Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 356, 590-598 (2000)

Previous Section Next Section Title Page Table of Contents

3. A planetary companion around HD 75289. Down towards the mass of Saturn

3.1. Stellar characteristics of HD 75289

HD 75289 (HIP 43177) is a bright ([FORMULA]) G0 dwarf in the southern Vela (Sail) constellation. The HIPPARCOS catalogue lists a colour index [FORMULA] and a precise astrometric parallax [FORMULA] mas (ESA 1997) which puts the star at a distance of 29 pc from the Sun. Thus the absolute magnitude [FORMULA], typical for a G0 dwarf. This invalidates the supergiant classification usually given, for example in the Bright Star Catalogue (Hoffleit & Jaschek 1982).

This discrepancy was already pointed out in Gratton et al. (1989) who defined the star to be a "nearby metal-rich small-mass dwarf near the turn off". In their high-resolution spectroscopic analysis they derived a gravity [FORMULA] from the ionization equilibrium and an effective temperature [FORMULA] K from the excitation equilibrium of the FeI absorption lines. The latter is in good agreement with [FORMULA] K derived from the Geneva photometry (Grenon 1981, private communication). From their individual line abundance measurements, Gratton et al. also estimate a metallicity index [A/H] = 0.2 in agreement with the [M/H] = 0.28 value from the Geneva photometry and with the [Fe/H] = 0.29 value derived from the calibration of the surface of the CORAVEL cross-correlation dip (Pont 1997). This metallicity is higher than the average value (slightly negative) of the metallicity of the stars in the CORALIE sample, as is the case for most of the other stars with giant planets.

Using a bolometric correction BC = -0.0488 (Flower 1996 2), the star luminosity is estimated to be [FORMULA] [FORMULA]. According to the evolutionary tracks of the Geneva evolution models with appropriate metal abundance (Schaerer et al. 1993), the position of the star in the HR diagram indicates an age comparable to the age of the Sun (but with large uncertainties) and a mass [FORMULA]. The derived mass is slightly higher than the typical mass of G0 dwarfs because of the high metallicity of the star.

From the emission flux in the middle of the CaII H line after correction for the photospheric flux, we obtain the chromospheric activity indicators [FORMULA] = [FORMULA] and [FORMULA] (Santos et al. 1999b), very close to [FORMULA] and [FORMULA] given in Henry et al. (1996). These values are typical for stars with low chromospheric activity level (Henry et al. 1996). From the calibration of Donahue (1993) also quoted in Henry et al. (1996), they lead to an age estimate of 5.6 Gyr compatible with the age derived from the evolutionary tracks (large uncertainties in both determinations).

Observed and inferred stellar parameters are summarized in Table 1 with the references of the corresponding observations or methods used to derived the interesting quantities.


Table 1. Observed and inferred stellar parameters for HD 75289 and HD 130322 with the corresponding references

3.2. CORALIE orbital solution for HD 75289

Between November 1998 3 and October 1999 we obtained 88 precise radial velocities of HD 75289. The distribution of the photon-noise errors of individual measurements ([FORMULA]) peaks around 4.5 m s-1 with a mean value [FORMULA][FORMULA][FORMULA] = 5.5 m s-1. A Fourier transform (using the approach by Deeming (1975, 1976)) of the temporal data exhibits a strong periodic signal around P = 3.5 days, as shown in Fig. 2 (upper panel). Horizontally-dashed lines in the figure represent 0.5, 0.05 and 0.001 false-alarm probabilities derived from Monte-Carlo random realisations with the same temporal sampling and r.m.s. as the real data. The significant secondary peaks are aliases as indicated by the aliasing and residue windows in the figure (lower panels). The best-fit Keplerian model (Fig. 3) yields an accurately constrained orbital period of [FORMULA] days, a very small eccentricity [FORMULA], compatible with a circular orbit according to the Lucy & Sweeney (1971) test at a 92% confidence level, and a semi-amplitude [FORMULA] m s-1 of radial-velocity variation. At such a small distance from its parent star and following Guillot et al. (1996), the planet equilibrium temperature at the surface is estimated to be around 1260 K. The complete set of orbital elements with their uncertainties are given in Table 2 for the circular and "eccentric" solutions.

[FIGURE] Fig. 2. Fourier power spectrum of the radial-velocity measurements (upper panel) for HD 75289. The middle panel gives the alias spectrum due to the data sampling (Deeming method) and the lower panel the residue after removing the contribution of the main peak and its aliases. The 3.5-d periodicity is obvious in the velocity data (mean peak and aliases). In the upper panel, horizontal lines represent (from bottom to top) 0.5, 0.05 and 0.001 false-alarm probabilities

[FIGURE] Fig. 3. Phase-folded radial-velocity measurements obtained with CORALIE for HD 75289. Error bars represent photon-noise errors


Table 2. CORALIE best Keplerian orbital solutions derived for HD 75289 and HD 130322 and inferred planetary parameters. For HD 75289, the eccentricity is not significant according to the Lucy & Sweeney (1971) test and we thus also provide the circular orbital solution

Using the previously derived 1.15 [FORMULA] mass for HD 75289, the best-fit orbital parameters imply a companion minimum mass [FORMULA] [FORMULA] and a separation [FORMULA] AU between the star and the planet. The inferred planetary minimum mass corresponds to 1.4 times the mass of Saturn and is the lightest found to date for an extra-solar planet around a solar-type star. Despite the low mass of the planet, the velocity variation is accurately determined and the error on the minimum mass (without taking into account the uncertainty in the primary mass estimate) is less than 2%.

From the relation between the activity index and stellar rotation period (Noyes et al. 1984) we derive a period [FORMULA] = 16 days for HD 75289. Assuming that the orbital and rotation axes coincide, a "statistical" equatorial velocity [FORMULA] may be derived from the star radius and then the orbital plane inclination is obtained from the measured projected rotational velocity [FORMULA] [FORMULA] km s-1 (calibrated from the width of the CORAVEL cross-correlation dip; Benz & Mayor 1984). Using the simple relation between stellar luminosity, radius and effective temperature [FORMULA] and with the stellar parameter values given above, the radius is estimated to be [FORMULA] [FORMULA], slightly larger than the typical radius for a G0 dwarf with solar metallicity. This leads to a value [FORMULA] km s-1 very close to the quoted [FORMULA]. Taking into account the uncertainty on [FORMULA], a maximum mass of 0.51 [FORMULA] is estimated for the planet.

The r.m.s. to the Keplerian fit is 7.5 m s-1, similar to the velocity r.m.s. for the orbital fit of the previously-discovered CORALIE planet around Gl 86 (Paper I). The residuals around the solution show no evidence of a possible additional companion on a short- or intermediate-period orbit.

A set of 5 CORAVEL radial-velocity measurements of HD 75289 obtained over slightly less than 14 years, between 1983 and 1997, present a dispersion of [FORMULA] m s-1, the level of precision of CORAVEL. Older measurements from the literature (Gratton et al. 1989, Parsons 1983, Catchpole et al. 1982, Przybylski & Kennedy 1965) do not show any clear evidence of radial-velocity variation. We thus conclude that HD 75289 is very probably a single star contrarily to the "potential binary" characteristics reported in the literature.

3.3. Transit candidate

The discovery of hot Jupiters has major theoretical and practical implications. For favourable inclination (edge-on orbital plane), high-accuracy photometric eclipse observations become possible. The transit light curve provides the inclination angle and the radius of the transiting body (using the on-transit and off-transit flux ratio). The probability of such a transit is proportional to ([FORMULA]+[FORMULA])/a (stellar radius over star-planet separation, e.g. Borucki & Summers 1984) and is typically about 10% for a hot Jupiter on a 4-day orbit. Combining the photometric data of a detected transit with radial-velocity measurements it then becomes possible to determine the real mass of the companion and therefore to determine the nature of the transiting body. Finally, from the real mass and radius, we get the mean density of the companion, bringing thus strong constraints to theories of planet and brown-dwarf formation and atmosphere models. Such an observation would also represent a direct detection of an extra-solar planet.

With its 3.5-day period, the HD 75289 system is a good transit candidate. Furthermore, as discussed above, a favourable geometry could be expected from activity indicator and rotational velocity considerations. During the winter 1998-1999, an intensive campaign of high-precision relative photometry has been initiated on the 50-cm Danish telescope at La Silla in order to detect a possible transit. Future results of this ongoing monitoring will be described in details in a coming paper (Santos et al., in prep). Up to now the photometric measurements have already demonstrated the luminosity stability of the star at a few millimagnitude level.

During the preparation of this paper, a planetary companion transiting in front of the star HD 209458 was independently observed by Charbonneau et al. (2000; two transits in September 1999) and Henry et al. (2000; in November 1999).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: April 10, 2000