In late August, more than 600 separate wildfires ravaged California, killing seven people. Meanwhile, two tropical cyclones struck the Gulf Coast only days apart: first Tropical Storm Marco and then Hurricane Laura, the latter of which killed 26 people in the United States and tied the record for the strongest storm to hit Louisiana. Extreme events such as these signal a worrying trend. In the coming decades, as temperatures continue to climb, seemingly isolated climate disasters will begin to overlap, their impacts becoming more than additive. Scientists expect to see more intense tropical cyclones and more heat waves. Each disaster could compound the damage of the next, with less and less time for people to recover in between.

Many observers assess the threat of climate change in terms of the frequency or severity of extreme events. They have viewed each crisis—be it a Texas hurricane or a California wildfire—as distinct from others. But consider how people feel on the fourth day of a heat wave as opposed to the first. Their resilience begins to drain away. Viewing weather events as independent occurrences is like trying to understand a movie by looking at a series of brief clips; they are important plot points, but not the whole story. In fact, viewing climate change as the accumulation of individual events underestimates the threat, because such events do not take place in a vacuum. As recent research shows, features of the climate interact with one another—interactions that exacerbate the impact on people and ecosystems. 

Two interactions are particularly worrisome. First, as extreme events become more intense and more frequent, they will increasingly occur close together in time and location, worsening the overall impact. Alone, a single extreme event—such as a hurricane or a wildfire—can devastate wide areas. But back-to-back climate catastrophes compound the misery of each. The second type of interaction is longer term. It happens when one of the earth’s mechanisms for regulating the climate—systems involving air, the ocean, land, or ice—runs amok, setting off a chain reaction involving other such mechanisms. 

These new risks to the planet should challenge the conventional wisdom on fighting climate change. In the United States and other wealthy countries, efforts to adapt to global warming have always played second fiddle to efforts to reduce carbon emissions. This emphasis is understandable, since if greenhouse gas emissions are not restrained, successfully adapting to climate change will be impossible for most of humanity: countries will suffer major damage, and lives will be lost. Adaptation has also seemed less attractive because it involves no global silver bullets. But policymakers no longer have the luxury of downgrading adaptation, because climate change’s devastating effects are no longer in the future; they are occurring now.


Extreme events can wreak havoc on society. In 1953, a powerful storm in the North Sea killed more than 2,000 people in the Netherlands and the United Kingdom. In 2003, a searing heat wave gripped western Europe, contributing to somewhere between 35,000 and 70,000 deaths. In 2012, Superstorm Sandy struck the northeastern coast of the United States, causing nearly $80 billion in damage. These episodes unleashed cataclysmic consequences, but few communities had to face them more than once over several generations. Such events occur worldwide multiple times per year but rarely in the same place. The North Sea storm, for instance, appears to have been a once-in-a-century event for its region; the 2003 heat wave, a once-in-500-year event; and Sandy’s flooding of New York City, a once-in-250-year event. 

But soon, some once-in-a-lifetime catastrophes will become annual debacles. As temperatures rise, the odds that such events will occur at any specific location in a given year are growing quickly, particularly in coastal areas. By 2050, many such areas around the world will face flood levels every year that only recently occurred once per century. 

When extreme events strike the same location more frequently, the confluence can be more devastating than the sum of its parts. Consider a string of extremely hot days in one particular place—the odds of which, computer climate models confirm, are growing rapidly. A few consecutive days of unusually hot weather is manageable, but a week or two, far less so. As a heat wave goes on, the electrical grid struggles to supply enough power for all the air conditioning being used. Blackouts are triggered. With no air conditioning, the human body’s own system for mitigating heat breaks down, too. Some die of heat stroke and respiratory disease. For those who lack air conditioning (which is a majority of the world’s population), many of whom live in aging, urban apartments that are slow to cool naturally, the risk is greatest. As hot days bunch together, such households will see long stretches without relief, since the indoor temperature lags a day or two behind the outdoor temperature. 

Dogs stranded after Hurricane Florence in Leland, North Carolina, September 2018
Dogs stranded after Hurricane Florence in Leland, North Carolina, September 2018
Jonathan Drake / Reuters

Another scientific finding suggests that the problem does not stop there: in many locations in the future, episodes of high humidity will be more likely to accompany hot days than they are now. Heat plus humidity equals more human misery than heat alone or humidity alone—and more than the sum of the two. Above a certain threshold, the human body can no longer dissipate its own metabolic heat through perspiration. More and more often, in an expanding area of the world, outdoor activity involving any significant effort—farm labor, construction work, or even a soccer game—will be life threatening.

The devastation caused by multiple extreme events is not hypothetical, as the 2017 hurricane season showed. In August of that year, Hurricane Harvey struck the Gulf Coast of the United States, deluging parts of the areas around Houston, Texas, with more than four feet of rainfall and causing over $90 billion in damage. A couple of weeks later, Hurricane Irma flattened parts of the Leeward Islands, in the Caribbean, while striking a glancing blow to Puerto Rico. Just two weeks after that, Hurricane Maria made a direct hit on Puerto Rico, destroying its infrastructure and causing about 3,000 deaths. At some point in their paths of destruction, each storm was classified as Category 4 or 5, the highest levels of intensity.

Like heat waves, consecutive hurricanes of this magnitude can exacerbate misery. The link in this case was not necessarily geographic or temporal. Hurricane Maria made landfall in Puerto Rico 26 days after and over 2,000 miles away from Hurricane Harvey’s strike on Texas. But these two events connected at a distance. The damage in Puerto Rico was made worse because the U.S. Federal Emergency Management Agency had exhausted its financial and personnel resources on the previous two storms, especially Harvey. Add to that the island’s financial troubles, its lack of representation in Congress, and the Trump administration’s hostility—perhaps not surprising, given the island’s overwhelmingly Hispanic population—and the result was gross mismanagement that worsened the disaster. 

Even extreme events scattered across the world can compound one another. Consider crop failures. About 15 percent of the world’s grain is consumed not in the country where it was grown but after being exported. The biggest exporters of grain—Argentina, Australia, Russia, Ukraine, and the United States—are spread out around the world. That is a good thing from the perspective of food security, because it minimizes the chances of simultaneous crop failures. But global warming is increasing those odds. Yields of corn, soybeans, and other key crops fall sharply as temperatures rise and the amount of water they receive falls. As a result, there is now a growing possibility of simultaneous crop failures in two far-apart breadbaskets—something that could disrupt the global food supply and lead to malnutrition and, in some places, widespread starvation.


Beyond the prospect of extreme events coinciding or connecting, another sort of interaction is just as worrying: those among the various systems that drive the climate. Scientists have long worried about tipping points—thresholds beyond which small changes in the global temperature can lead to rapid, disruptive effects. For example, if large portions of the Greenland and Antarctic ice sheets melt and disintegrate, a process already underway along their fringes, the global sea level will rise much more rapidly than it has for thousands of years. As Arctic permafrost across North America and Eurasia melts, it will release large amounts of methane and carbon dioxide, further increasing the rate of global warming. If a key ocean current in the North Atlantic slows down as a result of global warming, climate at the high latitudes will be disrupted. Although scientists have looked at these possibilities with a wary eye for decades, they have been unable to nail down the exact temperature at which these rapid responses would occur—or to determine if precise tipping points even exist. 

But if such thresholds exist and were crossed at relatively low temperatures, the result would be disastrous: the widespread dislocation of ecosystems and societies with little window of opportunity to adapt. Even worse, there’s evidence to suggest that several of these phenomena would interact. If a threshold in one system were crossed, there might be a ripple effect, causing thresholds in others to be crossed, too. For example, a rapid loss of Greenland ice would pour water into the surrounding sea, slowing ocean circulation. Because this current normally channels warm water northward, its slowing would create something akin to a series of rear-end collisions in a traffic jam, causing a warming of Antarctic waters on the other side of the world. That, in turn, would have another knock-on effect, speeding the breakup of parts of the Antarctic ice sheet. The global sea level would surge even higher.

When these building blocks of the climate are examined individually, the chances of crossing multiple thresholds more or less simultaneously appears small. Some of these phenomena are unlikely to occur in this century or even the next without a major trigger. But that is precisely what the interaction of these various systems might create: one system may go haywire, triggering the disruption of others. At this juncture, there is still a great deal of uncertainty about how high that probability is. But the potential for such interactions adds another reason to be extremely cautious about venturing beyond the targets set in the 2015 Paris climate agreement—keeping warming to well below two degrees Celsius above preindustrial levels and trying to limit the increase to 1.5 degrees Celsius. Exceeding those targets would mean entering a climatic terra incognita.


The interaction of extreme events creates risks of an entirely new type and magnitude. Using computers to predict when and where such events may occur is of little immediate help, since modeling of those events is in its early stages. Nor can one extrapolate from past experience, since the climate is evolving well outside of what humans have lived through. It’s not that a confluence of risky climate events at a particular place and time is entirely new. But what is new is that the likelihood of some confluences is increasing rapidly and globally.

Further complicating predictions is the question of how people and governments will respond. People who are not directly involved in an extreme event tend not to remember the lessons learned from such past events long enough to prepare for the next. Some studies suggest that it takes multiple similar incidents to leave a deep enough impression to convince them to learn from their experience and adapt accordingly. Only then will they think ahead and act to protect lives and property or get out of harm’s way by relocating to safer terrain.

Climate change’s devastating effects are no longer in the future; they are occurring now.

Even highly developed countries are underprepared for climate risks, especially in certain geographic areas, economic sectors, or demographic segments. Before Hurricane Katrina struck in 2005, New Orleans had an emergency escape plan, but it didn’t consider poorer people who didn’t have cars, most of whom were Black. As a result, many stayed in their houses and drowned or wound up in New Orleans’s Superdome, which had been set up as a shelter. Other countries may be able to handle one threat but completely overlook another. Japan, for example, has millennia of experience dealing with earthquakes, floods, and typhoons, and its disaster-risk-management system is the envy of the world. Yet the country failed to prepare for a new type of disaster that arose in 2011: an earthquake triggering a tsunami, which flooded a nuclear reactor.

As these examples suggest, although governments can learn through experience with individual disasters, they are almost never ready for new combinations of them. That does not offer much reason for optimism when it comes to preparations for climate change. Indeed, in a world where climate risks rarely interact, governments are already inadequately planning for potential disasters. As those risks increasingly compound one another, governments will lag even further behind the threat.


Nearly all accounts of the climate problem from scientists and other experts end with a plea for rapid reductions in greenhouse gas emissions. But governments should emphasize adaptation equally. That means developing forward-looking policies to protect people, infrastructure, ecosystems, and society. It means restructuring or replacing perverse incentives that encourage people and industries to settle in exposed areas. It means giving more resources to international agencies to help the least developed countries. Most of all, it means thinking many years ahead to gather extensive resources and political will for often unpopular policies. Very little of this job can be done quickly. Adaptation should have begun in earnest decades ago.

Emphasizing emission reductions but not adaptation to climate change is misguided, because no matter what happens to emissions over the next 30 years, the planet will get significantly hotter. Trapped heat that has been absorbed by the oceans over decades is bound to emerge, warming the earth. Years of emissions have accumulated in the atmosphere and will have a lagged effect on the climate. Although the world may be capable of meeting the targets set in the Paris agreement, it is more likely than not that it will fail to do so. Ever since international climate change negotiations began in 1991, countries have talked the talk more than they have walked the walk. If the targets aren’t met, climate change will produce more events that a greater number of governments will either have to learn to adapt to at a very high price or altogether fail to manage.

Even achieving the Paris targets would not be a free pass to avoid adaptation.

Even achieving the Paris targets would not be a free pass to avoid adaptation. Attaining those goals would give the world some welcome breathing room. But the resulting warming would still create serious consequences, such as a hundredfold uptick in the frequency of floods along large swaths of the world’s coasts. It is true that no amount of adaptation will be enough if emissions remain unconstrained, because that would lead to warming that would go far beyond what humans have ever experienced. But it is also true that no amount of emission reduction will be enough to spare communities that do not also adapt.

Governments must also remember that the ability of people and places to adapt to climate change is highly unequal, largely because of unfair arrangements determined too often by racial, gender, ethnic, age, or other differences. Many of the interactions between extreme events will become apparent only suddenly, so accommodating them will require extra flexibility to respond rapidly—a capacity that much of the population in less developed countries and major segments of wealthy countries have long been deprived of.

The bottom line is that few if any countries are sufficiently prepared to deal with what is in store. A yawning gap has opened up between what they know about the risks of climate change and what they are doing to reduce them. In the riskier new era of climate change, the longer countries take to close that gap, the more painful and deadly the outcomes.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • MICHAEL OPPENHEIMER is Albert G. Milbank Professor of Geosciences and International Affairs at Princeton University.
  • More By Michael Oppenheimer