Skip to main content

Suffocating the Ocean

Oxygen-depleted oceans have preceded many mass extinctions in Earth’s past, including the worst one of all 252 million years ago. Are hypoxic dead zones from California to Namibia a harbinger of the next extinction?

By Moises Velasquez-Manoff


(Illustration: Tim Tomkinson)

It was crabbers who first reported something amiss. In 2002, they began pulling in traps full of corpses. (Crabs should be alive when you catch them.) And they mentioned something else: Little octopuses had followed their crab lines to the surface, as if fleeing inhospitable conditions below.

Then heaps of dead crustaceans began washing ashore along a stretch of Oregon’s coast. When scientists sent a robotic submersible offshore, they discovered mile upon mile of dead crustaceans, the water brown and murky with detritus.

The killer was low oxygen, or hypoxia. Nearly all animals require oxygen to live, and, that year, dissolved oxygen had fallen so low off Oregon’s coast that whatever mobile creatures could had fled, while more-sessile life had simply suffocated.

“You look out, all you see is blue ocean. It looks exactly the same as 50 years ago. But it’s not.”

Every year since, those hypoxic waters have appeared in late summer and early autumn. In 2006, they became anoxic, meaning they lost all their oxygen. “You didn’t see a single fish in a day,” says Jack Barth, a professor at Oregon State University, “just the piles of crab carcasses and worms that had come out of the bottom, sort of wafting in the current.” When scientists examined the roughly 50-year record of oxygen measurements from the region, they couldn’t find a single comparable event in the past. The hypoxia, it seemed, was unprecedented.

And then it spread.

In 2013, low-oxygen water showed up off the California coast just north of San Francisco. The following year, crabbers pulled in dead crabs in Half Moon Bay, just below San Francisco. Further south, the Monterey Bay Aquarium registered a decline in the oxygen content of the water it pumps in from the ocean.

“It seems like something has changed,” says John Largier, head of the Coastal Oceanography Group at the University of California-Davis. He tries to remain skeptical, but he suspects “large-scale global change.”

These suffocated patches of ocean aren’t just bad for fishermen and their catch; they represent a change in the ocean that has, at times in Earth’s past, heralded mass extinction.


Hypoxia occurs naturally in fertile waters. Plankton blooms and, as it dies, it rots, leeching oxygen from the surrounding water. Human effluent is often responsible for these low-oxygen expanses, or dead zones. When fertilizer or sewage washes down rivers, it prompts algal blooms and then oxygen depletion.

But the low-oxygen zones appearing along the western coast of North America aren’t due to runoff, at least not entirely. They probably reflect a number of related forces acting in concert. Along the western edge of North America — and the western edges of other continents — winds that cause upwelling have strengthened in some areas in recent decades, pushing water out to sea with greater force. That has intensified the upwelling of nutrient-rich water along the coast, leading to more phytoplankton, and more hypoxia as the blooms die.

The ocean is also changing independently of the winds. Warm waters hold less oxygen than cold, so as the Pacific has warmed, its ability to absorb oxygen from the atmosphere has declined. And how the ocean has warmed is also likely playing a role. The ocean is naturally stratified, like a layer cake. Deep waters are, because they’re separated from the atmosphere, always somewhat oxygen-depleted.

The oceans became hot, stagnant, and inhospitable to animals, reverting to a primeval state where microbes reigned supreme.

As the upper layers have absorbed heat from the atmosphere — about four small nuclear devices’ worth per second, according to one estimate — they’ve become more buoyant. That relatively warm, buoyant surface water acts like a lid on the ocean, impeding aeration of deeper waters. It’s this already oxygen-poor water that may be upwelling along western North America — meaning that the water in Oregon and California’s hypoxic zones is being fed by a much larger, growing mass of low-oxygen water in the Pacific.

“When you look out at the ocean, all you see is blue,” says Lisa Levin, a biological oceanographer at the Scripps Institution of Oceanography in La Jolla, California. “It looks exactly the same as 50 years ago, even though it’s not.”

When we think about the ocean and climate change, we often imagine ocean acidification, which will make life difficult — and, as it progresses, impossible — for some shell-forming organisms. We learn of melting ice caps and sea-level rise, which could require that humans retreat from the coastline and abandon low-lying regions, like Florida. We don’t hear much about hypoxia. Yet in some parts of the ocean, like the northeastern Pacific, hypoxia may be a greater near-term threat than acidification. And its emergence in multiple other places has caught some in the scientific community by surprise.

“It’s been a large potential problem that’s generally been under-appreciated,” says Curtis Deutsch of the University of Washington.


This story first appeared in the May/June 2016 issue of Pacific Standard.

Almost anywhere anyone has looked in recent years — including the Atlantic Ocean — those deep, naturally low-oxygen waters have pushed upwards, in some places by an average of three feet per year. It’s unclear how marine life will respond. Off Southern California, where a northerly current has grown warmer and less oxygenated in recent decades, some sea urchins are moving into shallower water. The ecosystems also seem to be simplifying. Like underdeveloped economies, less diverse ecosystems may be less productive — yielding fewer of the goods, such as fish, that we like. They’re also more prone to collapse.

New creatures have also arrived. The man-sized Humboldt squid typically inhabits waters off the coast of Mexico and points further south. In the early 2000s, they began pushing north, frightening divers along the California coast, and eventually fouling fishing nets as far north as Alaska. The squid, it turns out, are unusually tolerant of low-oxygen conditions. They seem to have surfed a growing expanse of oxygen-depleted water northward.

More broadly, as fish squeeze into a narrower band of ocean, catches may improve, at least for a while. But there’s a potential downside to catching too many fish — overfishing. Some speculate that the decline in recent decades of large fish everywhere may have been driven, in part, by rising low-oxygen waters that have essentially pushed fish toward fishermen.

For those who study the climate of Earth’s past, the mounting signs of ocean hypoxia inspire both unease and a sense of vindication. In the past two decades, paleontologists have developed a new theory about mass extinctions — including the worst of all, the End Permian die-off 252 million years ago. Ocean hypoxia features prominently. Then, as now, the trigger was a rapid accumulation of heat-trapping gases in the atmosphere. The oceans became hot, stagnant, and inhospitable to animals, reverting to a primeval state where microbes reigned supreme.


In 1980, scientists Luis and Walter Alvarez, a father-and-son team, proposed a new idea for the disappearance of the dinosaurs 65 million years ago: an asteroid strike. As evidence, they pointed to a 65-million-year-old layer of clay, found around the world, rich in an element called iridium. Iridium is rare on Earth, but common in asteroids. That layer, they argued, contained the pulverized remains of a six-mile-wide asteroid. A 110-mile-wide impact crater later discovered off the coast of the Yucatan in the Gulf of Mexico lent credence to their idea.

Earth has experienced five such extinction spasms, interspersed with numerous smaller ones, and many argue we’re now entering a sixth extinction — one caused by human activity. Before the Alvarez hypothesis, scientists often blamed volcanic activity for these die-offs. After Alvarez, many suspected that collisions with celestial bodies might have been responsible.


(Illustration: Tim Tomkinson)

Proponents found some evidence to support the idea — a little iridium, and what looked like crystals formed during impacts. But the closer scientists peered at these other extinctions, the more things didn’t jibe with the celestial impact model. The geological record often suggested a long, drawn-out process more than a single, catastrophic event. And the evidence of fluctuations in atmospheric carbon — meaning greenhouse gases — indicated that heat-trapping gases often built up before mass extinctions.

The picture that began to emerge was less of a sudden cataclysm and more of a series of insults that pushed ecosystems over a threshold beyond which they began to rapidly disintegrate. This gradualist view fits particularly well with the End Permian die-off, sometimes called the mother of mass extinctions.

In that era, Earth’s continents had converged into a single land mass called Pangea. Massive volcanic eruptions occurred in what’s now Siberia, spewing magma over an area nearly the size of Australia. The eruptions, known as the Siberian Traps, might have been survivable. But as the molten rock pushed out of the Earth, scientists think it ignited beds of coal laid down during a previous epoch, releasing huge quantities of carbon dioxide over a relatively short period of time — maybe tens of thousands of years. The warming atmosphere heated the oceans from the top down, turning vast portions of them stagnant and hypoxic.

What probably took a few thousand years of volcanoes during the end-Permian will have taken modern civilization, with its tailpipes and smokestacks, a few hundred years.

Scientists can see evidence of this prehistoric hypoxia in marine rocks laid down at the time, which contain chemical signatures from microbes that thrive in low-oxygen conditions. And there’s something else: a peculiar fingerprint indicating the presence of noxious sulfur compounds.

Deep in the ocean live anaerobic microbes that emit the poisonous gas hydrogen sulfide as a byproduct. Lee Kump, head of the Department of Geosciences at Pennsylvania State University, thinks that, as the low-oxygen waters expanded in the End Permian ocean, these microbes surfaced. The gas they spewed may have poisoned the atmosphere and torn holes in the ozone layer. A die-off that began in the ocean spread to land. Over 90 percent of marine species, including trilobites, went extinct, as did 70 percent of terrestrial species.

Today, there’s one place where scientists can glimpse End Permian–like conditions in the ocean. Off the coast of Namibia in southern Africa, a powerful upwelling occurs that brings those hydrogen sulfide-producing bacteria to the surface. From space, the plume is bright green — one likely hue of those sulfidic, End Permian oceans. The coast occasionally stinks of rotten eggs — hydrogen sulfide. Hordes of lobsters sometimes emerge from the sea, fleeing the toxic waters.

To the degree that levelheaded, skeptical scientists express worry at all, what worries them about the hypoxia today is that it resembles the beginning of what they think happened in the past.


In 1972, the British chemist James Lovelock proposed that Earth was a complex, living system — a superorganism — that could self-regulate and maintain conditions conducive to its own survival. He called this idea the Gaia hypothesis, after the Mother Earth goddess in Greek mythology.

As scientists’ understanding of extinctions has progressed, however, it increasingly seems that Gaia occasionally becomes homicidal. University of Washington paleontologist Peter Ward describes this idea as the Medea hypothesis in his 2009 book of the same name. Medea, another figure from Greek mythology, killed her own children.

“What almost certainly happens when you warm the planet is that you destabilize other carbon reserves. It’s a snowball response, and those responses can greatly amplify the original trigger.”

Far from self-regulating, Ward argues, if the Earth’s climate is pushed far enough, it can enter a spiral of self-destruction. That’s what he and others think happened during the End Permian.

“All of the greenhouse extinctions happen in the same way,” he says. They start as low- oxygen conditions deep in the ocean, and move upwards, squeezing multicellular life from the ocean. Eventually, “you breed a whole new group of microbes at the surface,” he says. Those microbes wreak more havoc.

If Ward’s story has a moral, it’s that mass extinction can begin in the corner of our planet that’s most out of sight — the deep ocean.


We’re not immediately near End Permian concentrations of atmospheric carbon dioxide, which may have been as high as 8,000 parts per million. By comparison, in 2015, we passed 400 parts per million. There are other important differences between now and 252 million years ago. Today, we have seven separate continents. During the End Permian, there was only that single supercontinent. With its limited coastal refugia — places to flee from the increasingly toxic ocean — End Permian geography may have exacerbated the die-off.

On the other hand, Kump notes, we’re releasing carbon into the atmosphere 10 times faster than the rate of release during the End Permian.

The speed of global warming is important — maybe even as important as the absolute amount of energy added to the climate system. The difference in temperature between the upper ocean and layers lower down — the severity of stratification — is what drives, at least in part, ocean hypoxia. During the End Permian, carbon dioxide may have doubled more than once, warming the planet by about six degrees Celsius. If current trends continue, in 50 years, the amount of carbon dioxide in the atmosphere will have doubled compared to the pre-industrial level of 275 parts per million. By century’s end, in the most pessimistic scenarios, it may double again.


(Illustration: Tim Tomkinson)

What probably took a few thousand years of volcanic activity during the End Permian will have taken modern civilization, with its tailpipes and smokestacks, a few hundred years.

And then come the positive-feedback loops. “What almost certainly happens when you warm the planet is that you destabilize other carbon reserves,” Kump says. “It’s a snowball response, and those responses can greatly amplify the original trigger.”

Arctic permafrost will, as it thaws, release more greenhouse gases. Undersea methane ice along the continental margins will also melt. Methane is much shorter-lived than carbon dioxide, but it’s around 30 times more potent as a greenhouse gas. And scientists have already observed methane gas plumes rising from the seafloor near Siberia and Norway’s Svalbard islands. It’s not always clear if they’re growing in size, or if they’ve been bubbling for some time. But what is apparent is that, as the sea warms, more methane ice will become gas.

We’re inadvertently replicating End Permian conditions in other ways as well. Back then, massive quantities of nutrients flowed into the ocean, likely causing plankton blooms that exacerbated ocean hypoxia. The reasons for this runoff are unclear. Maybe plants died en masse, prompting runaway erosion that flowed into rivers and fertilized the ocean. Or perhaps warmer temperatures accelerated the natural weathering of rock, freeing nutrients.

Earth has experienced five extinction spasms, interspersed with numerous smaller ones, and many argue we’re now entering a sixth extinction caused by human activity.

Today we’re also fertilizing the sea, mostly through the byproducts of agriculture and the burning of fossil fuels. The levels of nitrogen, which is critical to plant growth, entering into the ocean have likely doubled compared to pre-industrial times. The world’s largest dead zones, including at the mouth of the Mississippi River and in the Baltic Sea, are man-made.

We don’t have to go back 252 million years to find widespread ocean hypoxia. When Earth warmed at the end of the last Ice Age, which ended about 10,000 years ago, the entire eastern Pacific basin lost oxygen. Sarah Myhre, a postdoctoral research scholar at the University of Washington, studies sediment cores spanning the period. The warming ocean triggered an ecosystem collapse throughout the eastern Pacific, from Chile to Alaska.

Nothing went extinct ultimately — plants and animals re-colonized from other areas of the ocean when favorable conditions returned. But parts of the eastern Pacific went devoid of all but single-celled life forms for perhaps thousands of years.

“Environments we care about, and rely on, are on the table in a future with abrupt warming,” Myhre says. “We need to come to terms with [whether] those are environments we’re willing to sacrifice.”

Last year was the hottest on record. The year before, scientists recorded the highest mean sea-surface temperatures ever, largely due to a warming of the northeastern Pacific, the region that’s seeing more coastal hypoxia. Some scientists I spoke to, however, were hesitant to over-interpret the hypoxia. That reluctance was partly due to real uncertainty: Decades-long cycles operate in the Pacific, and they might also explain the low-oxygen conditions. Maybe more important, they didn’t want to encourage a fatalistic attitude. “It’s not good to write an article that the sky is falling and we’re all going to die,” says Lisa Levin of the Scripps Institution of Oceanography. “Because it’s not exactly true.”

A growing number of economists take a different approach. They argue that, rather than downplaying potential catastrophe, we should contemplate it squarely and incorporate the possibility, and the nail-biting uncertainty it inspires, into our plans.

Historically, economists have thought about climate change in a cost-benefit framework, where the costs of transitioning to carbon neutrality now are weighed against the costs of a hotter world later, and the costs of adapting to it. The outcome of this decision-making will depend on what you imagine that hotter world to be. If climate change is a matter of incremental shifts — a little warmer and wetter here, a little drier there — the most rational approach may be to do nothing; to instead invest your money in business-as-usual schemes and use the wealth generated to adapt in the future.

This approach only makes sense, however, if the future really is a slightly warmer version of the present. What many paleontologists think, though — and what research on the End Permian suggests — is that tipping points exist beyond which Earth’s climate shifts dramatically, leading, occasionally, to a worldwide ecosystem collapse.

Harvard University economist Martin Weitzman calls such events “fat-tailed.” They’re unlikely but not outside the realm of possibility. And if they occur, they’re cataclysmic. So it behooves us, he and others argue, to work strenuously now to prevent these game-changing events later.

We apply this sort of thinking in daily life all the time. It explains why we fasten our seat belts when we get in the car; why we purchase home and traveler’s insurance; why we wear helmets while biking. We’re constantly modifying our behavior and paying out hard-earned money to manage unlikely but potentially disastrous events.

In this view, mitigation is less about preventing more heat waves in August, or not having to protect New York City with dikes, than it is a form of “planetary insurance,” as one economist called it, against End Permian-like events — catastrophes that begin, scientists think, in ways similar to what’s happening in the northeastern Pacific today.