Astron. Astrophys. 364, 137-156 (2000)

1. Introduction

The modelling and interpretation of the ISO-SWS (Infrared Space Observatory - Short Wavelength Spectrometer) data require an accurate calibration of the spectrometers (Schaeidt et al. 1996). In the SWS spectral region (2.38-45.2 µm) the primary standard calibration candles are bright, mostly cool, stars. The better the behaviour of these calibration sources in the infrared is known, the more accurate the spectrometers can be calibrated. ISO offered the first opportunity to obtain continuous mid-infrared spectra between 2.38 and 45.2 µm at a spectral resolving power of , not polluted by any molecular absorption of the earth's atmosphere. Therefore our knowledge on the mid-infrared behaviour of the stellar calibration sources is limited. Refining the synthetic reference spectra used to calibrate the SWS can only be done by refining the model atmospheres of the stars. A full exploitation of the ISO data can therefore only result from an iterative process in which both accurate observations and new modelling are involved. In order to obtain a reliable convergence in such an iterative process, one needs: (1) a sample of bright stellar objects spread over a large range in spectral type; (2) a thorough understanding of cool star atmospheres and of the influence of the various parameters - e.g. , g, and chemical composition - on the emergent spectra. With this in mind, stars with spectral types ranging from A0 - M8 were observed with ISO-SWS. It is important to cover a broad parameter space in order to distinguish calibration problems from problems related to the model and/or in the generation of the synthetic spectrum. Concerning point 2 above, it is necessary to perform an intensive study of the influence of different stellar parameters on synthetic spectra. The results of such a study will be presented with the red giant Tau as a test case. The results for 15 other stars in our sample will be presented in forthcoming papers.

This research has been done within the framework of the proposal STARMOD (Title: Accurate modelling of cool-star atmospheres; P.I.: C. Waelkens; C.I.: M. Cohen, L. Decin, Th. de Graauw, L.B.F.M. Waters) and the ZZSTARM proposal (Title: Accurate modelling of cool-star atmospheres; P.I.: L. Decin; C.I.: M. Cohen, C. Waelkens, Th. de Graauw, L.B.F.M. Waters). Some calibration data have been provided by the SWS Instrument Dedicated Team (SIDT) in the framework of a quick-feedback refining of the model Spectral Energy Distribution (SED) of the calibration sources used for the SWS calibration.

So far, the analysis of the discrepancies between the ISO-SWS data and the corresponding synthetic spectra has been restricted to the wavelength region from 2.38 to 12 µm, since the lack of comprehensive molecular and atomic line lists hamper fast progress at longer wavelengths (12-45 µm). Furthermore the brightness of the stars drops quickly in this wavelength region so that the same signal to noise ratio will not be achieved. A third point is that the SEDs may also be affected by unknown circumstellar contribution.

This paper is organized as follows: in Sect. 2 the observations are described and the data reduction procedures are discussed. In Sect. 3 a summary of the literature concerning Tau is presented, on the basis of which the starting values for our modelling stellar parameters are selected. The effect of changing stellar parameters on the synthetic spectra of K and M giants is analyzed in Sect. 4. In Sect. 5 the method of analysis is outlined, while in Sect. 6 the results are discussed. In the last section, Sect. 7, the main conclusions are summarized.

© European Southern Observatory (ESO) 2000

Online publication: December 15, 2000
helpdesk.link@springer.de