We report the discovery of WASP-5b, a Jupiter-mass planet orbiting a 12th-mag G-type star in the Southern hemisphere. The 1.6-d orbital period places WASP-5b in the class of very hot Jupiters and leads to a predicted equilibrium temperature of 1750 K. WASP-5b is the densest of any known Jovian-mass planet, being a factor of 7 denser than TrES-4, which is subject to similar stellar insolation, and a factor of 3 denser than WASP-4b, which has a similar orbital period. We present transit photometry and radial velocity measurements of WASP-5 (= USNO-B1 0487-0799749), from which we derive the mass, radius and density of the planet: MP= 1.58+0.13−0.08MJ, RP= 1.090+0.094−0.058RJ and ρP= 1.22+0.19−0.24 ρJ. The orbital period is P= 1.6284296+0.0000048−0.0000037 d and the mid-transit epoch is TC (HJD) = 245 4375.62466+0.00026−0.00025.
Several wide-angle, ground-based surveys for transiting extrasolar planets have now been successful, including HAT (Bakos et al. 2002), TrES (O'Donovan et al. 2006), WASP (Pollacco et al. 2006) and XO (McCullough et al. 2005). In the Southern hemisphere relatively faint (V= 16-17) transiting systems have been found by OGLE (Udalski et al. 1992; Pont et al. 2007) and by Weldrake et al. (2008). The WASP-South survey is the first to find much brighter transiting systems in the south (Wilson et al. 2008). We report here the second planet found by the WASP-South and CORALIE collaboration, a dense, very hot Jupiter transiting a V= 12.3 star with an orbital period of 1.6 d.
Transit searches are most sensitive to large, Jupiter-sized planets in very short orbits. WASP-5b, like the recently announced WASP-3b and WASP-4b, has an ultrashort period below 2 d (Pollacco et al. 2008; Wilson et al. 2008). We discuss the characteristics of these ultraclose, highly irradiated planets.
The WASP-South observatory is sited at the Sutherland station of the South African Astronomical Observatory. It consists of an array of eight cameras, each with a Canon 200-mm f/1.8 lens backed by an e2v2k × 2k CCD, covering 7.8°× 7.8°. WASP-South started operating in 2006 May with a strategy of tiling up to eight pointings with a cadence of 5-10 min, taking two 30-s exposures at each pointing. The WASP hardware, observing strategy, data reduction pipeline and archive are described in detail by Pollacco et al. (2006). The data were detrended and transit searched as described in Collier Cameron et al. (2006), and candidates were selected as described in Collier Cameron et al. (2007).
WASP-5 (=1SWASP J235723.74-411637.5 = USNO-B1 0487-0799749 = 2MASS 23572375-4116377 = NOMAD1 0487-0881175) is listed in the NOMAD (Naval Observatory Merged Astrometric Dataset) catalogue as a V= 12.3 star, and was observed by WASP-South from 2006 May 13 to 2006 November 14. Transits of WASP-5b were detected by two WASP-South cameras, where they overlap at the field edges. A total of 4642 data points were obtained with camera 7 and 4108 data points with camera 8 (Fig. 1).
During follow-up, a transit was observed in the R band using EulerCam on the 1.2-m telescope at La Silla on 2007 October 10 (Fig. 2). The telescope was heavily defocused to allow long exposure times (2 min) resulting in an rms scatter of 1-mmag. A further transit was observed in the Sloan Digital Sky Survey (SDSS) i' band on 2007 October 13 using HawkCam2 on the 2.0-m Faulkes Telescope South (FTS) at Siding Spring Observatory (Fig. 2).
Radial velocity data were obtained with the CORALIE spectrograph on the Euler 1.2-m telescope. CORALIE had recently been upgraded, as described in Wilson et al. (2008). 11 radial velocity measurements were taken over the course of one month (Table 1; Fig. 3), establishing WASP-5b as a planetary-mass companion.
We used a line-bisector analysis to look for asymmetries in the spectral line profiles, as could be caused by contamination from an unresolved eclipsing binary (Queloz et al. 2001; Mandushev et al. 2005). Such a binary would produce bisector spans that vary in phase with the photometric period with an amplitude comparable to the radial velocity amplitude. This is not seen in our data (Fig. 3, bottom panel), supporting the conclusion that the radial velocity variations are due to a planet.
The CORALIE spectra, when co-added, give a signal-to-noise ratio (S/N) of ~40, which is suitable for a preliminary photospheric analysis of WASP-5. The analysis was performed using the uclsyn spectral synthesis package and atlas9 models without convective overshooting (Castelli, Gratton & Kurucz 1997). The Ha, Na i D and Mg i b lines were used as diagnostics of both Teff and log g. The metallicity was estimated using the photospheric lines in the 6000-6200 Å region. The parameters obtained from this analysis are listed in Table 2.
We also used the Galaxy Evolution Explorer (GALEX) near-ultraviolet (NUV) flux and magnitudes from the NOMAD, DENIS (Deep Near Infrared Survey) and 2MASS (Two Micron All Sky Survey) catalogues to estimate the effective temperature using the infrared flux method (IRFM; Blackwell & Shallis 1977). This gives Teff= 5610 ± 250 K, which is in close agreement with the spectroscopic analysis. These results imply a spectral type in the range G2V-G6V. The Li i 6708 Å line is not present in the co-added spectrum, giving an upper limit on the lithium abundance of . This implies an age of ≳2 Gyr for a star of this effective temperature (Sestito & Randich 2005). The value of v sin i= 3.4 km s-1 obtained from the CORALIE spectra indicates an age of 1.7-4.4 Gyr (Bouvier 1997). Comparison of the temperature and log g* with the stellar evolution models of Girardi et al. (2000) gives maximum likelihood values of M*= 0.99 ± 0.08 M⊙ and R*= 0.97 ± 0.06 R⊙. The distance of WASP-5 (d= 300 ± 50 pc) was calculated using the distance modulus, the NOMAD apparent visual magnitude (V= 12.3) and the absolute visual magnitude of a G4V star (V= 4.9; Gray 1992). We assumed E(B-V) = 0, which is reasonable at such a distance and in a direction out of the galactic plane.
Spectroscopic Orbital Solution and Planetary Parameters
The CORALIE radial velocity measurements were combined with the WASP-South and EulerCAM photometry in a simultaneous Markov chain Monte Carlo (MCMC) analysis to find the system parameters. This process is described in detail in Collier Cameron et al. (2007) and Pollacco et al. (2008). The FTS data were not used owing to their higher red noise. The best-fitting solution to the radial velocity data gave an eccentricity of e= 0.03 ± 0.03. Our adopted MCMC solution fixes the eccentricity at e= 0, which is expected for such a short-period planet.
In optimizing the MCMC solution we used constraints (Gaussian priors) on the stellar parameters, setting M*= 0.99 ± 0.08 M⊙ and log g*= 4.30 ± 0.20. In order to balance the weights of the photometry and radial velocities in the MCMC analysis, we added a systematic error of 7 m s-1 to the radial velocities (as might arise, for example, from stellar activity) to reconcile χ2 with the number of degrees of freedom; the error bars in Fig. 3 do not incorporate this error. The optimal MCMC parameters are shown in Table 3.
The discovery of WASP-3b (Pollacco et al. 2008), WASP-4b (Wilson et al. 2008) and WASP-5b, all transiting, Jupiter-mass planets with orbital periods of less than 2 d, increases the number of such systems known from four to seven.1 It is notable that they show a large disparity in density (Fig. 4). WASP-5b is the densest of these systems, whereas WASP-4b is the least dense, with a density a third that of WASP-5b. This illustrates that planets with a similar orbital period and radiation environment must still have different compositions or past histories.
For example, WASP-5b will receive the same insolation and thus has a similar estimated equilibrium temperature (1753 K) to TrES-4 (1720 K), which has a longer orbit about a more luminous star (Mandushev et al. 2007). Despite this, WASP-5b is denser by a factor of 7 (1.22 ρ compared with 0.167 ρJ for TrES-4). Burrows et al. (2007) postulate that the bloating of the underdense planets could be due to an enhanced atmospheric opacity, which could be absent in WASP-5b.
Although it is denser than other planets with a similar orbital period, WASP-5b is still within the expected theoretical range (Fortney, Marley & Barnes 2007); depending on the age of the system, a core mass of ~50-100 M⊙ is expected, with a younger planet requiring a more massive core. However, if the bloating of some planets were due to an extra energy source that is common to all hot Jupiters, then the core of a dense planet such as WASP-5b would have to be more massive to compensate (Fortney et al. 2006; Guillot et al. 2006). If the core mass is toward the higher end of the estimate and the metallicity is no more than solar then there could be a discrepancy with the relationship between core mass and stellar metallicity postulated by Burrows et al. (2007). High-resolution spectra with better S/N are required to reduce the uncertainties in the metallicity and age.
WASP-5b (log g*= 3.48) has the highest surface gravity of planets with orbital periods less than 5 d. Thus it could be a good system for studying evaporation as a function of surface gravity. In comparison HD 209458b is heavily bloated (log g*= 2.97) and appears to be undergoing evaporation (Vidal-Madjar et al. 2003, 2008; Ben-Jaffel 2007). Given its high equilibrium temperature, WASP-5b is also expected to have a prominent secondary transit, which should be detectable with Spitzer (e.g. Harrington et al. 2007). Further, with the moderately bright host star having a moderate rotation rate of v sin i= 3.4 km s-1, the Rossiter-McLaughlin effect (40 ± 8 m s-1) should be detectable (Gaudi & Winn 2007), allowing us to check the alignment between the planetary orbit and the stellar spin axis.
The WASP consortium comprises the Universities of Keele, Leicester, St. Andrews, the Queen's University Belfast, the Open University and the Isaac Newton Group. WASP-South is hosted by the South African Astronomical Observatory (SAAO) and we are grateful for their support and assistance. Funding for WASP comes from consortium universities and the UK's Science and Technology Facilities Council.