Using an Atmosphere to Weigh a Planet

Reading an exoplanet’s atmosphere from the spectra of light that passes through it. Image Credit: ESA

A new study finds that the mass of an exoplanet can be determined solely by looking at the starlight that passes through the planet’s atmosphere.

When sunlight streaming through a planet’s atmosphere reaches our telescopes, the light’s spectra acts like a calling card for the different gases that make up that atmosphere. Water vapor has one type of spectral signature, for instance, while carbon dioxide has another. In that beam of light, astronomers can even see how many molecules of each type of gas are packed at different altitudes in the planet’s atmosphere, indicating whether it’s a diffuse and airy halo, or a dense shroud of high pressure.

The new method, called MassSpec, uses this information about a planet’s atmosphere to determine how heavy an exoplanet is. Knowing the mass can indicate whether a planet is a Jupiter-like gas giant where life as we know it is not possible, or a rocky world more similar to Earth.

“The first step is to characterize the atmosphere,” says Julien de Wit, a PhD student at MIT and the study’s lead author. “Once you know what’s there, then you can tell more about what’s below the atmosphere — in other words, the mass of the planet.”

The method currently can be used to determine the mass of gas giant Jupiter-like planets. But with the upcoming James Webb Space Telescope (JWST), it also should be able to determine the mass of Earth-like planets. The scientists say their method also can complement or supplement other methods that astronomers normally use to determine a planet’s mass. These other methods for determining mass run into trouble for low-mass planets, planets orbiting far away from their stars, or planets orbiting faint or active stars.

“Determination of an exoplanet’s mass is a key to understanding its basic properties, including its potential for supporting life,” writes de Wit, along with his co-author Sara Seager, a professor of physics and planetary scientist also at MIT.

The study was published in the Dec 20 edition of the journal Science.

A Planet’s Silhouette

Astronomers have detected more than 1,000 exoplanets to date, and while many are massive gas giant planets like Jupiter, quite a few are smaller, rocky worlds like Earth.

Artist’s impression of a planetary transit, and the resulting dip in the light curve. Image Credit: NASA

But how do astronomers know when they’ve found a Jupiter or an Earth? Nearly every exoplanet has been found by studying how stars react to planets orbiting around them. “Radial velocity” measurements look at the gravitational dance between a star and planets as they orbit each other — the star’s very slight movement back and forth in the sky can be measured, and based on the amount of wiggle the mass of the star’s dance partner (or partners) can then be determined.

Some planets cross in front of a star from our point of view. When such a planetary “transit” occurs, the planet blocks some of the star’s light from reaching Earth. This brief dip in starlight can then indicate the size of the planet. Sometimes, radial velocity and transit observations can be performed on the same planet, allowing astronomers to get the planet’s density and, from that, an idea of its internal structure.

But these methods have their limits, and can’t be used to find every planet out there. Radial velocity measurements tend to favor the discovery of massive Jupiters orbiting close to bright stars, because it’s easiest to see when such a planet exerts a hefty gravitational tug on its relatively lightweight stellar companion. If a star is very active, erupting coronal mass ejections or pulling other stellar shenanigans, that confuses the picture planet-hunters try to piece together from the data. Transit detections, meanwhile, depend on perfect geometry — if a star’s orbital plane is even slightly titled in a different direction from our viewpoint, we won’t be able to see any dip in starlight as planets rotate around their star.

Artist’s impression of HD189733b after the discovery of its blue color by Berdyugina et al. (2011) using polarimetry. Image Credit: Darya Rioa, Wikicommons

Even for the planets that are found with these methods, there’s only so much you can determine about the planet’s characteristics, like its mass, radius, or atmosphere. Without getting a photograph of a planet — and only a very few have been imaged directly as mere points of light — other methods are needed to tell whether any given planet has oceans, continents or even life.

A Glimpse of the Gas

Getting information on a planet’s atmosphere is perhaps the best way for astronomers to determine a planet’s habitability, given current technology. Astronomers learn about a planet’s atmosphere by simply measuring starlight: a star’s light will pass through a planet’s atmosphere during a transit, so if the viewing geometry is right, we can then read what sort of elements are in the atmosphere — gases like oxygen, nitrogen, or methane, for example — based on the spectra of that light.

MassSpec uses information gained from studies of a planetary transit — the planet’s size as well as its atmospheric temperature, composition, and pressure profile — to derive the planet’s mass.

The scientists applied MassSpec’s concept to the gas giant HD 189733b. This planet is a distant 63 light years away from Earth. One notable finding about this world is that it’s blue. The blue color was determined by polarimetry, the first time this method was able to see visible light reflected off a planet.

The planet is extremely close to its Sun-like star, completing one orbit every 2.2 days. (The planet actually has two suns; a dim red dwarf star also is a part of this system.) Being so close to its main star, astronomers assume the planet is tidally-locked, meaning it has a permanent day and night side (i.e., the same side always faces its sun). Even so, astronomers think the atmosphere circulates the heat, distributing it globally so there’s not one hot and one cold side.

Still, life as we know it would not exist on the planet. Being so close to its main star, the planet is blazing hot — measurements from the Spitzer Space Telescope give a temperature range of 940 to 1223 Kelvin (1,232 F to 1,742 F). Data from Spitzer allowed astronomers to make a brightness map of the surface temperature of this planet. de Wit says the temperature of the planet helps enhance the signal of the light spectra from the planet’s atmosphere.

“The hotter the planet, the puffier its atmosphere will be,” says de Wit, allowing more light to pass through the extended atmosphere for astronomers to analyze. But de Wit says they can look at more than just hot Jupiters — the MassSpec method can be used for cooler gas giant planet orbits orbiting much farther away from their stars, as long as the planet transits and has an atmosphere.

Follow-up studies of the gas giant HD 189733b have found sodium, potassium, water vapor, and possibly other gases such as carbon monoxide in the planet’s hazy atmosphere.

The layers of Earth’s atmosphere. What might the layers of an exoplanet’s atmosphere be like? Image Credit: Kelvinsong, Wikicommons

de Wit says they used this planet as their “proof of concept” partly because it is so well-studied. Their method pegs the mass of this planet at 1.15 Jupiter masses, which is very close to previous calculations that used radial velocity measurements.

The MassSpec method extracts a planet’s mass partly through its influence on the “atmospheric scale height”.

“Looking at the pressure profile of the atmosphere will give you an estimate of the mass that’s compressing the atmosphere,” says de Wit. That’s because the mass of a planet determines its gravity, and the gravity of a planet – along with its temperature and atmospheric “mean molecular mass” (the average weight of the constituent molecules of an atmosphere) – sets the stage for how the atmospheric gases are stacked from the ground up.

de Wit says a good future application of the MassSpec method could be to look at visible light data obtained by the Hubble Space Telescope for WASP 33 b, another gas giant that orbits extremely close to its star.

“WASP 33 b is the hottest planet known,” says de Wit. “But its star is so active we don’t have a constraint on its mass.”

Future telescopes like ESA’s proposed EChO mission also could provide the needed data to use this method on other gas giant and close-by super-Earth planets. And when JWST launches in 2018, the method could be used to quickly determine the mass on “trickier” worlds like potentially habitable Earth-sized planets.

However, Adam Burrows, professor of astrophysical science at Princeton University and head of Princeton’s astrobiology program “Planets and Life,” is skeptical that this method can be used to determine the mass of exoplanets as accurately as the authors claim.

“Their method depends on determining with precision a planetary atmosphere’s temperature and mean molecular weight, and assumes that the interaction of the stellar light with the atmosphere is understood,” he states. “As we are seeing now with numerous exoplanets — including HD189733b — such measurements all have a substantial margin of error, and therefore a lot of uncertainty. For instance, the presence of hazes at a potentially broad range of pressure levels is complicating the extraction of physical information.”

He acknowledges that the method might be helpful for planets where radial velocity techniques are difficult to make, but also notes, “I would not bet against the long-term usefulness of radial velocity as the technique of choice to obtain exoplanet masses.”

“Nevertheless, this paper highlights the regions of parameter space in which we may otherwise be blind, and the need to explore novel methods with which to constrain the physical properties of these exotic and intriguing exoplanets," he concludes.

Jonathan Fortney, associate professor of astronomy and astrophysics at the University of Santa Cruz in California, says the new paper is impressive, "but I think that clouds may confound the rosy picture they paint."

Fortney adds, "It is exciting that a new route to obtaining planetary mass has been found. I think that in practice, for the small planets, it may only be rarely used, but likely for the most interesting planets — those that are the targets of dedicated observing campaigns to obtain transmission spectra."