Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 320, 185-195 (1997)

Previous Section Next Section Title Page Table of Contents

5. Stellar properties

We have estimated the effective temperatures [FORMULA] from spectral types given by Krautter et al. (1996 ), using the calibration of Bessel & Brett (1988 ) and Schmidt-Kaler (1982 ).

For the determination of bolometric luminosities [FORMULA] we calculated extinctions towards the individual stars using their ([FORMULA]) colours and intrinsic colours from Bessel (1991 ). Then apparent bolometric luminosities were calculated by normalizing a blackbody to the dereddened flux in the [FORMULA] band, where the effective temperature of the star was chosen as blackbody temperature. We have checked this method by using the methods described in Alcalá et al. (1996b ), i.e. (i) integrating the spectral energy distribution and (ii) using a bolometric correction in [FORMULA]. All three methods are found to give consistent results with very small ([FORMULA] dex) systematic differences. Systematic offsets of this size are also found by Alcalá et al. (1996b ), see Fig. 1 in their paper. We use the method of blackbody normalization in order to follow the procedure of Hughes et al. (1994 ), thus enabling us to compare our results directly with their study of Lupus CTTS.

From spectroscopic parallaxes of field stars Hughes et al. (1993 ) estimated a distance of 140 [FORMULA] pc for the Lupus SFR. At this distance the angular extent of the area surveyed by Krautter et al. (1996 ) (about [FORMULA] degrees) converts to a linear extent of about [FORMULA] pc. As WTTS were found not only near the dark clouds, but spread over the whole investigated area, we have to assume that the spatial distribution of WTTS extends radially also a few tens of parsecs from the clouds, thus providing an additional source of error for the bolometric luminosity [FORMULA] of individual stars. However, assuming that the distribution of the stars is more or less uniform in all directions, the mean distance of the new WTTS in Lupus might not be much different from the mean distance of the CTTS.

Other sources of errors encountered in the determination of [FORMULA] and [FORMULA] are thoroughly discussed in Hughes et al. (1994 ) and Hartigan et al. (1991 ). Errors in log [FORMULA] are small ([FORMULA] dex), but errors in the estimate of [FORMULA] can introduce errors in log [FORMULA] on the order of [FORMULA] dex.

The stellar radii were determined using the effective temperature [FORMULA] and the bolometric luminosity [FORMULA]. As discussed by Alcalá et al. (1996b ), this method yields results in good agreement with the Barnes-Evans method (Barnes & Evans  1976 ). Based on this consistency, as well as on the study of the scatter in the Barnes-Evans relation by Lacy (1977 ), we estimate an rms error of about 0.2 dex in [FORMULA].

Following Hughes et al. (1994 ), for the determination of masses and ages we use the evolutionary tracks of D'Antona and Mazzitelli (1994 ) with Canuto and Mazzitelli convection and Alexander opacities (Alexander et al.  1991 ). Again, this choice is taken in order to compare our results to the previous study of Hughes et al. (1994 ). Fig. 2 shows the H-R diagram of the new WTTS in Lupus studied by us. In Table 7 the derived stellar properties are given.

[FIGURE] Fig. 2. H-R diagram of Lupus TTS. Filled triangles denote CTTS, filled squares WTTS in regions of high obscuration, open squares WTTS found in regions of low obscuration. Data for CTTS and the WTTS known prior to ROSAT are taken from Hughes et al. (1994). As discussed by Huges et al., the two CTTS below the ZAMS are heavily veiled, thus the extinction may be incorrect.
Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998