web analytics

Dr David Kipping




There is currently a widespread and perfectly understandable pre-occupation with discovering exoplanets in the habitable-zones of their host stars. These worlds are just the right distance from their parent star that liquid water could be present on the surface, given the right atmospheric conditions. But venture a little further out and you hit the so-called “frost-line”, an important boundary in understanding how planets form. Finding worlds this far from their host star is very difficult with the transit technique and yet such planets should be common. In our recent paper, we discovered the first example of a transiting planet near the frost-line Kepler-421b.

A year on Kepler-421b lasts for 704 days, making it the longest orbital period transiting planet yet found. We discovered the planet by detecting the tell-tale decrease in brightness of the parent star as Kepler-421b passed in front - a “transit” event (see figure at the bottom). It did this just twice in the 4.35 years for which NASA’s Kepler Mission patiently stared at it, the telescope which has found hundreds of transiting planets but never before at such a long period. The amount of light blocked by the planet betrays her size and we find that Kepler-421b is nearly the same size as Uranus, which is about four times larger than the Earth. Using a special technique I developed called Asterodensity Profiling, we can also exploit the transit light curve shape to measure the orbital eccentricity of Kepler-421b, which works out to be ~0.04, almost exactly the same as Uranus as well. It is probably reasonable to assume Kepler-421b looks similar to Uranus or Neptune then (see artist’s impression, credit David Aguilar) and adopting a Uranus-like albedo we estimate the temperature of this world would be a chilly -130 degrees Fahrenheit.

Kepler-421b’s very long year makes it more than merely a new record-holder, it makes it the first transiting planet discovered near the so-called “frost-line”. In our solar system, the frost line is essentially the dividing point between the rocky interior planets (Mercury, Venus, Earth & Mars) and the outer gaseous giants (Jupiter, Saturn, Uranus & Neptune). When the solar system was first forming, it was at this special distance that the temperature was cold enough for ice grains to form. Any closer to the Sun and these grains would be boiled off. These ice grains started to stick together and form planetary embryos which then went on to form the gas giant planets. It is for this very reason that the gas giant are rich in ice and water but the rocky planets started out their existence as very dry worlds (water got delivered to the rocky planets later via comets).

Depiction of the frost line. Image credit Pearson Education and Addison Wesley

The frost-line is not a fixed point in space but actually moves inwards over time as the protoplanetary disc of a young star evolves. From this disc, planets form and their properties will be determined (in part) by whether they were inside or outside the frost-line during their birth. In the case of Kepler-421b, we estimate that the frost-line would be the same distance as the planet’s present location when the system was about 3 million years old. It is thought that planets form over the first 3 to 10 million years, which means that Kepler-421b could have formed exactly where we see it today. That’s quite remarkable because all of the other known gas giant transiting planets found to date must have formed further out and migrated inwards later on. More over, the fact that Kepler-421b is a Uranus-sized planet and not a Jupiter-sized planet may because it formed in the final stages of this planet formation era and thus there was less material left over to build planets by this point. Of course, we don’t really know how this planet formed because no-one was there to watch it happen, but these clues may help us decipher the history of this world.

In general, planets further away from their host star have a lower probability that their orbital inclination is small enough that they appear to pass in front of their parent star. A planet like Kepler-421b has a minuscule 0.3% chance of transiting, and yet here she is. To make things worse, with only two transits observed by Kepler, the signal is much harder to find than a planet on a shorter orbital period which would hundreds of transits over the same time range. Put together, these effects generally reduce the yield of long-period planets in transit surveys by a scaling of P^(-5/3). This means that a hot-Neptune (~3 day orbital period) is ~9000 times easier to find than a frost-line-Neptune, like Kepler-421b. Although it’s a bit premature to go through an occurrence rate calculation, since we have just a single planet in this regime, it seems likely that these cold-Neptunes are everywhere, but just very hard to find.

Transit light curve of Kepler-421b. Blue and red points denote the two different transit epochs observed, offset in time by 704 days.