Is Climate Science Physics?
How we answer this question can help determine who understands, and who enters, the field.
Is climate science physics? As a theoretical physics graduate student at the University of California, Berkeley in the early 2010s, I hadn't given the question much thought. I read about climate change in the newspaper, but had the naive and condescending impression that Earth science was so much stamp collecting, and of little interest to a physicist.
That changed when I became a teaching assistant for Richard Muller's popular course, “Physics for Future Presidents.” The course opened my eyes to the physical underpinnings of atmospheric science, climate change, and energy technology. After years of wandering the wilderness of theoretical physics, I grew excited about the prospect of applying the physics I loved to front-page problems.
When an ex-physicist turned up in Berkeley's Earth and Planetary Science Department and started probing for recruits, I decided to take the plunge, leaving the lofty heights of quantum gravity for the earthlier realm of atmospheric science.
The first step was to learn the basics, and I reveled in the application of fundamental physics to the climate — Navier-Stokes equations to atmospheric flow, Planck's radiation law to the greenhouse effect. It was a thrill to learn that Newton's laws applied as well to trade winds as it did to point masses, and that thermodynamics governed phenomena from cloud drops to thunderstorms.
But as the basic coursework waned and I turned to research, the honeymoon ended. Culture shock set in. Even though climate science was based in physics, it didn't feel like physics. The papers we read sometimes had no equations at all, and the arguments felt, at times, loose and hand-wavy. My job was to run computer simulations of the tropical atmosphere, but I had only ever worked with pencil and paper; it took me a week just to make a plot, and twice that to get the model to compile on the supercomputer. The climate science literature was summarized in thousand-page reports by the Intergovernmental Panel on Climate Change, which focused less on physics and more on historical records and model projections. In some instances, different models produce very different projections, giving rise to "spaghetti plots" for variables like global mean surface temperature (Fig. 1).
I yearned for the elegance of physics, the Fermi problems and Feynman lectures, but found myself mired in spaghetti and Fortran instead. I wondered if I could muster the resolve to graduate, let alone continue in the field.
Eventually, however, I found my footing. Following the example of my advisor and many others, I learned to sniff out problems where back-of-the-envelope calculations could yield insight. (An early success: solving a Poisson equation that described how buoyant fluid parcels accelerate.) Obtaining results on my own gave me confidence that the gulf between physics and climate could be bridged, and that I could play a role.
I also realized that, on their own, these calculations were not enough. Unlike the idealized physical systems I studied as a student, Earth's climate is complex. With so many variables across so many scales, the climate behaves in ways that the underlying equations on their own can’t predict, well-known as those equations might be. So climate scientists have developed model hierarchies, an array of modeling approaches that link to one another like rungs on a ladder. The simplest models look like textbook physics, solvable with pen and paper; the most complex require the world’s largest supercomputers. For my physics-driven work to be meaningful, I had to connect it with the real-world complexity of comprehensive climate modeling.
Inspired, I crossed the country with my family to Princeton, New Jersey, to begin a fellowship at the Geophysical Fluid Dynamics Laboratory (GFDL). GFDL was an early nursery for climate science, and — in the 1960s and 1970s — produced the world's first climate models, pioneered by GFDL scientist and Nobel Prize winner Syukuro Manabe. Moving to GFDL meant working in The Room Where It Happens — where algorithms are developed and models run, and where the complexity of climate science is both a familiar fact and constantly debated.
Outside the lab walls, however, conversation was eroding. Princeton University itself was home to several skeptical physicists, and communication between them and Princeton's climate scientists, to the degree it had ever existed, had broken down. On the national level, apocalyptic doomsaying and righteous denial sailed past each other on the airwaves, and trust in science felt dangerously low.
These tensions reinforced my suspicion that the complexity of Earth's climate was an issue not just for climate modeling but also for climate communication — both for our colleagues in physics and for the public. Uncomfortable just burying my head in research, I wanted to try to reach both audiences.
In 2019, a few colleagues and I formed a group called Climate Up Close and made it our mission to speak with ordinary people, far from academic centers. We ventured out into the country, to central Pennsylvania and the Florida panhandle, and held events in churches, synagogues, and libraries. We tried to dispel misinformation, answer questions, and build trust with our willingness to listen. It was an effort to close the gap with the public.
But what about the gap with physicists? I wondered if the simple models my physicist's heart sought might also be a communication tool. If key aspects of climate science could be explained on a blackboard, à la Fermi and Feynman, could this help hesitant colleagues understand it? My mentors liked the idea, and encouraged me to offer a handful of lectures at Princeton on the "physics of climate." No PowerPoint spaghetti — just chalk on blackboard. The turnout was overwhelming, and made clear that many colleagues yearned to bridge this gap, too.
I realized that explaining climate science to the public, or to physicists down the hall, required starting where I had as a graduate student: With clear, simple essentials. Of course, in climate science, the reductionist approach so vaunted in physics must be married to an acceptance of the Earth's complexity — but to climb a ladder, it helps to start on the bottom rung.
While the lectures were a success, the pandemic put a halt to the outreach. My colleagues and I buried our noses in research. But then, a miracle occurred: The 2021 Nobel Prize in Physics was awarded to climate scientists, including GFDL's Suki Manabe, the father of climate modeling. Climate science now had pride of place in physics. GFDL scientists gathered for a week to celebrate, our pride compounded by the joy of gathering in person for the first time in a year and a half.
In the months after, I found myself invigorated. My mentors and I wrote a retrospective on Manabe's key paper, detailing his ingenious distillation of the essential elements of climate into a 1-D model that produced the first credible simulation of global warming. I turned back to my lecture notes for the "physics of climate," expanding them to show how Manabe's results could be sketched on a blackboard. I joined APS's Topical Group on the Physics of Climate, itself energized by the 2021 Prize and actively organizing events and lectures. The Nobel committee had sent a clear message to the world, and I was running with it.
So, yes — climate science is physics. Earth obeys the laws of fluid mechanics, thermodynamics, and radiative transfer, and these laws are encoded in the climate models that have led to consensus. But to understand the Earth system, we need that hierarchical ladder — models familiar to physicists on the bottom rung; models of intermediate complexity, like Manabe's 1-D model, on the middle rungs; and global climate models, like GFDL’s, on the top rung. Moving between rungs requires a physicist's penchant for identifying what is essential and what is not. And this ability to cut through complexity is also what’s needed to communicate climate science, to both the public and fellow scientists, and build trust and confidence in our results.
In this sense, climate change is a physics problem par excellence: take several fields of physics, link them across an enormous range of scales, and then parse what emerges. This is a challenge worthy not only of our best Earth scientists, but our best physicists. Recognition is spreading: For example, this year's candidates for the APS presidential line are both climate physicists.
But on the ground, obstacles remain. Climate scientists in the US are rarely housed in physics departments, and there is little support for young physicists to transition into climate science, as I did. But I’m optimistic that physics departments and funding agencies will also soon answer the call to bridge the gap — once they realize that climate science is physics, and that physicists themselves can help make that clear.
The views expressed herein are in no sense official positions of the Geophysical Fluid Dynamics Laboratory, the National Oceanic and Atmospheric Administration, or the Department of Commerce.
Nadir Jeevanjee is a physical scientist at the Geophysical Fluid Dynamics Laboratory at the National Oceanic and Atmospheric Administration (NOAA).