Astron. Astrophys. 345, 635-642 (1999)

3. The quiet photosphere: centre-to-limb variations

Before we determined the facular contrast as a function of limb angle, we first checked the centre-to-limb variation (CLV) of the intensity at different wavelengths produced by the quiet-sun model against the measurements made by Neckel & Labs (1994) . This allows us to test in how far the intensity calculations approach the solar behaviour at all limb angles, and can give us a rough indication of the importance of effects that we have neglected, such as non-LTE, granulation and other inhomogeneities.

The dashed line in Fig. 2a shows the disk-integrated flux divided by the intensity at disk centre as measured by Neckel & Labs (1994) (see their Fig. 3a) and the solid line shows the same quantity as obtained with ATLAS9 and the radiative-equilibrium solar model atmosphere of Kurucz (1992a) .

Fig. 2. a  The dashed line shows the disk-integrated flux of the Sun divided by the intensity at disk centre as measured by Neckel & Labs (1994). The solid line shows the same quantity as calculated with ATLAS9 and Kurucz's solar model atmosphere. b  The CLV of the quiet-sun flux for a number of selected filters. The lines show the fits from Neckel & Labs (1994), the symbols show the intensities obtained with ATLAS9. The numbers next to each line indicate the central wavelength of the narrow-band filters in nm.

Our calculated flux-to-intensity ratios are consistently lower (by less than 1%, however) than the measurements of Neckel & Labs (1994) . A comparison between measured and modelled CLV of a number of selected filters (plotted in Fig. 2b), indicates that this is mainly due to our lower intensities at intermediate limb angles ( to 0.6). Apparently, small departures from radiative equilibrium are present in the solar atmosphere, which is not surprising in view of the presence of solar convection.

The calculated disk-centre intensities also agree reasonably well with the disk centre measurements by Neckel & Labs (1984) and Burlov-Vasiljev et al. (1998a, 1998b). In the blue, our calculations are somewhat closer to the measurements by Neckel & Labs (1984) , but they show better agreement with the data by Burlov-Vasiljev et al. (1998a) in the red. The average deviation between the calculated and measured fluxes over the range of 330 to 1050 nm is of the order of 1.5% for both data sets, the deviation between the two observed intensity sets being 0.8%. When only the wavelength range between 450 and 1050 nm is taken into account, the deviation between all three datasets is below 0.7%.

© European Southern Observatory (ESO) 1999

Online publication: April 19, 1999
helpdesk.link@springer.de