Weathering New Worlds

I attended ExoClimes 2010, at the University of Exeter, a workshop about the diversity of planetary atmospheres within our solar system and beyond. The cliché forecast for southwest England had me prepared for rain, but it stayed hot and sunny. Weather prediction will always be unreliable, but how well can we predict climate change? Increasingly, we’re banking on accurate climate models for the future well being of civilization. If our models really are any good, shouldn’t they also predict the climate on other worlds?

The results are mixed for our neighboring planets. We still can’t say why Mars started off warm and wet. Clearly, its atmosphere was thicker, but if we just pump up the carbon dioxide in our models, it never gets hot enough. We understand, in terms of basic climate principles, why Venus’s thermostat became stuck at the hottest setting as the warming young Sun triggered a runaway greenhouse, but we don’t know when this catastrophe occurred.

Planets are diverse and quirky, so three or four examples are insufficient to prove any pattern or improve our models. We need many more examples, and fortunately, we’re about to get them as we slowly learn more details about exoplanets. The only ways we have right now of gaining any hints is through painstaking analysis of datasets at the hairy edge between noise and meaning. Scientists (many of them young) are displaying a fantastic amount of cleverness to bravely try new ways to use the available tools to glean information about these distant worlds. 

Meanwhile, modelers are trying to understand possible exoplanet climates. At the workshop it was fascinating seeing astronomers trying to talk to terrestrial climate modelers. Our knowledge of exoplanet climates is so sparse. Each is at best a few numbers — mass, distance from a star, and in some cases, vague inferences about temperature or atmospheric composition. In contrast, Earth climate models are supplied with dense grids of data — millions of points of temperature, humidity, and wind velocity. The mismatch in perspectives made communication challenging. 

Climate physics must be universal even if local conditions are infinitely variable. When we predict Earth’s future climate, we change a few variables such as the abundance of greenhouse gases, and the model tells us what the new temperature pattern will be. A similar exercise should also work for another planet: take a garden variety Earth climate model, alter the gravity, the atmospheric composition, or the amount of sunlight, and you should get results that match observations. Unfortunately, it’s not so easy in practice. Our best models are Rube Goldberg contraptions cobbled together from older models, modified and tweaked to add new physics, assumptions, and constraints. Nobody really knows how they work. 

Efforts are underway to develop more universal climate models that will work for Earth as well as for Mars, Venus, Titan, and all the coming exo-Earths with arbitrary combinations of planetary variables. Exoplanets are a transformative watershed for planetary science. They will test our theories and deepen our insights in so many ways that have been impossible when we knew nothing about other planets.

Earth climate modelers might be slow to realize it, but they need exoplanets. During the workshop I felt like I was seeing a future era of climate understanding that will be made possible when we can study our planet’s qualities in the context of thousands of its planetary peers.

This article originally appeared in print in the January 2011 issue of Sky & Telescope. Subscribe to Sky & Telescope.