He wants to bring ideas from theoretical physics to the realm of environmental science—a bridge that he hopes will enable scientists to better understand the climate system and predict the changes that are headed our way.

Marston explains that the climate system lends itself well to study within the framework of a branch of his discipline called statistical physics.

"Basically, climate is the statistics of the weather. And in statistical physics, which is a big area of physics, we try to find the statistics of whatever we're studying—like a gas—directly, without trying to model the motion of all the individual molecules," he says. "The hope is that similar thinking can lead to a better understanding of the climate system, and also maybe a more efficient or more accurate simulation of the components of the climate system."

One of the ways that climate phenomena are different from other statistical systems, however, is that they have intrinsic structure; that is, the random and uniformly chaotic movement that characterizes molecules of a gas in statistical physics is at odds with the comparatively organized arrangement found in climate features such as clouds, or the jet stream.

But despite this difference, Marston believes that he is able to identify ways to describe these elements in physical terms by looking for similarities between their structures and other, better-understood physical phenomena.

"Once we recognize these structures," he says, "then there's some underlying simplicity that can be exploited."

This is not Marston's first foray into physical pattern-seeking. He describes the moment when he realized that his postdoctoral work on atomic scattering bore a striking mathematical similarity to his seemingly unrelated, more recent work: modeling flows in the atmosphere and oceans.

"In physics, there's often a recapitulation of the same themes," explains Marston. "They reappear in different areas, and it's kind of surprising somehow that that happens. It just seems that the natural world is well-suited, or at least some elements of it, to a mathematical description, and sometimes the same mathematical description appears [over and over again]."

Marston is currently chair of the American Physical Society's Topical Group on the Physics of Climate, a relatively new unit that he hopes will encourage more physicists to start working on the problem of climate change.

"I think it's the number one problem facing humanity," he says. "We don't have any viable path right now to stabilizing the climate, and I don't think people realize how bad things might get."

Marston is hopeful that a more physically rigorous understanding of climate phenomena will strengthen climate models and enable scientists to make more accurate predictions of climate change—especially at the regional level, where uncertainties abound.

"That's a long-term project," he says. "And maybe all the climate change will occur before we actually are able to predict it."

He laughs ruefully. "But we try to do what we can."