4. Unlikely explanations
4.1. A bright spot on the star
The apparent similarity of the mean magnitude level of the star and the minimum level on JD 4918 is very interesting. This could be explained very naturally by a bright phenomenon which disappears behind the star during the JD 4918 measurements.
A bright spot on the rotating star would have been able to produce such an effect. However Pic is an A5V star with probably no convective zone reaching its surface and thus probably with no photospheric bright spot. In any case, with and a rotation period of 16 hrs, a spot on the star is visible during less than 8 hours, and measurements on JD 4917 and 4919 are spread over more than 7 hours. During this time, the putative bright spot should have strongly changed the magnitude measurements, which did not occur.
The interactions between a shower of FEBs and the stellar photosphere might have produced a temporary brightening. However, again in this case one might expect to see some rotational modulation, which is not observed. We conclude that the photometric variations of Pic are probably not due to a bright spot on the stellar surface.
4.2. A cloud of backward scattering dust
In the same manner, we could expect that a cloud of backward scattering dust moving behind the star could be responsible for the observed variations. The increased emission can suddenly be hidden behind the star, to explain the coincidence of the mean magnitude level and the minimum observed on JD 4918.
The variation of the star brightness (F) due to a back scattering cloud can be described as follows:
with , where is the scattering phase function normalized by , is the scattering efficiency and is effective scattering radius of the dust cloud at a distance d from the star. With and for the zodiacal dust or a planetary ring (Hong 1985) we must have . This would require a very large dust-cloud with a radius approximately equal to its distance from the star. Such a large cloud is inconsistent with the short duration of the dip in the lightcurve and could not be totally hidden if it is larger than the star.
4.3. Gravitational lensing
As already suggested by Lecavelier des Etangs et al. (1995), the abrupt drop in Pic could be due to a planet passing in front of the star. If this is true, we need to find an explanation for the brightness increase before and after the occultation due to the environment of the putative planet.
In this case, the well known microlensing effect (e.g., Pacziski 1986) is in fact not efficient. The Einstein radius of a planet around Pic is
where and are the mass of the putative planet and Jupiter respectively, d its distance to Pic. This radius is smaller than the planet size, so gravitational lensing can not account for the observed brightness increase.
4.4. Refraction by the atmosphere of a planet
The expected brightness increase due to refraction by the atmosphere of a planet is:
where H is the scale height of the atmosphere ( km) and is the refraction angle. We assume a distance of km and a planet radius of km. Just before the planet occultation the diffraction angle is . So diffraction by the atmosphere of the planet would produce a brightness increase of . This effect is too faint to explain the observed brightness increase.
Note however, that if the planet is surrounded by a large disk or torus of gas with a scale-height of , this explanation of the brightness increase might be acceptable.
© European Southern Observatory (ESO) 1997
Online publication: March 24, 1998