Kepler finds more 'Earth-like planets', but are they really like Earth?

The number of confirmed planets orbiting other stars has just jumped by 1,284 with NASA’s new analysis of data from the Kepler space telescope.

That takes the total number of known exoplanets to 3,264, with more than two-thirds (2,325) having been found by this one incredible space observatory alone.

But it is only a tiny subset of those new planets that have caught many people’s imagination. As NASA’s announcement says:

[…] nearly 550 could be rocky planets like Earth, based on their size. Nine of these orbit in their sun’s habitable zone, which is the distance from a star where orbiting planets can have surface temperatures that allow liquid water to pool. With the addition of these nine, 21 exoplanets now are known to be members of this exclusive group.

It’s a great hook – “NASA finds Earth-like planets around distant stars” – but is all the hype justified? And what does this mean for the future of our search for planets like the Earth, and for life elsewhere?

Kepler: the first great exoplanet census

The reason that Kepler has been so astonishingly successful at finding planets comes down to its design. It was built specifically to conduct the first great census of planets around other stars.

To do this, Kepler has played a numbers game. In the first phase of the mission, the spacecraft pointed at a single region of the night sky continuously for just over four years.

With a field of view of 115 square degrees, this allowed it to continuously monitor the brightness of more than 100,000 stars, looking for the telltale “winks” of a planetary transit in front of the star.

Only a small fraction of all the planets out there will have orbits lined up in a way that allows them to transit their stars from our point of view. But given that we now know that most stars host planets (another of Kepler’s exciting revelations), a small fraction of 100,000 stars still gives many stars, and many planets to play with.

Finding Earth-like planets

Kepler’s data tells us two things about a given planet. The first is how long it takes to orbit its host, which in turn tells us how distant the planet is from that star.

The other thing we learn about the planet is its size relative to its host. This comes directly from the fraction of the star’s light that is blocked during a transit; a bigger planet blocks more light than a smaller one.

When we combine this with what we know of the star, it allows us to estimate the actual size of the planet. This is where our Earth-sized planets come in to the picture.

Thanks to its exquisite sensitivity, and the fact its view isn’t obscured by the Earth’s atmosphere, Kepler is capable finding planets the size of Earth, and even smaller!

The smallest planet Kepler has found to date is smaller than Mercury, a truly amazing achievement.

The leap of faith

How, then, do we go from saying a planet is Earth-sized, to considering it Earth-like?

Well, first, we have to work out whether Earth-sized means Earth-mass. Is the planet we’ve found rocky, metallic or gaseous? To do this, we need to find out the mass of the planet, and thereby its density.

This requires challenging follow-up work that is all but impossible for the smallest Kepler planets, particularly given how faint their host stars are.

Hence the guarded wording in the NASA announcement that “nearly 550 could be rocky planets like Earth, based on their size”.

Then we have questions about the climate and nature of the planet, and what it means to be truly Earth-like? This is a really complicated question to answer, going far beyond simply knowing a planet’s size and distance from its host.

A first step on that journey comes from the consideration of the habitable zone, the region around a given star where a planet like the Earth could feasibly have liquid water on its surface.

The exact reach of that region depends on how strict, or how relaxed your assumptions are. But it does provide a nice first cut to whittle out those planets that would simply be much too hot or much too cold (for a given set of assumptions, of course!).

And this is where our knowledge of the orbital distance of the planet in question comes in to play, coupled with our understanding of its host star. Around cool, dim hosts, the habitable zone is expected to be fairly close in. Around those hosts that burn hot and bright, it would be more distant.

Combine our knowledge of the planet’s orbit, and of the star’s brightness, then we can estimate whether the planet could host liquid water, so long as it is something like the Earth, of course!

Whittling down the sample

Taking all this into consideration, the new Kepler announcement is still a fantastic achievement:

• 1,284 new planets, of all sizes;
• of these, almost 550 are small enough to be thought likely to be rocky planets, such as Mercury, Venus, Earth and Mars; and
• of those 550, nine orbit their stars within the habitable zone, at just the right distance that they could, possibly, host liquid water on their surfaces.

Kepler (and other, ground-based transit searches) are strongly biased to finding large planets, very close to their host stars.

The closer in the planet, the shorter its orbital period, so the more transits you see, and the easier it is to be sure you’re really seeing a planet.

So to find even this small number of potentially Earth-like planets suggests that, in the wider scheme of things, such planets are likely common. But we can’t say for certain that any of those worlds are truly Earth-like.

To be sure that the planets we find are truly Earth-like, and to find planets that are good targets for the search for life elsewhere, we have to wait for the next generation of planet search programs.

The future search

Kepler has done an astonishing job of revealing what is out there – the first census of the exoplanet sky – but it is Kepler’s successors that will drive the search for truly Earth-like worlds.

Kepler’s direct successor, the Transiting Exoplanet Survey Satellite (TESS), will launch next year. Where Kepler observed more than 100,000 faint stars in one patch of the sky, TESS will instead scan the whole sky, looking at brighter stars that are much easier to study with follow-up work.

Around the world, astronomers will be racing to follow up TESS’s discoveries, in a truly global endeavour. As a result, in just the next few years, we may well find the first planets that can truly be said to be Earth-like.