Astron. Astrophys. 322, 266-279 (1997)

4. Conclusions

We have investigated the potential of the Na I D doublet as chromospheric diagnostics in a cool dwarf of  K, and solar metallicity. Using an extended grid of model chromospheres, we have computed both H I and Na I non-LTE spectra in a variety of situations corresponding to a broad span of chromospheric activity.

From a comparison of the spectra of the two elements, we have shown that the information to be gathered from the D lines can effectively complement that obtained with other diagnostics, such as H or Pa . In particular, confirming previous investigations, we find that H I lines are chiefly sensitive to the structure of the upper chromosphere-lower transition region. In contrast, our calculations indicate that the cores of the D lines are nearly insensitive to the structure of the lower transition region, while being a good probe of the conditions in the middle-to-lower chromosphere.

Moreover, while the H I lines, when in emission, seem to depend mainly on the the density of the upper chromosphere and/or lower transition region, the emission cores of the D lines appear to depend both on the density and on the thickness of the chromosphere. For example, at a given chromospheric density, the extension of the chromosphere can have a large effect on the amount of inner-core self-absorption.

We have also pointed out how the Na ionisation equilibrium and, consequently, the emerging D -line profile can be affected by the presence of an active corona. The coronal radiation acts on the Na I lines mainly via the dependence of the emission core on the optical thickness of the chromosphere in the D doublet. This effect is normally not important, but may be significant for very active stars.

Another effect that should be accounted for in cool atmospheres, is the additional thermalisation due to inelastic collisions with hydrogen. We have included this effect in all our calculations, using a formula appropriate for Rydberg atoms. We have showed that the inclusion of such collision rates can only marginally alter the Na I D line profiles in the models we have considered. To some extent, this remains true even if the alternative, probably overestimated Drawin cross-sections are adopted. This is because much of the line profile is formed in regions where departures from LTE are not dramatic, while the inner cores form in regions where the electron density is high enough to reduce the importance of collisions with hydrogen.

Finally, we have included the effect of the severe atomic and molecular line haze in so cool an atmosphere by suitably adapting to our grid of chromospheres a model-specific opacity table. This has enabled us to estimate the importance of a realistic treatment of photospheric background opacities. The effect is especially important in the broad damping wings of the D doublet, but it is not negligible even in the emission cores produced by the most active models.

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de