## 1. IntroductionIn a previous paper (Solanki & Unruh 1998, henceforth Paper I), we have shown that the spectral irradiance variations cannot be modelled by a simple temperature increase alone. They can be relatively well fitted, however, with a three-component model consisting of quiet-sun, spot and facular contributions. Our earlier approach was relatively crude in the way the active flux was calculated: it relied on the simplifying assumptions that the flux was given by Planck's function at each wavelength point and that the formation height of the quiet-sun and the facular contributions was the same. The simple model of Paper I was compared with a far wider variety of data by Fligge et al. (1998), henceforth Paper II). On the whole, the model fared reasonably well, suggesting that the basic approach is sound. There were, however, differences to the observations that demanded improvements. In this paper, we drop some of the original simplifications and model the different contributions much more accurately. While the assumption of LTE is kept, we now calculate the intensities and fluxes using Kurucz's ATLAS9 (Kurucz 1992a; Castelli & Kurucz 1994) spectral synthesis code. In addition, we also present the first calculations of the spectral facular contrast as a function of limb distance. These results are compared with the relevant observations. Knowledge of the spectral facular contrast as a function of limb angle is an important ingredient when modelling the irradiance from a known surface distribution of faculae and spots. There is considerable debate about the plage contributions to the Sun's irradiance variations, and in particular about the question of whether an additional contribution, arising e.g. from latitude-dependent temperature changes (Kuhn et al. 1985, 1988; Kuhn & Libbrecht 1991) or from changes in the magnetic network (Foukal et al. 1991), is needed. The current paper is the next step in improving the modelling and understanding of spectral irradiance variations. We stress, however, that we do not use the centre-to-limb variations of the spectral facular contrast to reconstruct the irradiance in this paper. This will be the subject of a further publication. In the following section we introduce our model. We then calculate the intensities of the different components. Sect. 3 compares the model limb-dependent intensities of the quiet Sun to the observations of Neckel & Labs (1994) and Sect. 4 looks at the colour dependence and the limb behaviour of the facular contrast. In Sect. 5 we integrate the intensities to obtain fluxes and proceed to compare our results with disk-integrated solar data in an analysis similar to the one performed in Papers I and II. © European Southern Observatory (ESO) 1999 Online publication: April 19, 1999 |