A new, full analysis of Kepler data finds at least 300 million Earth-size planets in the habitable zone around Sun-like stars in our galaxy.

Kepler concept
An illustration representing the legacy of NASA's Kepler space telescope. After nine years in deep space collecting data that revealed our night sky to be filled with billions of hidden planets – more planets even than stars – NASA’s Kepler space telescope ran out of fuel needed for further science operations in 2018.
NASA / Ames Research Center / W. Stenzel / D. Rutter

NASA’s Kepler telescope was retired a few years ago, but ongoing analyses of its data, both by professional astronomers and citizen scientists, are still producing new results.

The mission's primary goal was to try and estimate the prevalence of Earth-size planets on Earth-like orbits around Sun-like stars. But understanding the occurrence of such planets has proven difficult, even though Kepler has found more than 2,600 exoplanets (and counting). Now, an international collaboration led by Steve Bryson, a researcher at NASA Ames, has announced a refined estimate.

The team, including NASA scientists, SETI researchers, academics, former-Keplerites, and other planet hunters, performed a statistical analysis that combined Kepler’s planet catalog and stellar data from the European Space Agency’s Gaia observatory. They found that about half of the Sun-like stars in our galaxy could have a rocky planet in their habitable zones.



Making up for Missing Data

Team member Michelle Kunimoto (MIT) says this estimate is more reliable than previous ones: “Most previous estimates did not take into account that planets are more or less common around stars with different temperatures,” she says.

The Kepler mission used the transit method, detecting planets by the slight dimming of a host star when its planet passes in front of it. This technique revealed planets by the thousands, but it’s easier to find gas giants orbiting close to their host star rather than Earth-size planets on farther-out orbits. Kepler is sure to have missed lots of rocky worlds.

To account for this effect, Bryson’s team worked with Kepler’s planet-detection “pipeline,” which has become well-versed in both authenticating and debunking planet candidates, as well as finding ones that were previously missed. Experience has taught them which ‘dips’ in the light are caused by interference from objects like eclipsing binaries, stellar companions, or planets from other systems. Isolated blips also sometimes turn out to be planets with longer orbits.

“We figured a way to measure how many planets we were missing. It’s a huge number,” Bryson explains. “And then we had to figure how many were typically false positives. That's also a huge number.” Taking all this information, the team estimated the number of rocky planets with ½ to 1½ times Earth’s mass in the habitable zone around Sun-like stars. The team analyzed all their stellar and planetary data via two different techniques and compared the results. They matched.

 “It was a relief when the answer was reasonable, not a million planets, or zero,” Bryson says.

At Least 300 Million

Earth-like planet around Sun-like star
This illustration depicts one possible appearance of the planet Kepler-452b, the first near-Earth-size world to be found in the habitable zone of a star similar to our Sun.
NASA Ames / JPL-Caltech / T. Pyle

The study, soon to be published in The Astronomical Journal, predicts that there are at least 300 million habitable-zone rocky worlds in the Milky Way. A handful of these are within a few light-years of Earth. This result assumes that the section of the sky Kepler monitored for four years is representative of the whole galaxy.

It is important to note that this new estimate does not tell us where to find an exoEarth, or what fraction of those above-mentioned 300 million worlds actually has life. But the result does suggest, based on an analysis of a large amount of astronomical data and with a high degree of confidence, that potentially habitable Earth-size planets around Sun-like stars are common.

“We have absolutely no information about whether planets in the habitable zone are in fact inhabited,” Bryson said. “That will be the purpose of future missions. And this prediction helps the endeavor by making it hopeful; because it’s much more likely that a direct imaging telescope will actually be able to succeed in imaging a rocky planet in the habitable zone.”

Bryson sees this project as a kind of ode to the Kepler mission. Indeed, most of the original team members, including the mission’s principal investigator Bill Borucki (NASA), were involved in the study.

“We're not claiming this is the final answer,” Bryson says. “But it is the best answer the Kepler team knows how to do.”

Comments


Image of Rod

Rod

November 5, 2020 at 9:35 am

An interesting study here indicating perhaps 300 million exoplanets in the HZ around sun like stars in the Milky Way for mass range 0.5 to 1.5 earth masses. I checked this site, http://exoplanet.eu/catalog/ using MS SQL there are 17 exoplanets with masses 0.5 to 1.5 earth masses listed, most red dwarf hosts like TRAPPIST-1 system. I found 21 such exoplanets listed at this site, https://exoplanetarchive.ipac.caltech.edu/index.html, also mostly red dwarf star hosts. Orbital periods < 30 days.---Rod

You must be logged in to post a comment.

Image of Robert-Casey

Robert-Casey

November 5, 2020 at 4:33 pm

The Sun has two of these planets, Venus and Earth. Except maybe in its clouds, Venus is not habitable. And we know about Earth.

You must be logged in to post a comment.

Image of Rod

Rod

November 5, 2020 at 10:10 pm

Robert-Casey, very good observation. I went back and reran my MS SQL query for stars with masses 0.9 to 1.1 solar masses. The .eu catalog shows 1385 exoplanets. Descriptive stats shows some interesting results for planet masses. The average size is 2.38 Jupiter masses, min 1.3 earth masses and max 62 Jupiters. Most are large exoplanets reported as confirmed around stars 0.9 to 1.1 solar masses (could be detection methods favor this). The exoplanetarchive site shows 1290 exoplanets reporting for the same host star mass range.

You must be logged in to post a comment.

Image of Harry Powell

Harry Powell

November 9, 2020 at 3:42 pm

Has anyone done an estimate the number planets missed because the Earth is not close to the plane of THEIR ecliptic? i.e. they are not "conjunctions"? I am thinking Venus only crosses the Sun twice every 130-something years, so Earth could only be found by a race living on a planet located close to the plane of our ecliptic! I am a bit disappointed that I have never read any mention of this factor which must be at least one order of magnitude!

You must be logged in to post a comment.

Image of DavidFoss

DavidFoss

November 6, 2020 at 9:57 pm

I did a rough calculation based on the size of the universe and came up with star with
"Earth like" planets being about 16 light years apart on average. Cube root of (Pi x 200 KLY x 2 KLY / 300K) Does anyone agree with that or have a different estimate?

You must be logged in to post a comment.

Image of Fred Ringwald

Fred Ringwald

November 8, 2020 at 5:08 am

Please explain how you got the figure of 300 million Earth-size planets in the habitable zone of Sun-like stars. On what page of the preprint does it occur?

With 400-600 billion stars in the Milky Way Galaxy, about half of which are single, about 20% of which are main-sequence stars with temperatures between 4800 and 6300 K (what this preprint defines as Sun-like), about half of which having Earth-size planets (with 0.5-1.5 Earth masses) in their habitable zone (the essential result of this preprint), shouldn’t that figure of 300 million be 20-30 billion (since 400-600 billion * 0.5 * 0.2 * 0.5 = 20-30 billion)?

You must be logged in to post a comment.

Image of Monica Young

Monica Young

November 9, 2020 at 10:19 am

Hi Fred, the paper quotes "eta_Earth," which is the fraction of Sun-like stars with Earth-size planets in the so-called habitable zone. Note that Ms. Rimmer quotes this as "at least 300 million" — that is, the 300 million is the most conservative estimate based on the numbers in the paper. It could be many, many times more than that!

You must be logged in to post a comment.

Image of Fred Ringwald

Fred Ringwald

November 9, 2020 at 11:39 pm

Yes, 300 million is less than 300 billion: it's 1000 times less. Where does that extra factor of 1000 come from? Please ask Mr. Rimmer how he got the figure of 300 million Earth-size planets in the habitable zone of Sun-like stars. Or was it simply a matter of writing "million" when it should have been "billion"?

You must be logged in to post a comment.

Image of Monica Young

Monica Young

November 10, 2020 at 11:40 am

The number 300 million is correct for the more conservative estimates the authors published for eta_Earth. (They published a range, it may be more than 300 million, that's a lower limit.)

You must be logged in to post a comment.

Image of Rod

Rod

November 14, 2020 at 10:23 am

I did another MS ACCESS descriptive stat report on these two exoplanet sites, http://exoplanet.eu/catalog/, and https://exoplanetarchive.ipac.caltech.edu/index.html

The NASA archive site shows 4301 confirmed exoplanets. Average radius is 0.379 Jupiter. Min is 0.03 and max 6.9 Jupiters. The eu site, 4374 confirmed exoplanets, average radius 4.79 earth radii, min essentially zero (my format in query), max 73 earth radii. The new report with 300 million or more earth like planets is perhaps zeal for SETI 🙂

You must be logged in to post a comment.

You must be logged in to post a comment.