It's all in the atmosphere: exploring planets orbiting distant stars

The second of a two part series that looks at what astronomers can find out about the planets that are discovered orbiting other stars in our galaxy.

When the Moon passes through the Earth’s shadow during a total lunar eclipse, it isn’t plunged into darkness, but instead takes on a ruddy hue. This is caused by the sunlight passing through our planet’s atmosphere, and being refracted into Earth’s shadow.

When a planet transits its host star, the main bulk of the planet obscures the star’s light, causing the wink used to detect it. As with the total lunar eclipse, a small proportion of the star’s light will pass through the planet’s outer atmosphere, en-route to Earth.

As a result, we can use that transmitted light to reveal the nature of the planet’s outer atmosphere.

As the star’s light passes through the exoplanet’s atmosphere, the molecules that make up that atmosphere absorb some of the light. Each molecule absorbs only at certain specific frequencies, giving a unique spectral fingerprint.

But the contribution of the planet’s atmosphere is tiny, almost imperceptible. To reveal it, astronomers obtain spectra of the star before, during, and after the transit.

They can then subtract the out-of-transit spectrum from the in-transit spectrum, removing the star’s contribution, and leaving just that of the planet itself.

This technique is called transmission spectroscopy, and has already proved hugely successful. It has allowed astronomers to detect a wide variety of chemical species in the atmospheres of hot Jupiters (including water and carbon dioxide).

The spectra also carry hints as to the structure of the planetary atmosphere, allowing astronomers to find evidence of haze, cloud layers, and even temperature inversions on these distant worlds.

Studying planets by the light

As well as studying the atmospheres of planets by transmitted light, it is also possible to study the light the planets reflect back from their host stars. And some planets are even hot enough that we can detect the light they emit themselves.

To observe these thermal emissions, and starlight reflected from a planet’s atmosphere, astronomers employ a technique called emission or occultation spectroscopy.

Here, instead of observing spectral absorption features in the starlight after it has passed through the planet’s atmosphere, astronomers directly measure the infrared light being emitted and reflected by the planet.

To measure the thermal emissions, a star is observed just before its planet moves into its secondary eclipse (when the planet goes behind its host). It is then observed during the secondary eclipse, and finally after the secondary eclipse is complete.

Astronomers then subtract the in-eclipse observations from those taken before and after.

So when we subtract the light observed during the eclipse from that observed just before or after, we are left with just the component reflected, or emitted (if the planet is hot enough) by the planet itself.

As was the case with transmission spectroscopy, the emission spectrum of a planet can reveal its atmospheric composition. Even the colour of an exoplanet tells us something. Observations of the hot Jupiter HD 189733b, for example, revealed the planet was a spectacular cobalt blue.

Just like the Moon and Venus, these planets exhibit phases. Thus, if we extend this technique away from the time just before and after secondary eclipse, we can learn even more about the planet. As it moves around its orbit, a planet will turn from showing its night side to its day side.

So if we can measure the planet’s temperature (its brightness, in the infrared) as it orbits its star, we can map the distribution of day-night temperatures across its atmosphere.

Such temperature maps can reveal whether heat from the day-side is being redistributed efficiently to the night-side of a tidally locked hot Jupiter, revealing the planet’s weather.

Some hot Jupiters have been shown to have large temperature contrasts between their day and night sides, suggesting inefficient heat redistribution.

The search for life

To date, most of the atmospheres we have studied are those of the easiest targets, the bloated hot Jupiters. But as techniques have improved, and new instruments have come online, astronomers are probing smaller and smaller worlds.

In doing so, they are preparing the tools we will one day employ to search for evidence of life beyond the Solar system.

In the coming years, it is likely that we will discover the first truly Earth-like planets orbiting other stars, and the search for life upon them will begin in earnest.

The first targets will be small, rocky worlds. They will orbit at just the right distance from their star to potentially host liquid water on their surfaces. By the time we’re ready to search for signs of life upon them, we will have found tens, or hundreds of possible targets.

Those planets will then be whittled down to the most promising few on the basis of a wide variety of factors, from the star they orbit and the system in which they move to the precise nature of the planet itself.

Once the best targets have been chosen, the observations will begin. To determine whether the targets are truly habitable will require their atmospheres to be studied in detail. By using the same techniques that currently reveal the nature of hot Jupiters, astronomers will uncover their composition and climate, before going on to search for any evidence of life (biosignatures).

To carry out these observations of such tiny and distant worlds presents an enormous observational challenge for astronomers. At visible wavelengths, an Earth-like planet would be ten billion times fainter than its host star. Even in the infrared, where things are a bit better, the planet would still be ten million times fainter.

The James Webb Space Telescope (JWST), scheduled for launch in 2018, should enable astronomers to study the atmospheres of nearby super-Earths (planets a bit more massive than the Earth), another step along the road to truly habitable planets.

But to characterise a true Earth analogue will require a different approach. At the moment, it seems likely that direct imaging will be the only viable option. And as of yet, such observations are beyond us.

See also, part 1: It’s all in the rotation: exploring planets orbiting distant stars