Webb's Exoplanet Data Are Almost Too Good

Good science takes time.

That’s not a popular refrain in our 24-hours news cycle — especially when there’s a new space telescope returning crystal-clear views of the infrared universe nearly every day. But for astronomers drinking from the firehose that’s coming from the James Webb Space Telescope, time is exactly what they need.

Webb’s Early Science Release program has been delivering exquisite data on a variety of celestial targets. Some of the most anticipated of these are the spectra of exoplanets. But a study in Nature Astronomy urges caution in interpreting these chemical fingerprints of alien worlds.

Astronomers often gather exoplanet spectra as the planet passes in front of its star. As starlight passes through the sliver of atmosphere encircling the planet, the alien air leaves its mark, absorbing light at specific wavelengths corresponding to molecules in the atmosphere. One of the first pieces of data the Webb team released was such a spectrum of WASP-96b, which showed clear wiggles and bumps indicating the presence of water vapor in the hot giant’s atmosphere.

That detection, says Julien de Wit (MIT), who led the Nature Astronomy study, isn’t in question. “There’s no problem related to that first level of interpretation,” he explains. It’s the second level, for example, how much water, where things get complicated.

That’s not the fault of Webb: The spectra it has captured so far contain more detail than we’ve ever had before. Now it’s the theory that the data are compared against that needs improvement.

Prajwal Niraula (MIT), de Wit, and colleagues found that the models astronomers use to decode such spectra are imprecise. For example, one group of astronomers might use the models to say that an exoplanet’s atmosphere is 5% water; another group might find it’s 25% water. And it would be difficult to say who’s right.

Those challenges extend all the way down to rocky exoplanets. Several hot, Earth-size(ish) worlds will be targeted as part of Webb’s first year of observations. “We do expect the limitations we've highlighted to also affect our study of rocky exoplanets,” confirms Niraula.

“To date, no studies have been published regarding the interpretation of Webb exoplanet spectra, so our study is timely,” de Wit says. He adds that that doesn’t mean we should stop taking data: “We do need to keep on gathering this data even if it takes us a couple of years to fully decrypt what is encoded in them optimally.”

How long it takes to get models up to speed to accurately interpret Webb data depends on the molecule in question. “For instance, for carbon monoxide (CO), there's not much work needed at the level of accuracy required by JWST,” says team member Iouli Gordon (Center for Astrophysics, Harvard & Smithsonian). For water measurements in deep atmospheres, though, lab measurements and theoretical calculations will take a few years to catch up to observations. Some molecules, like ethane, will need a lot more work.

Caroline Morley, a theorist who specializes in studying exoplanet atmospheres (and who wasn’t involved in this study), agrees the study is coming at a good time. “This is the kind of thing that practitioners of the field know very well, but users of the models (i.e., observers fitting their data) perhaps do not,” she says.

But she thinks we might not need to wait so long to begin understanding what Webb is telling us about exoplanets, because at least one approach to dealing with these challenges is already in use: Comparing planets with their bigger cousins, brown dwarfs.

Brown dwarfs are massive worlds (more than 13 Jupiters) that cannot sustain fusion in their cores; in other words, they’re failed stars. Yet these stellar wannabes act a lot like gas giants, with similar composition and other characteristics. “They have the same temperatures as exoplanets,” Morley notes. “Temperature is the main planet property that will change the molecules that are present and the shapes of the molecular bands we observe.”

And past studies comparing lab-based measurements with brown dwarf spectra have shown very good agreement. “My instinct is that the conclusions are therefore a bit overstated,” Morley says. “Nonetheless the paper is a nice contribution to the field, and hopefully helps get some of this [modeling] work (which is ‘under the hood’ work that often doesn’t get the recognition it deserves) funded in the coming years.”

This issue is far from the only challenge facing those using Webb data. Some have found that the calibration of brightness in individual wavelength bands changed after Webb was launched, which can affect distance estimates for far-away galaxies. A post-launch study also found that one of the large hexagonal mirrors will occasionally tilt a tiny bit (up to tens of nanometers, much less than the width of a human hair), causing changes that can, for example, affect observations of an exoplanet crossing the face of its star.

These are known issues, though, and while some will resolve on their own (tilt events are supposed to decrease over time as the telescope’s inner stresses relax), astronomers are working hard to sort out the others. All they need is a little time.