Robert Pincus: Cloud Story

Cloud Story
Robert Pincus
This article appeared in the Spring 2000 issue of the UHS Journal, a newsletter put out by the Alumni Relations office at San Francisco University High School.

Abstract

Why study clouds? The short answer is that clouds influence the exchange of water and moisture between land, ocean, biosphere and atmosphere, and these exchanges are what drive both short-lived weather and long term climate. Will it rain this week in the Central Valley? How might the climate change in the next twenty years as a result of human activities? We can't make accurate predictions until we understand clouds and the way they interact with the rest of the world.

I have vivid, sensual memories of the fog in San Francisco, of walking through the mid-summer cold and dampness, seeing halos around car headlights. I eventually went to graduate school to study clouds, but it was years before I recognized that growing up with the Bay Area fog had formed the intellectual germ of a career. In my apprenticeship years I learned how light filters through innumerable tiny drops or crystals, what changes it brings about in the cloud by its passing, and how to glean knowledge about the cloud from the light which emerges. This is the received knowledge of my profession, the background each new member struggles to learn and understand.

Science is storytelling. The stories have to follow certain rules (they can't contradict observations, for example, and they should be quantitative), but a scientific theory is valuable only if it provides a narrative that helps make sense of the world. When science is taught in school the provisional nature of the narrative is glossed over and the tales are presented as strict reportage. But more than one description may make sense (light can be both wave and particle), and no single story explains everything. Scientific research is the processes of finding new ways to look at the world and weaving an explanation that encompasses a broader view. Sometimes a new story comes from noticing what people have been ignoring. In young areas of science like mine the stories can change pretty rapidly.

To relate the whole history of even a single, simple cloud would take a lifetime. Each is composed of an astronomical number of nearly invisibly small drops: a volume of cloud just ten yards on a side contains about as many drops as there are stars in the galaxy, but each is only a thousandth of an inch across. Elaborating the chronicle of every drop simply isn't possible. Instead, we tell abridged, abstract versions of the cloud's story, focusing on different details in different contexts in a balance between what we need and what we can do. Satellite meteorologists may describe a cloud in terms of the temperature at the cloud top and the total amount of water in each cloudy column; the climate modeler might refer to the same entity in terms of the concentration of liquid water and amount of cloud at each location; both understand that such terse characterizations leave much to the imagination.

In any endeavor it is easy to confuse description with reality, to see only what you know how to express. When you look at clouds over the ocean near the California coast, do you see the uniformity or the variability, the sameness or the differences from place to place? For many years scientists admired these clouds for their constancy: a good fable about low clouds over the ocean could apply to large portions of the ocean surface. But during the last decade a small group (myself included) have been rethinking the way we describe clouds. We've been paying more attention to the subtle variety, spinning yarns about the ways clouds change from place to place, and trying to see if admitting that clouds are ever-changing is worth all the extra work it requires.

The computer programs used to predict the weather and future climate are based on a schematic description of the world. The models divide the planet into discrete areas, implicitly assuming that the atmosphere within each grid cell is uniform. But real clouds aren't constant inside the coarse blocks the model delineates, and the predictions made using these chunky cartoonish clouds are often in error. In other words, by ignoring clouds' small-scale texture we're likely to make mistakes in large-scale cloud predictions. Prompted by more powerful computers and novel ways of observing clouds, my colleagues and I have begun to change one of the state-of-the-art climate models to include an account of this rich structure. The new view encompasses both the average and the variations, and it seems as if the model will become more accurate and more robust. As we've thought about this approach we've discovered that there are many things we don't know about clouds' gentle variations. What we need to do now is learn to see what we've ignored in the past and to find the means to describe it.

I like my job for a lot of reasons: it is intellectually satisfying, it benefits people besides myself, it allows for a mix of solitary and social work. What I most love about the work, though, is that it allows me to spend my time living in the physical world and telling stories about what I perceive. And the more I learn to see, the more beautiful the world appears.