Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 331, 619-626 (1998)

Previous Section Next Section Title Page Table of Contents

3. Selection of stars

The effective temperatures of luminous, cool stars have been fairly well determined by lunar occultation to spectral types as late as M6, and additional measurements with optical and infrared interferometry have confirmed the correctness of this calibration at the level of approximately [FORMULA] K. Extending this calibration to still cooler stars faces a number of difficulties. Though luminous, they are relatively faint in the visible, hence have not been the subject of intensive scrutiny during the classical era of stellar astronomy. They are variable in flux, color and spectral type. Their luminosities are not well known. Relatively few stars have even been classified as M8 giants.

As with many difficult problems, the best procedure is the simplest. The source list (Table 2) for the calibration program includes the "prototype" examples, that is the stars which have been selected, on basis of brightness, minimal variability, and "normal" spectrum to be primary or secondary spectral type standards (Morgan & Keenan 1973). Possible additions to this list were considered, but in each case the problems of poorer spectral type information, greater variability or lack of variability information, and increasingly uncertain extinction (for fainter stars) caused greater concern than the shortness of the list. The most immediate improvement in this list would be to add a few comparable southern stars which were not accessible from Mt. Hopkins. No doubt this list can be further improved, but possibly not without significant ancillary preparation.

In addition, the well studied stars [FORMULA]  Boo and [FORMULA]  Tau were observed as consistency checks against other interferometers and occultations, and the intermediate type star [FORMULA]  Oph was added to check the connection of our scale to previous ones for earlier stars.

Stars later than M8 were excluded. Types M9 and M10 have been used only for photometric classification of long period variables, which are known to follow a different effective temperature calibration than the normal giants (van Belle et al. 1996).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: February 16, 1998