Astron. Astrophys. 326, 249-256 (1997)

2. The sample stars

The sample consists of two dM stars of the same spectral type: Gl 588 and Gl 628, in addition to AD Leo, already modelled in Paper I. The main parameters of the three stars are given in Table 1. The data sources are indicated in the notes at the end of the Table. The value R/R are calculated using the relation of Pettersen (1980)

where the colors in the Johnson system are obtained from those in the Cousin system trough the relation given in Legget (1992). The values of R/R , that can be obtained with the various relations found in the literature, differ within 10-15% and this uncertainty propagates on the surface flux causing an uncertainty of 20-30%.

Table 1. Parameters of the sample stars: V magnitude, Cousin B-V, V-R, R-I colours, trigonometric parallax (in mas), mass M, radius R and gravity g, relative to their solar values, logarithm of the conversion factor C from the observed to surface flux, rotational velocity (in Km/s), and Mg II, Ca II, and X-ray surface fluxes in erg cm^-2 s^-1. Negative fluxes represent absorption lines.

GL 588 and Gl 628 are very low activity stars. In particular, Gl 628, with a very low Mg II flux and no detected Ca II flux prior to our observations, had been included in a list of "basal stars" by Mathioudakis et al. (1994). They defined these stars as those that have the minimum level of chromospheric activity. Moreover, Gl 628 is a standard star of the UBV system, but shows long term variability at the 99% confidence level (Weis 1994). Also Gl 588 has been classified as a low activity star by Andrews et al (1990).

AD Leo, on the other hand, is a well known flare star, which shows the Balmer lines in emission, and much larger levels of Mg II and Ca II flux. It has also a much higher X-ray emission than Gl 628 and Gl 588 (Fleming et al. 1995). We refer the reader to Paper I for a review on the literature on this star.

The spectra we use for Gl 588 and Gl 628 were obtained by us in May and June 1989 at ESO, La Silla. Low resolution spectrophotometry was obtained at the ESO 1.5m telescope equipped with the B&Ch spectrograph in the spectral range 3900-8600 Å. To minimize the effects of slit losses and atmospheric differential refraction, a wide (8 arcsec) slit was used; spectrophotometric standards were observed with the same configuration every night. The data have been reduced using the MIDAS package (Banse et al. 1988) and standard absorption coefficient for La Silla have been adopted. Due to the large slit, the real resolution of the spectra was set by the stellar disk. Although this is slightly variable from one object to another, the intrinsic optical quality of the ESO 1.5m telescope and the hand guiding available at the time of the observations set the stellar image to a diameter of around 2 arcsec, which correspond to a resolution of 7 and 14 Å at the blue and red ends of our spectra, respectively. Since the low resolution spectra have been used only to set the basic parameters of the stellar models and to rectify the high resolution echelle spectra, the difference in resolution between the two gratings used in the red and blue part of the spectrum is not relevant for our results.

High resolution spectra were obtained using the CASPEC spectrograph at the 3.6m telescope (Pasquini and D'Odorico 1989). CASPEC was used in combination with the F/1.6 Short Camera. With a slit width of 2 arcsec, the nominal resolving power of this configuration is 18000. The real resolving power has been measured as the FWHM of unsaturated, unblended lines in the comparison (Th-A) spectra, finding values of R 24000, in good agreement with the nominal one. A spread of 10 in resolution over each frame is present, due to the intrinsic focus spread of the camera.

In order to minimize the effects induced by possible long term variability in the chromospheric lines, all observations of the same star were performed as close in time as possible, and were completed within one night. The error expected on the absolute calibration of our spectra is about 30 %.

© European Southern Observatory (ESO) 1997

Online publication: April 20, 1998
helpdesk.link@springer.de