Syukuro Manabe. (Courtesy of Nobel Prize Outreach. Photo by Clément Morin)

Syukuro Manabe forecast global warming from a Princeton lab. He was right.

By Carolyn Jones | Images from Shutterstock unless otherwise attributed

When young Syukuro “Suki” Manabe gazed at the sky above Japan, with World War II raging, there were few indications he would change the scientific world. A contemplative child from the mountains of Shikoku, Manabe was more interested in clouds than in schoolwork. 

Yet it was this daydreaming that led to the first complete atmospheric model of the Earth — one that would accurately predict climate change. Manabe, now 94, forecast that rising levels of carbon dioxide would one day warm the globe. 

For his insights into our shifting climate, which he expressed mathematically years before they came true, Manabe is known as the father of modern climate science. These insights unfolded primarily in Princeton. 

Manabe’s path from rural Japan to central New Jersey seemed charmed. When his blend of curiosity and doggedness met a town churning with intellectual energy, his potential was set aloft. Conditions were favorable for success. The Institute for Advanced Study (IAS) had already pioneered the science of weather forecasting, while a unique partnership between the federal government, which employed Manabe, and Princeton University, fueled that nascent research. When sky-dreaming Manabe converged with a town famed for its scientific edge, climate prediction was born. 

Manabe, now a resident of Stonebridge Senior Living in Montgomery, shared the 2021 Nobel Prize in Physics for his work. Over his lifespan, fossil fuel combustion has driven up the concentration of atmospheric carbon dioxide (CO2) by roughly a third. Our planet is more than two degrees Fahrenheit warmer, and climbing. With uncanny foresight, Manabe predicted it all.

Nobel Prize medal. (Courtesy of Nobel Prize Outreach. Photo by Clément Morin)

The Infinite Forecast

As a boy, this path was not obvious. Despite coming from a family of physicians, Manabe did not seem cut out for science. He was, by his own admission, a clumsy child with little interest in memorizing facts. What fascinated him instead was weather.

Powerful typhoons, known elsewhere as hurricanes or cyclones, slam the islands of Japan. They bring lashing rain and widespread destruction. Manabe wondered: Could they be predicted?

In the 1940s, little was known about atmospheric phenomena. Weather forecasting was not accurate, and Japanese households tended to rely on almanac-style publications instead. For Manabe, this lack of knowledge was a lure.

In 1949, he enrolled at the University of Tokyo to study meteorology. By then, something intriguing was happening across the Pacific Ocean. In Princeton, Hungarian polymath John von Neumann had developed one of the first electronic computers at the IAS. The machine had just convincingly predicted weather 36 hours in the future, a stunning achievement for the time. Von Neumann next asked if the machine could run models for a longer period. Dubbed the “infinite forecast,” he pitched the idea to the U.S. Weather Bureau, which agreed to fund the research. 

Manabe, left, works on research with Joseph Smagorinsky, the first director of GFDL, in the 1970s. (Photo courtesy of GFDL)

Von Neumann appointed his deputy as head of that unit. Joseph Smagorinsky, an ambitious young American meteorologist, had a head for math and a sense of what to do with it. He was determined to make the first numerical calculations of long-range weather. This research was to begin at the Weather Bureau but would eventually shift to the National Oceanic and Atmospheric Administration (NOAA), be renamed the Geophysical Fluid Dynamics Laboratory (GFDL), and relocate to Princeton University’s Forrestal campus. 

Yet before that could happen, Smagorinsky needed the talent. He had been tracking the work of Tokyo’s elite research meteorologists, particularly a quiet young man whose intuition graced everyone else’s papers. Even before Manabe had finished his dissertation, Smagorinsky recruited him as a visiting scientist to the Weather Bureau, then based in Washington, D.C.

Many years had passed since the war ended, but Japan’s economy was still on its knees. Job prospects for graduates were dismal, as was the funding for research and technology. Indeed, the U.S. was the only nation with computers powerful enough to process the ideas Smagorinsky hoped to develop. Understanding this, and despite speaking little English, 27-year-old Manabe accepted Smagorinsky’s offer. He joined an exodus of scientists headed for the U.S. 

In September 1958, dressed in a dark suit and tie, Manabe flew through the troposphere on a propeller plane toward Washington, D.C. That year, a scientist at the Mauna Loa Observatory in Hawaii, Charles David Keeling, had begun measuring the amount of carbon dioxide in the atmosphere. Keeling’s recording station registered an average of 315 CO2 molecules per million molecules of air (ppm). Manabe passed through these molecules — a trace atmospheric gas few passengers were thinking about — en route to his new life. 

(Illustration by Johan Jarnestad/The Royal Swedish Academy of Sciences)

Holding a Mirror to the Earth

Imagine throwing a giant net over the Earth. It creates gridlines covering the planet. Now stack those nets from the ocean depths to the atmospheric edge. The result is thousands of fictional cubes, several layers deep. These cubes, called grid cells, are the climate modelers’ playground.

Climate modeling — Manabe’s field — is the science of simulating the Earth’s atmosphere. Each grid cell offers a digital analog to a real place on Earth. Modelers use mathematical equations to describe how effects such as ocean currents, wind patterns, and warm air interact inside each cell. They also study how cells interact en masse. 

These models aim to predict the way our swirling atmosphere affects the weather tomorrow, or the climate decades into the future. They can also be used to pose hypotheticals. What happens if we remove the forests from this digital planet? What if we warm up that ocean? What if we pump in more CO2? 

When Manabe was in graduate school, only a handful of scientists worked on such complex puzzles. Today, there are thousands, comprising physicists, oceanographers, and mathematicians. There are now several climate modeling shops globally, all working on slightly different models. One of the most prestigious — the GFDL — is in Princeton.

In the bucolic surroundings of Princeton’s Institute Woods, von Neumann imagined being able to forecast long-range weather. Smagorinsky — who as an Air Force meteorologist had made weather observations from the nose of bomber planes — understood how to unlock this dream. With differential equations and powerful computing, Smagorinsky’s team would create a numerical model that so accurately captured atmospheric interactions that it would be like holding a mirror to the Earth. Manabe came to Princeton to help make those dreams come true.

But the intellectual task was daunting. The imaginary net he and Smagorinsky cast over the globe was rudimentary at best. It only partially covered the world, and the grid cells themselves were huge.

Their ambitions were limited by their hardware. They were working with an IBM 701, a high-speed (for the time) calculating machine. There were only 19 such devices in the country. Though considered the acme of technology, the computer had limited memory — a resource Manabe’s formidable calculations would put to the test.

The machine ran on fat stacks of punch cards; each clipped with holes denoting mathematical values the computer would read as instructions. Manabe programmed these cards himself. 

But given the complexity of the calculations, programming was a painstaking task. Manabe had to define each grid cell using lines of latitude; then define the values inside each cell — such as temperature, air pressure, or wind; and then program the instructions for how each value would change. His efforts resulted in thousands of data points grinding away for every moment in measured time. 

It took seven years for Smagorinsky and Manabe’s first numerical prototype of the Earth to be ready. Known as the general circulation model, it created a framework for how the atmosphere flows over a moving globe. It was hardly headline-making science, yet it set the scene for significant future achievements. The model included radiation and convection — two complex processes — allowing Manabe to begin “unraveling the mystery of climate” as he later described the process to a reporter. Now the fun could begin. 

The CO2 Jackpot

By now it was 1965. Men’s hair was growing, women’s hemlines were rising, and Bob Dylan had just switched to an electric guitar. But for Manabe, the fun was unfolding inside his lab. Meanwhile, CO2 had ticked up to 320 ppm.

To refine his general circulation model, Manabe tinkered. Too few atmospheric variables and the model would be inaccurate. Too many and the model would crash. For Manabe, who loved detail-oriented challenges, the uncertainty was the prize. He relished experimenting with the various components of the atmosphere — clouds, water vapor, ozone, greenhouse gases — mixing them up like a chemist. 

“I was playing and enjoying myself,” he later told a journalist.

But he realized that to capture the complexity of radiation and convection, two vital atmospheric processes, he needed to lean toward simplicity. Accordingly, he devised a mini model that only captured the vertical movement of energy through the atmosphere. 

Moreover, Manabe had a hunch that carbon dioxide might be more influential than thought. Scientists had long speculated about its warming effects, so for curiosity’s sake, Manabe adjusted his model’s variables. When the computer spat out the result, the output was astonishing. 

After Manabe doubled the CO2 in the atmosphere, the digital atmosphere warmed about three degrees. It was the intellectual version of hitting the jackpot — though no one knew it at the time.

Nadir Jeevanjee, an atmospheric scientist at GFDL whose work on radiation, convection, and climate is directly informed by Manabe’s insights, explained that this intuition is what made Manabe’s work so groundbreaking. 

“That number is still our best estimate of how much the Earth will warm due to the doubling of CO2,” he said. “When Suki was putting together his simplified radiative-convective model, it turns out he was making just the right assumptions that stand the test of time.”

A Countryside Move

In 1968, the GFDL moved to Princeton. Smagorinsky had cut a unique deal with Princeton University to create a geophysical fluid dynamics program for graduate students. Manabe and his colleagues, who would remain federal employees, would train the students and serve on Princeton University faculty.

In exchange, GFDL scientists would be part of an intellectual environment that afforded collaboration opportunities with the academic community. Their prime goal was experimentation. The task was to develop climate models with no precise application in mind, in the understanding that an intellect freed from pressure would allow the imagination to soar.

By now, Manabe had married Nobuko, whom he had met in Tokyo in 1960, and had two daughters, Nagisa and Yukari. Their move to Princeton offered a stark contrast from Washington, D.C., which had been roiled by youthful discontent with the establishment. Yet in Manabe’s new hometown, university students politely protested the Vietnam War in a jacket and tie.

“Princeton was a countryside town,” Manabe recalled. 

That pastoral feel may have been enhanced by the location of GFDL’s office, which had been built for the team in the still mostly rural Forrestal campus four miles from Nassau Hall. From Manabe’s glass-fronted corner office, he gazed out at woodlands and fields.

The building was an architectural gem. Sleek and understated, its modernist array of reflective boxes conveyed the sophistication of a serious research institution. Manabe recalled being warned by a fellow scientist that this building was so luxe they would grow comfortable and stop producing good work. Unlike most forecasts coming out of the building, that prediction was wrong.

Instead, Manabe’s CO2 discovery continued to power his inquiries. Could he pair this new insight with increasingly complex algorithms showing how energy moved through the atmosphere as well as the oceans? 

At that time, his older daughter, Nagisa, then 5, recalls watching him carefully as he worked. She confidently informed her kindergarten friends that her father was an artist. At home, she had seen him adding bright spikes to the monochrome acetates he used for presentations. At his office, colleagues would huddle round to discuss his colorful chalkboard designs.

“I thought my dad was a coloring specialist,” Nagisa says.

She did not learn until later that her father’s colorful spikes indicated the expected temperature changes as CO2 levels rose — something he would soon show in a simulation of the globe that now included oceans.

In 1969, in partnership with his longtime collaborator, Kirk Bryan, Manabe debuted the first climate model to include the laws of geophysical fluid dynamics applied to both the atmosphere and the ocean. Despite only covering a sector of the globe, the model expressed the way water and air flow together on a spherical shell on a rotating planet. Crucially, the model worked.

Manabe speaking at Princeton University. (Princeton University, Photo by Denise Applewhite, 2021)

Melting Ice

Yet the purity of Manabe’s theorems was pitted against the dust of reality. The computing, though increasingly powerful, struggled with the intricacy of his calculations.

By now, computers had moved from punch cards to magnetic tapes, but even so, it took years for the machines to run his calculations. Operators would retrieve his tapes from storage, blow the dust from them, and jam them into the computers. Often they failed to work, and had to be sent away for cleaning. Manabe’s next big discovery emerged in 1975, but the work had been completed five years earlier. The computer had taken more than two years to run the tapes and a further three years to analyze them.

Nonetheless, the more Manabe refined his calculations, the more confident he was of the global warming thesis. His 1975 scoop with colleague Richard Wetherald confirmed it. Together they showed how increased CO2 would cause the Arctic to warm faster than the rest of the globe. This process, now known as Arctic amplification, predicted that the polar ice cap would melt.

By then, atmospheric CO2 was at 331 ppm. The world was still decades away from the geopolitical scuffles that would emerge because of Arctic amplification — rising sea levels, a navigable Northwest Passage, and Greenland as a presidential bauble. Yet in the mid-70s, that course had already been set, and Manabe had foreseen it.

By the late 1970s, a scientific consensus was building, and a new term was entering the scientific lexicon: global warming. By then, the Manabe family had moved to a stately yellow house on Princeton Avenue. Manabe was frequently lost in his models, his imagination compelling him toward ever more ambitious calculations. 

He would drive the winding country lanes to GFDL but was often so lost in thought he would stop paying attention to the road. He had several crashes and at one point, he says, his license was suspended. Reflecting on those moments recently, he mused how risky it was for his family to ride with him. “I’m glad they’re still alive,” he said with a chuckle. 

Meanwhile, Nagisa and Yukari had figured out a trick. They could delay bedtime by asking Manabe about his work. He would use a lamp and a globe to demonstrate the warming effects of the sun on a world increasingly changed by CO2. 

Yukari would grow up to become a scientist herself — a physician specializing in infectious diseases. Nagisa would go on to own River Stoan Farm, an organic farm in Kingston, where the growing seasons would gradually shift as a result of that warming sun. 

Notably, by the time Nagisa and Yukari were graduating from Princeton High School in the early 1980s, carbon dioxide concentrations had spiked to more than 340 ppm. Manabe’s simulations had been predicting it for years. Would anyone pay attention? 

A Climate Warning

In 1988, a severe drought gripped North America, sparking wildfires, dust-storms, and successive heatwaves. Congress set up an urgent hearing for scientists to tell legislators what they knew. James Hansen of NASA was punchy when discussing the link between climate change and the buildup of greenhouse gases.

“It is time to stop waffling so much and say that the evidence is pretty strong that the greenhouse effect is here,” the New York Times reported him as saying. To this day, when people talk about when global warming entered the public consciousness, they often refer to that moment.

In addition to his own research, Hansen had also been relying on predictions from Manabe’s simulations to make these claims. In fact, Manabe also spoke to the same committee. Yet, unlike Hansen, he failed to make an impact. 

“They weren’t too impressed by this Japanese guy who has this accent, whereas Jim Hansen made a bombshell impression,” Manabe said later.

Scientists are inherently cautious. Manabe, more comfortable at his chalkboard than delivering real talk to politicians, did not want to speak in definitives. He accepted that public communication wasn’t his strength. Besides, his curiosity about the workings of climate was tugging him toward what would come next. 

Now that the broad strokes of climate change had been painted, Manabe wanted to understand the impacts. How would river discharge be affected? In what way would soil moisture change? Manabe left it to others to convey the urgency of what his models showed. He went back to work, peeling more layers from the onion. By then, global CO2 had climbed to 352 ppm.

World Recognition

In 1990 the first Intergovernmental Panel on Climate Change (IPCC) report landed with a slam. It was a direct response to the 1988 Congressional hearings, and it quickly established scientific legitimacy for climate science. Manabe’s work was integral to the report.

By then, CO2 had spiked to 354 ppm. The global temperature was almost half a degree warmer. 

Manabe was not surprised. He could already see the effects at home. Princeton University’s Lake Carnegie was no longer a reliable skating pond. When the Manabe family had arrived some 20 years earlier, he recalled skating on the lake with his children and resting on the ice when he grew tired. But from 1980, the probability of finding thick enough ice upon which to glide or sky-gaze had shrunk.

In 2007, Al Gore shared the Nobel Peace Prize jointly with the IPCC for their efforts in raising awareness about climate change. As one of the scientists whose work underpinned the report, Manabe received a citation. Between 1992 and 2018, he would go on to garner ever more prestigious awards, including the Crafoord Prize in 2018, given to scientists working on groundbreaking work in fields not yet recognized by the Nobel Committee. 

Meanwhile, Manabe continued to make forecasts. As global warming accelerated, his predictions from decades before were proven again and again. Yet he shied away from queries about what he thought humans should do. I’m not a crusader, he would say.

In the early morning of October 2021, Nobuko took a phone call. It was from the Nobel Prize Committee in Stockholm announcing that Manabe had been jointly awarded the Nobel Prize in Physics for his work on climate modeling. Manabe was shocked. Not only was this the first time the Nobel had been given for applied physics, but also, why should he be the one to deserve it? He had merely been having fun and seeing where his curiosity took him.

Still Questioning

Today, NASA data shows that the 10 most recent years are the warmest on record. At the time of writing, atmospheric CO2 is at 429 ppm. Yet the scientific consensus about climate change is again under attack. In a bid to undercut the experts pointing toward the main cause of global warming — the combustion of fossil fuels — the current U.S. administration is instead withdrawing from science. 

Yet while climate models are not perfect, climate scientists have come a long way from the days when young Manabe was drawing pressure charts by hand. Today, in the U.S. alone, climatologists crunch more than 210 million observations a day to produce forecasts for hurricanes, severe weather, air quality, and the future climate. The output is used not just by citizens for their weather apps, but also by public health experts, military planners, and the logisticians in charge of moving cargo around the world.

Moreover, as the policies unleashed by today’s leaders continue to belch carbon dioxide into the atmosphere, those climate models — the descendants of Manabe’s foundational work — will continue to predict the outcome. 

That outcome need not necessarily be dire. When von Neumann first imagined Princeton as a place to study atmospheric processes, he saw it as a forge for preparedness. He cautioned that even more than war or nuclear threats, climate problems could unite the interests of all nations.

Indeed, the GFDL still houses scores of weather and climate modelers working toward humanity’s long-term interests. Despite funding threats, the group continues to reveal insights about the Earth’s atmosphere and oceans, and to train new generations of Earth scientists.

Meanwhile, Manabe is still gripped by the curiosity that powered his career. When recalling his life experiences for Princeton Magazine, he again referred to his fascination with weather. 

“Typhoons are the most interesting phenomenon in the atmosphere,” he said, adding with a playful smile: “But in the universe, the spiral galaxy and the hurricane look so similar. Why?”