4. Photometric, spectral and magnetic variability
According to Renson & Manfroid (1978), the photometric variability period of For is P = 1.89 days. Borra & Landstreet (1980) found their measurements of the effective magnetic field appear to be variable with the periods 1.877, 1.892 and 1.908. With this period, the photometric curves of Renson & Manfroid (1978) vary as the function (Mathys & Manfroid 1985):
To determine the value of P, we have performed a least-square fitting with the previous function of the magnitudes () of For measured with the Hipparcos satellite (European Space Agency 1997). The resulting period is 1.89250.0002 days, the given error represents the change on the period that increases the of a unit. Almost the same period value (1.89240.0004) is obtained by using the Phase Dispersion Minimising routine distributed with IRAF package. Fig. 1 shows the Deeming (1975) power-spectrum graph of Hipparcos data and the associated spectral window. It is clear that aliases can be excluded. Peak position is at 1.8927 days, confirming the previous period value.
Unfortunately Renson & Manfroid (1978) photometric data have not been published, however from Mathys & Manfroid's (1985) fitting of these data, we find that on JD = 2 443 456.147 there was a photometric minimum. Thus, we adopted the ephemeris:
that is, within our fitting error, the period which forces the Hipparcos and Renson & Manfroid (1978) light curves to have the minimum at the same phase.
With this ephemeris, we have phased the Hipparcos photometry, the Borra & Landstreet (1980) measurements of the effective magnetic field (Fig. 2) and the equivalent widths, measured here, of the strongest lines (Fig. 3). We note that the light maximum shows the phase of the magnetic positive maximum.
As is expected for a magnetic chemically peculiar star, For is characterised by spectral variability. Fig. 3 shows the equivalent width variations, phased with the previous ephemeris. The variability of iron and chlorine lines is in phase with the light variability. In contrast, helium, magnesium, silicon and chromium lines are out of phase.
Borra & Landstreet observations were obtained from JD = 2 443 498.581 to 2 443 741.888. They were so close to the light minimum instant determined here that a negligible error (0.005) is expected for the phase relation between light curves and magnetic variation. Because of the time interval between Hipparcos photometry and our spectroscopic observations, a 0.1 error is possible for the previous phase relations between the light/magnetic and spectral variations.
© European Southern Observatory (ESO) 2000
Online publication: July 7, 2000