Exoplanet #431 in the bag

The HATNet team at Harvard have discovered a new exoplanet. It is a hot Jupiter type planet, orbiting its star a good 200 parsecs away, once every four and a half days. It has a mass of around two and a quarter Jupiters and a radius three twentieths greater than that of Jupiter. The star is pretty bright at just better than tenth magnitude and so can be seen with a moderate amateur telescope. For those who’d like a photograph of the area, the coordinates are right ascension: 17h 20m 27.96s, declination: 38 degrees 14′ 31.8″ (J2000).

There now follows a summary of the paper behind the discovery, which may include technical notation.

The star known in this survey as HAT-P-14 (the star with the fourteenth planet discovered by HATNet) is more normally known by the romantic name GSC 3086-00152. It has a visual apparent magnitude of 9.98 and is a mid-F dwarf 1.386 times as massive as the Sun with a radius 1.486 as large. It is slightly metal rich and has an effective temperature (ie a black body spectrum associated with a temperature of) 6600K.

The planet discovered orbits the star with a period of 4.63 days, each transit lasting 0.0912 days. Analysis reveals that the transit is close to grazing (the planet passes along the edge of the star’s disc as seen from here rather than its middle) and the planet itself has a mass of 2.232 times that of Jupiter and a radius 1.150 as large, giving a density of around 1.82 grammes per cubic centimetre. The position and slight eccentricity (0.107) of this 1.3 Gigayear old system can be explained by the planet having fifty times the mass of the Earth’s worth of heavy elements within it, accounting for 7% of its total mass. The system is in an excellent configuration for careful studies of its orbit to reveal any additional planets or even large exomoons perturbing the orbit.

HATNet is the Hungarian made Automated Telescope Network. It surveys stars between 8th and 12.5th magnitude using four widefield instruments at the Fred Lawrence Whipple Observatory (FLWO) in Arizona and two on the roof of the hangar servicing the Smithsonian Astrophysical Observatory‘s Submilimetre Array in Hawaii. It looks for dips in the light of these stars associated with a planet passing between the star and the instrument.

As the name suggests, HAT-P-14 follows 13 other Transiting Extrasolar Planets (TEPs) discovered and published by HATNet. It was discovered by the HAT-10 instrument during observing runs in May and July 2005. This instrument was in its commisioning phase and able to detect changes of 3.5 thousandths of a magnitude (mmag) in brightness. This particular star exhibited changes of 4.5 mmag, a prime candidate for the instrument. Examination of the recorded data showed a dip in brightness of the star at regular periods. Although transiting planets do this, this can also be caused by other effects. Having determined there is a dip through photometry, the group then began to look at the spectrum of the system to rule out some of these other potential causes.

Using the 1.5m telescope at the FLWO with a spectrograph that had a power of wavelength/(change in wavelength) = 35,000, the group assigned a rotational velocity of 11.7 km/s and a radial velocity of -20.8 km/s to the star. There was only the one stellar spectrum visible, which suggested this wasn’t a binary star system masquerading as one star with a planet.

Now armed with evidence that they had potential extrasolar planet, the team were able to get time on the Keck 1 telescope. Using this (with a spectrograph that has a resolving power of 55,000), they were able to both confirm the assessment given by their previous look at the spectrum and assign confidence to certain characteristics of the star itself. Looking at lines produced by Hydrogen, singly ionised Calcium and Potassium, the team could rule out the possibility of substantial activity in the chromosphere of the star that may have interfered with photometry. They were also able use this and improved photometry from KeplerCam on the 1.2m FLWO telescope to rule out another possible transit cause – blends. Blends are unresolved binaries in the background that act as if they’re part of the main star. In order to look for them, you look to see if the dip seems to be more to one side than in the middle. There was a little irregularity in the photometry, so the spectrum was investigated for signs of scattered Moonlight when the photometry was taken (a problem identified in the discovery of HAT-P-12). There was a little – enough to account entirely for the irregularity seen.

From the spectrum, the team were able to derive the effective (black body) temperature of the star and (via the Doppler shift) all the velocities associated with the system. From the photometry (brightness measurement), they determined photometry at a number of different wavelengths. They used these figures and models of stellar evolution to determine the star mass, radius, absolute luminosity (3.66 times that of the Sun), the age of the system, its distance and absolute visual magnitude.

With these parameters in, work could be done reproducing the observations of the planet through modelling. They used an eccentric Keplerian model for the planet and modelled the transit effect, accounting for limb darkenning of the star and other effects to get the eccentricity of the orbit. They also used methods for eliminating systemic error from the measurements, looking at the photometry against hour angle (for changes with time), hour angle squared (for changes with elevation) and stellar parameters such as the full width half maximum (size), elongation (shape), brightness (when transits weren’t occurring) and position of the star image on the CCD. Each of the five nights observations were made were considered as separate datasets, giving thirty parameters for analysis. Twenty other bright stars captured in the same field of view were also included, to look for global effects. By the end of all that, they had their planet’s characteristics and enough to publish.

HAT-P-14 is the 12th brightest of sixty-six stars known to have TEPs. It has been discovered that TEPs drop off dramatically in numbers below twice the magnitude of Jupiter. There’s no obvious observational reason for this sudden drop, suggesting there’s a physical mechanism behind it. All TEPs with an eccentricity greater than 0.2 are above 3 times the mass of Jupiter (but we’ve only found a couple, so this could mean anything). Observations of systems with planetary orbits misaligned with the rotational axis of the star also only include massive planets, but these too are rare in current observations.

HAT-P-14b is similar in mass to Kepler-5b, but a fifth smaller in radius. This puts it in the pM category of exoplanets, meaning it is likely to have temperature inversions in its atmosphere and to have large variations in its day and nightside temperatures. Just modelling the black body temperature associated with reflected light gives a temperature of 1570 degrees Kelvin.

One confusing thing about the planet is that it has a slightly but still significantly eccentric orbit while being close in to its host star. Using a tidal dissipation factor of 10^5, the time taken for a planet like HAT-P-14b to be brought into a near circular orbit by tidal effects alone is 0.33 Gigayears. This system is 1.3 Gigayears old. Either, something is perturbing the orbit of this planet (like another planet) or the assumed tidal constant is wrong. As the tidal constant is poorly known and 10^6 isn’t outside the realms of plausibility (which would make the time taken for tidal effects to bring the planet into line ten times larger), this is the main suspect.

The Hill radius of the planet, associated with moons, is 8.1 planetary radii (9.1 Jovian radii). The maximum distance for an exomoon to be in a stable orbit around this planet at this orbital eccentricity is 0.436 for a prograde moon and 0.824 for a retrograde one. If we assume an exomoon at maximum distance and a tidal dissipation constant of 10^5, then this means maximum moon masses of 0.002 times the Earth for a prograde moon and 0.12 Earth masses for retrograde. If we use a constant of 10^6, these figures go up by a factor of ten, suggesting a retrograde moon of 1.2 Earth masses is possible in a retrograde orbit about this star (life bearing planetary masses tend to be quoted in the 0.3-10 Earth mass range).

By making careful measurements of how the transit time, transit duration and transit impact factor alter over time, it is possible to determine whether or not there are other planets or even moons present in the system. Assuming the constant is 10^6 and there is a 1.2 Earth mass moon in this system in the optimal configuration, observables would be in the range of:

  • TTV = 4.5s
  • TDV-V = 1.05
  • TDV-TIP = 16-45s

Compare this to the similar planet TrES-26, where the TDV-TIP factor for such a moon would be 0.7s, that’s an indication of how much easier it is to spot sub-Earth or Earth sized planets or exomoons interacting with the orbit of this planet.

One to keep an automated eye on.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s