This seems like the same thing that happened at the end of the Paleocene epoch but at a vastly increased scale, and with vastly more destructive results. The Permian bottom waters were warmer than those of the Paleocene and much less oxygenated. The stage was set and needed but one more trigger, and it seems both had the same trigger—a short-term but massive infusion of greenhouse gases into the atmosphere changed the nature of the oceans. In the Paleocene epoch the source of that carbon dioxide was volcanism in the Atlantic Ocean region, whereas at the end of the Permian period the initial source of the heat was emission of vast volumes of carbon dioxide from the spectacular lava outpourings, perhaps one million cubic miles in volume, that today cover some 800,000 square miles and that might have covered nearly three million square miles when formed into what is known as the Siberian Traps. (Why igneous geologists call stacked-up piles of lava “traps” is beyond me.)
Now, it seems, events at the end of the Permian period can be related to changes in oceanography as well, with the addition of a kill mechanism from hydrogen sulfide that was microbially produced.
The main difference between the two events seems to be that the Permian event showed far more upwelling of poisonous bottom waters. In the case of the Paleocene event, some deep, near-shore basin underwent a change from oxygenated to less oxygenated, even into the shallows, after upwelling of the deep, warm water, in a manner happening today to the Gulf of Mexico, Gulf of California, and Puget Sound, among other such places. Deep, warm water also upwelled at the end of the Permian period, but it did so over a far greater area of the globe—virtually every shallow-water area (rather than just a few, as at the end of the Paleocene epoch) became filled with warm water without oxygen, even at the surface. And the Permian deepwater brought up poison not seen at the end of the Paleocene—it was rich in carbon dioxide and methane, which seems to have moved out of the solution in seawater and into the atmosphere as potent greenhouse gases (causing even faster planetary warming) as well as the deadly hydrogen sulfide gas, which, if it occurred a the end of the Paleocene, did so only at low concentrations.
IN A 1997 BOOK, MASS EXTINCTIONS AND THEIR AFTERMATH, ANTHONY Hallam of the University of Birmingham in England and Paul Wignall of the University of Leeds compiled what was known about all the mass extinctions in an excellent volume. At the time the impact hypothesis still held sway. Nevertheless, their data show that of the 14 mass extinctions they recognized, 12 of them were characterized by poorly oxygenated oceans, which they thought must have been a major part of the cause of the extinctions. There is no proposal about how the oceans got that way, however. With the models above, we now have a mechanism: perturbation or even stoppage of the thermohaline, conveyer current systems. It is time to stop looking at the “kill mechanisms”—low oxygen, heat, and perhaps excess hydrogen sulfide gas in water and air—and start looking at the driver of these changes, the atmosphere itself.