Astron. Astrophys. 336, 613-625 (1998)

5. Conclusions

We have presented the first systematic exploration of the Na I D lines in early M dwarfs of various activity levels. We have confirmed the prediction of ADB that these lines show similar behavior to that of previously known chromospheric diagnostics such as Ca II HK: a brightening of the line core with increased chromospheric pressure that culminates in a central emission reversal in the highest activity models.

For the dMe stars of high chromospheric pressure (Gl 388, Gl 494) and the one dM(e) ("zero H") star of intermediate chromospheric pressure, the model that provides the closest fit to the H line has a value of that is approximately 0.4 dex lower in column mass density than that which provides the closest fit to the Na I core. The discrepancy is significant in the sense that the model that provides the closest fit to one of these lines cannot even approximately fit the other. However, the discrepancy is small enough that models fit to each line are in the same general regime of chromospheric pressure. The consistency of the discrepancy among the three different stars suggests a systematic effect. For the dMe stars the Na I line cores are in emission and the determination of the chromospheric thickness (model series) that fits most closely the star is determined entirely by the shape of the core. Whereas the H line profile is negligibly affected by the chromospheric thickness (or, equivalently, chromospheric steepness), the Na I D emission core is affected significantly, varying from a double peaked profile with a clear central absorption reversal for the thicker Series models to a single peaked profile for the thinner Series models. For this reason, the Na I D line core is a powerful diagnostic complement to the H line for active M dwarfs.

The two dM stars of low chromospheric pressure (Gl 212, Gl 382) are unique in their ability to be fit. For Gl 212 we are able to fit both lines closely with a model of the same value of and the same chromospheric thickness. By contrast, the closest fit model to the H line of Gl 382 has a value of that is approximately an order of magnitude lower than that which fits most closely the Na I D core. For the dM stars, in which both lines have cores that are in absorption, neither line distinguishes chromospheric thickness as clearly as the Na I D emission core does for the dMe stars.

An important caveat to the use of either emission line is the sensitivity of the computed emission core contrast to the adopted value of both and . Lowering or raising by K approximately doubles or halves the predicted brightness contrast of the emission core. Lowering or raising by dex has a similar effect. As a result, the value of the chromospheric pressure that is inferred from a closest fit may be incorrect by an amount corresponding to 0.3 dex in the mass loading at the onset of the TR and the location of . Given the uncertainties in measured stellar parameters due to the faintness of M dwarfs, and the lack of measurements, this places significant uncertainties on the modelling of emission spectra in active M dwarfs.

© European Southern Observatory (ESO) 1998

Online publication: July 20, 1998
helpdesk.link@springer.de