When astronomers discovered the first exoplanet, they were shocked to find it seven times closer to its star than Mercury is to the Sun. Its proximity to a blindingly bright star made it extremely difficult to study. But Seager knew that before long, someone would find an exoplanet that, when viewed from Earth, would transit in front of its star and then disappear behind it. (The probability of a close-orbit exoplanet transiting its star is about 10 percent.) Astronomers could then use a method devised in the mid-1900s for studying eclipsing binary stars: by measuring the slight drop in the star’s light when the exoplanet passed in front of it, they could calculate the ratio of the planet’s area to that of its star.
Sure enough, in 1999 astronomers observed the seventh close-orbit exoplanet transiting its star some 904 trillion miles from Earth. A gaseous giant with temperatures that can top 1,300 °C, the exoplanet HD 209458b is classified as a “hot Jupiter.” It’s also one of just two known transiting exoplanets whose stars shine enough light through their atmospheres to give astronomers the data they need to do detailed studies.
The only way astronomers can learn about the atmosphere of another planet is to study the planet’s radiative transfer, or the propagation of light through its atmosphere. But HD 209458b is so close to its star that it completes its orbit in three and a half days, and even the Spitzer–which detects infrared light to the level of 1 part per 1,000–can’t on its own distinguish the planet’s light from the star’s. So as part of a research group based at NASA’s Goddard Space Flight Center, Seager and her colleagues used a simple calculation to isolate the light of the planet. With the Spitzer’s infrared spectrograph, the researchers measured the light of the star and planet together (when both are visible) and subtracted the light from the star alone (when it eclipses the planet).
The spectrograph, which functions like a prism, also separated the planet’s light into its component wavelengths. The researchers analyzed the light at each wavelength, looking for features that would identify molecules in the exoplanet’s atmosphere. In the February 22 issue of Nature, they reported gathering the first spectral data from HD 209458b over the course of two eclipses in July 2005–and explained their somewhat surprising findings.
To do their spectral analysis–and even to decide what to look for in the first place–the researchers used models of possible exoplanet atmospheres that Seager had developed. When starlight shines on a planet, photons encounter molecules in the atmosphere; depending on what molecule it hits, a photon might get absorbed, scattered, or reëmitted at a different wavelength. By observing the photons that emerge from the atmosphere, Seager can determine what molecules the atmosphere contains. “Photons are our currency,” she says.