Proc. ASA! ["[) 1987 Russell et al. 65 Stromgren photometry of supergiants in the Magellanic Clouds S. C. Russell, M. S. Bessell, and M. A. Dopita, Mount Stromlo and Siding Spring Observatories, The Australian National University. Abstract: This is a preliminary report on the first Stromgren—H/3 photometric survey of yellow supergiants in the Magellanic Clouds. Less luminous supergiants have been chosen in preference for this program in order to facilitate calculation of their physical parameters. This will not only provide a basis for future detailed spectral analysis of these stars, but will allow a better calibration of the colours derived from model atmospheres for this type of star. The colours of these stars are more affected by changes in metallicity than the hotter stars more frequently studied, and since they are also young, they reflect the metallicity of their local Interstellar Medium (ISM). If indeed the Stromgren abundance index mi can be calibrated successfully for these stars, then their observation in any galaxy will provide the metallicity of that particular ISM. Our results show that it is important for future colours to be calculated over the range of turbulent velocities important for supergiants. Introduction This paper details the results from an ongoing photometric survey of A5 to GO supergiants in the Magellanic Clouds. Stromgren — H0 photometry was chosen over other systems since it provides the maximum amount of information for the spectral classes we are interested in and has relatively good theoretical colours already available (Relyea and Kurucz 1978). Indeed, this was seen as an ideal opportunity to calibrate these colours for low gravity, high luminosity stars. The sample of high luminosity stars in our own galaxy suffers critically from interstellar absorption and reddening and hence from uncertain distance scales and unknown colours for model fitting. In the Clouds this is not a problem. Caldwell and Coulson (1985) showed that the intrinsic reddening for Cloud stars was small with a spread of the order of only a hundredths of a magnitude in b-y, while McNamara and Feltz (1980) showed that the Galactic foreground reddening in the direction of the Clouds is also very low. In addition, the distances of the Clouds are quite accurately known, thus allowing good calibration of the models of Relyea and Kurucz (1978). The spectral range A5 to GO is of special interest for several reasons. These stars form a natural group within the Stromgren system since for later spectral types there are no calibrations or models, while for earlier types the ci index becomes double valued and the b-y index becomes insensitive to temperature. In addition the crucial mi index for abundance is most sensitive in this range of spectral types, and loses all sensitivity for stars earlier than B9. The only Stromgren photometry previously obtained on such relatively late stars in either Magellanic Cloud was by Osmer (1973), with his survey of the brightest stars in the Small Magellanic Cloud (SMC). However, in a survey of 169 stars, he observed only 3 stars of type F0 or later, one of which was G5 and hence beyond the range of the available models, and the other two being of luminosity la. Being of such high luminosity and therefore low gravity, the physical parameters of the later two stars are more difficult to model. They suffer from a breakdown in the classical assumptions made in order to simplify the models. Thus the effects of non local thermodynamic equilibrium, non radiative equilibrium, unpredictable atmospheric turbulence, unpredictable mass loss and problems with the treatment of convection, are all likely to be important. With this in mind, less luminous stars were chosen for observation in the hope that these results may form a basis from which to extend the models to the more luminous stars. In this manner it is hoped that observations of the very brightest stars in more distant galaxies will furnish us with valuable information on the galaxies themselves. Background Theory In the Stromgren four colour system the y (5470A) filter transforms very accurately to the broad band V filter thus providing a measure of the magnitude. This filter and the b (4670 A)filterboth cover regions in which there is low blanketing and there are no strong lines, so the index (b-y) provides a good measure of the continuum slope. This correlates well with effective temperature for stars later than A3 but does depend to some degree on gravity. The index mi, defined as iy-b)-{b- y), is a measure of the blanketing in the region of the v (4160 A) filter, and is thus an indicator of metal abundance. The index Ci defined as (u-v)-(y-b), provides a measure of the Balmer jump, where the u (3500A) filter, being wholly beyond the Balmer discontinuity, is better suited for this job than the broad band U filter. The height of the Balmer jump correlates well with differences in gravity for spectral types of A5 and later. Moreover, since the blanketing in the region of the u filter is roughly twice that of the vfilterregion, the cj index is insensitive to variations in abundance. In addition to the basic four colour Stromgren system two morefiltersare often used, both of which are centred on the H/3 line, one being ten times the width of the other. The difference between the two measurements gives an indication of the H/3 line strength which is totally free from reddening. The H/3 index can thus be used in conjunction with the reddening dependent indices b-y and Ci to determine the reddening, at least up to a spectral type of GO, beyond which it loses all sensitivity. Relyea and Kurucz (1978) calibrated theoretical uvby colours from the Kurucz (1978) grid of model atmospheres for stars as late as GO, while the H/3 indices were calculated from the same model atmospheres by Schmidt (1979). Since molecular opacities were not included in the models they were not expected to hold true for temperatures less than 6000K. As a test of the models' validity, Relyea and Kurucz superposed the solar abundance grids on plots of all the stars in the Hauck and Mermilliod (1975) catalogue for the different indices. The https://www.cambridge.org/core/terms. https://doi.org/10.1017/S132335800002186X Downloaded from https://www.cambridge.org/core. IP address: 107.172.61.10, on 23 May 2019 at 12:07:58, subject to the Cambridge Core terms of use, available at