The K2-138 system: the first exoplanet system discovered by citizen scientists

by Justin Davenport

NASA has announced the discovery of a planetary system around star K2-138.  The discovery was the first of its kind, being made by citizen scientists using data from the K2 mission as part of the Zooniverse Exoplanet Explorers effort.  This detection is the first multi-planet system discovered exclusively by crowdsourcing – the practice of obtaining services and/or content by soliciting contributions from a large group of people, especially from the public, online community.

Exoplanet discovery background:

The age of exoplanet discovery, as part of humanity’s ongoing effort to map the universe and understand its workings, began in 1989 with the detection of what is now known to be the actual first exoplanet discovery, Gamma Cephei Ab.

Due to the burgeoning nature of exoplanet research in 1989, doubt was originally cast on this exoplanet’s existence (and its “discovery” was even retracted for a time), which was discovered via radial velocity – measuring observed Doppler shifts in the spectrum of an exoplanet’s parent star.  However, subsequent study of Gamma Cephei A finally led to the exoplanet’s official confirmation in 2002.

Artist’s impression of Gamma Cephei Ab, the first exoplanet to be discovered. Credit: NASA

From the time of the first exoplanet discovery until 2010, a relative handful of exoplanets were found – including the least massive planet ever discovered (just twice the mass of Earth’s moon)  – in various star systems, with a majority of those planets being gas giants.  But the discoveries were relatively few and far between and were largely complicated by Earth’s atmosphere and rotation that prevented sustained, continuous observations of stellar systems.

The answer to this problem was the space-based Kepler observatory.  Launched in 2009, Kepler’s mission was, in simple terms, to find Earth-sized planets orbiting other stars and determine how common Earth-sized planets are.  In so doing, it was known that Kepler would find numerous other planets much larger than Earth and would greatly increase our knowledge of extrasolar system composition and exoplanet frequency.

Kepler was a quantum leap forward in exoplanet discovery, detecting thousands of “Kepler objects of interest” (planet candidates) by staring exclusively for four years at a region of the sky in the constellations Cygnus and Lyra and transmitting the light curve data of every star in its field of view to professional scientists on Earth.

Kepler used four reaction wheels to keep itself pointed to the Lyra-Cygnus portion of the sky in an extremely steady state that prevented wobbles in the telescope’s orientation that could create distortion in the sensitive light curve data it was collecting.

To maintain this steady state, Kepler needed three of its four wheels to remain operational.  In 2012, one of the wheels failed, and a second wheel failed in May 2013 .

This second failure caused the end of the primary mission after exactly four years of observation (the original mission plan called for a 3.5 year observational campaign) because the craft could no longer stay steadily pointed toward the target without continuously using its limited supply of thruster fuel – which would be rapidly depleted if used in such a way.

Artist’s impression of the Kepler Space Telescope during its K2 extended mission. Credit: NASA/JPL-CalTech

Kepler mission scientists and controllers then devised a new method of keeping Kepler stably pointed at a specific location by using pressure from sunlight as the “third reaction wheel”, but the craft would need to point at a different location every 83 days to keep the Sun out of the telescope optics.

This new mission was dubbed K2, and started in 2014 and has continued since.

Crowdsourcing Kepler data – discovery of K2-138 system:

Overall, the Kepler mission has been a resounding success, far exceeding expectation.  And in this, NASA and professional scientists around the world found themselves with an unanticipated but good problem: Kepler was returning so much data that they couldn’t keep up.

Moreover, there were issues analyzing the data with computer algorithms that were found to be imperfect – flagging “false positives” while missing some signals that were planets.  

However, humans have a strong ability to recognize patterns; thus, a citizen science effort called Zooniverse Exoplanet Explorers was started to take advantage of this and involve the public in exoplanet discovery.  Zooniverse also engages citizens scientist in other astronomical and scientific endeavors.

While officially found last year after K2 observation campaign 12 data was uploaded by NASA to Zooniverse’s Exoplanet Explorers site, the discovery of the multi-planet system by Zooniverse crowdsourcing on was not officially confirmed until last week when NASA announced that a study on the data identified by citizen scientist had been analyzed and accepted for publication in The Astronomical Journal.

The K2 campaign 12 data was collected by Kepler from 15 December 2016 to 4 March 2017 and uploaded to Zooniverse in April 2017.  Within days, crowdsource efforts identified star K2-138 as a promising candidate for a planetary system with four transit signatures found in the initial crowdsource data review.

Follow-up observations from Earth-based telescopes were arranged for the system and not only confirmed the four exoplanets but also found a fifth.

A potential sixth planet in the K2-138 system was also flagged for further study.  Kepler data was found to contain two potential transits for this sixth planet, but three transits are required for confirmation.

In all, K2-138 is an orange K1V star approximately 85% of the mass and radius of the Sun.  It is slightly cooler and dimmer than the Sun (which is a G-class yellow star), but it is warmer and brighter than M-class red dwarfs.  It is approximately 600 light years away from Earth in the direction of the constellation Aquarius.

K2-138 system composition:

The K2-138 system’s five confirmed exoplanets are all larger than Earth but smaller than Neptune (sub-Neptunes), and the outermost confirmed planet orbits the star in just 13 days – meaning the system is very tightly packed, with all five planets orbiting closer to their star than Mercury orbits the Sun.

Exoplanet size comparison chart of the K2-138 system. Credit: NASA/JPL-CalTech

These worlds would be much too hot for life as we know it, with temperatures on these worlds ranging from 800℉ to 1,800℉.

The innermost planet, K2-138b, is approximately 1.57 Earth radii and orbits its parent star in just 2.35 days.  This exoplanet is thought to be a very hot “super-Earth” that could have a solid surface.  

The next world out, K2-138c, is 2.52 Earth radii and orbits in 3.56 days.  It is a “sub-Neptune” that is likely a gas planet.

Farther out, K2-138d is 2.66 Earth radii and orbits the star in 5.40 days; the next planet, K2-138e, is the largest planet discovered so far in the system and is 3.29 Earth radii and orbits the star in 8.26 days.

The outermost confirmed planet, K2-138f, is 2.81 Earth radii and orbits the star in 12.76 days.  The remaining unconfirmed sixth exoplanet candidate is thought to be 2.8 Earth radii and orbits the star in 42 days.  If confirmed, this sixth exoplanet would be designated K2-138g – assuming another planet in the system is not found and confirmed before it.

Artist’s conception of the K2-138 system, from an orbital vantage point around K2-138f (the farthest confirmed planet in that system). Credit: NASA/JPL-CalTech

At present, observations of the K2-138 system have not provided scientists enough data to estimate the masses of any of the K2-138 exoplanets, but it is believed the planets are between 4-7 Earth masses per mass-radius models.  

Follow-up observations with the NASA Spitzer or ESA CHEOPS (CHaracterising ExOPlanets Satellite) missions could yield exact masses for these planets using Transit Timing Variations, small variations in the timing of transits caused by the gravitational influence of other worlds, particularly in tightly packed systems like K2-138.

However, the available data does reveal that the K2-138 system has a near 3:2 resonance, meaning each planet has an approximately 50% greater orbital period than the next planet closer to the star.  This resonance will make it possible to measure the masses of the planets with Transit Timing Variations data from Spitzer and CHEOPS.

The CHEOPS (CHaracterising ExOPlanets Satellite) for the European Space Agency. CHEOPS is schedule to launch No Earlier Than late 2018. Credit: ESA

Once initial mass estimates for the exoplanets are determined, scientists can then combine those measurements with radial velocity measurements of the “wobble” imparted onto the K2-138 star caused by its planets’ orbits to further refine the planets’ mass estimates.

While the K2-138 system is the first extrasolar system to be discovered by crowdsourcing, it will likely not be the last.

NASA confirmed that the latest batch of K2 data has now been uploaded to the Zooniverse Exoplanet Explorer site – enabling more potential public discoveries of exoplanets and extrasolar system families.

As of 1 January 2018, there are 3,726 confirmed planets in 2,792 systems, with 622 systems having more than one planet. More so, according to NASA’s website, there are 4,496 exoplanet candidates currently awaiting confirmation.

Related Articles