Astron. Astrophys. 339, 858-871 (1998)

2. The database

A homogeneous set of 769 photometric standard stars for the ISO mission have been selected as suitable preliminary candidates with spectral types and visual magnitudes derived according to the Hipparcos Input Catalogue (Turon et al. 1992). High-precision broadband K magnitudes have been measured for 565 stars of the Northern Hemisphere in the TCS magnitude system (Hammersley, to be published) and for 204 stars of the Southern Hemisphere in the ESO magnitude system (van der Bliek, Bouchet & Manfroid 1996) with 35 stars in common. All the near-infrared data have been converted into the Johnson magnitude system, i.e. the standard reference photometry of the calibrations throughout this paper, according to the following transformations:

where is the standard error determined from the overall scatter around the mean relation. Since the TCS and ESO near-infrared colours have accuracies of about 0.02 mag, most of the scatter is likely due to the less accurate Johnson near-infrared photometry. The transformations are plotted in Fig. 1.

Fig. 1. Comparison between broadband K magnitudes against the broadband near-infrared colour on the Johnson scale. Top: Johnson and TCS systems. Bottom: Johnson and ESO systems. Solid lines: least squares fits [see Eqs. (1)]

Most of the 769 stars have their accurate trigonometric parallaxes measured by the Hipparcos satellite (ESA 1997). This provides a valuable step towards a fundamental calibration, since the interstellar extinction to get intrinsic integrated fluxes and colour indices has been derived according to the absolute distances from parallaxes. To be consistent with the bolometric flux measurements, I shall adopt below a visual absorption = 0.8 mag/kpc along the line of sight to each star (Blackwell et al. 1990). The near-infrared colour excesses will be derived according to / 1.1 (Wegner 1994).

Several conditions have to be met by the ISO standard stars, if their effective temperatures must be determined at the required target accuracy of 1 %. For this reason, 94 stars have been excluded from the actual calibration, since they are of O, B, M spectral-type and/or of luminosity class I. In addition, a subset of 127 objects is discarded because of their variability in the visible (more than 0.1 mag) or multiplicity according to the Hipparcos Input Catalogue or the absence of Hipparcos data. A further small number of 11 stars have their intrinsic colours out of the overall range covered by the actual calibration of A-F-G-K dwarfs and giants.

The final selected set of 537 stars includes 393 stars observed in the Northern Hemisphere and 144 stars observed in the Southern one with 25 stars in common. The bolometric fluxes along with the broadband near-infrared photometry are only available for a subset of 327 stars with 22 of these having K-magnitudes in both TCS and ESO systems. The bolometric flux measurements have been determined by Blackwell & Lynas-Gray (1998; hereinafter BL) for application of the semiempirical narrowband IRFM to the ISO standard stars. Tables 1 and 2 list the 393 Northern and 144 Southern stars, respectively, with HIC and BS identifications, spectral-type, observational V, K photometry, Hipparcos parallax and bolometric flux . The last two columns report the intrinsic broadband near-infrared colour converted into the Johnson magnitude system according to the Eqs. (1) and the final stellar effective temperature according to the calibrations discussed below.

© European Southern Observatory (ESO) 1998

Online publication: October 22, 1998
helpdesk.link@springer.de