Habitable exoplanets: not exactly habitable

GABE PEREZ-GIZ: Every few months, the press describe some recently discovered exoplanet as the closest thing yet to Earth's twin. But how much like Earth are these planets really, and do we even have the ability, yet, to answer that question?

Lots of headlines have touted the discovery of potentially habitable worlds, some of which might be "Earth's twin". Now, typically these stories are accompanied by pictures that look a lot like M-class planets from Star Trek, with a solid surface, liquid water, and a surface gravity that humans could at least function in. But unfortunately, those are all artist's renditions. These exoplanets don't actually look like that. Or, maybe they do. No one knows. In fact, we have no idea whether any of these planets has water or oxygen, let alone whether you could just walk out of a space ship and hang out there. It's not that the press or astronomers are lying. There's just a major disconnect between what many people imagine when they hear the phrase "habitable world" and what astronomers mean by that same phrase.

To astronomers, the phrase "habitable exoplanet" just means "exoplanet that lies in the habitable zone of its host star". And the habitable zone, in turn, is defined as the sweet spot of orbital distances from that star at which the energy from starlight would produce the right temperature on a planet's surface for water to remain liquid, provided the planet actually has a surface, and provided there's enough atmospheric pressure.

Now, the bad news is that being in a habitable zone isn't even remotely close to a guarantee of actual human habitability. For instance, Venus has a solid surface, and it has plenty of atmospheric pressure. Plus, if you expand its orbit just a teensy bit, it would be in the sun's habitable zone. But there's no liquid water there. So is the habitable zone, instead, a prerequisite for habitability?

Not exactly. A planet with a super-thick atmosphere, for example, could have surface water in a larger orbit than you'd ordinarily expect. Basically, the habitable zone is more of a guideline—a starting point to narrow down targets of interest. Those estimates you hear of an average of one habitable planet per star in the Milky Way are really just statements about this starting point.

But to make a more definitive assessment of habitability, actual habitability, for any of these exoplanets, you have to analyse their atmospheres. And here's the thing. We don't know anything about the atmospheres of any of the so-called Earth-like habitable worlds that get reported in the press. In fact, we can't be certain that they even have atmospheres at all. Take Kepler 186F as an example. In April 2014, this planet got a lot of press as the first confirmed Earth-size exoplanet in the habitable zone of its host star. These artist renditions make it look pretty sweet. But the only sense in which we've actually seen Kepler 186F is this graph. It shows the microscopic dip in starlight measured by the Kepler Telescope when the planet moved in front of its star. By knowing some properties of that star, then based on how much light gets blocked and for how long, astronomers were able to infer the radius of the planet and some features of its orbit, like its approximate distance from that star. Turns out Kepler 186F is about 10% larger in radius than Earth in an orbit around the size of Mercury around a fairly dim red dwarf star. So basically it's a planet like Krypton. Or maybe it's nothing like Krypton, because at 500 light years from Earth, Kepler 186F is just too far away to determine its mass or anything about its atmosphere with current or planned telescopes. All we know are its radius and the approximate size of it's orbit. That's it. Sure, you can put limits on its mass by considering extremes of what it could be made out of, and it seems likely from those bounds that it has a solid surface. But the atmosphere is, wait for it, totally up in the air. It could be Earth-like, or it could be wispy and barely there at all like on Mars, or a carbon dioxide super greenhouse like on Venus, or it could have no atmosphere at all. We just don't know.

Ok, what about habitable zone exoplanets that are closer to us than Kepler 186F is? Could we measure properties of their atmospheres and locate a truly habitable world like that? Well, here's where it gets interesting. You can measure the atmosphere of an exoplanet, but not if that exoplanet is an Earth-sized rocky body in a star's habitable zone. Let me explain the problem. To analyse the planet's atmosphere, you need to isolate the planet's light from that of its star, and see how bright that light is at different wavelengths. That graph of brightness versus wavelength is called an object's spectrum. Since different atoms and molecules emit or absorb particular wavelengths of light only, the spectrum tells you a lot about atmospheric composition. If you combine that information with the planet's mass, radius, and distance from its star, you can use models to get a rough picture of what the atmosphere is like. 

OK. So how do you isolate a planet's light? There are two basic methods. First, you can directly image the planet, improving the contrast by blocking out the star's light, kind of like putting your hand over your eyes on a sunny day to help see your surroundings. But this only works for planets that are in very large orbits outside the habitable zone, because closer in, the contrast is still too low. The planet just gets washed out by the star's light. Remember, planets don't glow very bright on their own. The second method takes advantage of what are called planetary transits. You take a spectrum of a star when the planet is in front of it. This will be the combined spectrum of the planet and star. And then you take another spectrum when the planet is behind the star. That gives you the spectrum of the star alone. Subtract the two, and you get the spectrum of just the planet. It's very clever. But unfortunately, this method only works for planets that are really close to their stars, because only planets that are close in will get hot enough to glow enough to improve the contrast ratio for that method.

So you see the problem. To be in the habitable zone, a planet has to be small enough, to be rocky like Earth instead of gaseous like Jupiter, and it can't be either too close or too far from its star, or it won't have liquid water. But unless it is really close or really far, its atmosphere cannot be measured easily, and you can't know whether it really might be habitable. It's a catch-22, but only because of the limitations of current instruments. There was a proposal for a terrestrial planet finder, or TPF, a space telescope that could have analysed the atmospheres of Earth-sized planets in Earth-like orbits around sun-like stars. But funding for the TPF was cut a few years ago. So for the forseeable future, pinning down the actual habitability of true Earth analogs isn't happening.

So does that mean identifying the only potentially habitable worlds is a waste of time? No, on the contrary. It's critical. We have to narrow the field, and we need to improve our census of what kinds of exoplanets are out there. It's important science. I just think it's also important to understand that despite what you read, we don't actually have the means to identify Earth 2.0 yet. But we could. The science is pretty much all there. And ironically, if headlines didn't make people think we'd already identified Earth 2.0, there might be more public outcry to make these missions happen. I don't see #iwantearth2.0 on Twitter yet. So what do you guys think? I know the exoplanet people need to pump up public interest in their work, and the press always needs good stories, but could the current way of reporting about habitable exoplanets be shooting the actual identification of Earth's twin in the foot?

This transcripts ends at 6:45. The remainder of the video discusses responses to comments from previous videos by PBS Space Time.