Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 341, 768-783 (1999)

Previous Section Next Section Title Page Table of Contents

7. Conclusions

We have detected magnetic fields of [FORMULA] kG strength in the classical T Tauri stars T Tau ([FORMULA] kG) and LkCa 15 ([FORMULA] kG). For the weak-line T Tauri stars LkCa 16 ([FORMULA] kG) and for the cTTS UX Tau A ([FORMULA] kG) the detection is only marginal (despite the formal [FORMULA] kG for UX Tau A see Sect. 6.1).

GW Ori clearly shows the limitations of the method. The reason that our method fails for the more rapidly rotating stars (e.g. GW Ori) is not the limited accuracy of the method for rotationally broadened single lines, but the fact that the number of unblended lines is too small in these stars. Here we restricted ourselves to blend-free lines, however the capability of the equivalent width method to deal with rapidly rotating objects could potentially be extended by including line blends in the calculation of the EWs.

A further possible improvement of the method would be an iterative determination of the parameters of the model atmospheres of the template star as well as of the field-free and field covered regions of the T Tauri stars. In a first step we derived here the magnetic field parameters of the T Tauri stars from roughly estimated model atmospheres, and the corresponding magnetic intensification of the spectral lines. Subtracting this intensification from the observed EWs, we get a correction to the initially observed mean excitation temperature of the included lines. This correction could be used to correct the effective temperature of the model atmospheres. However, since there is no known general relationship between excitation temperature of certain elements and effective temperature for stars of various spectral types, such a procedure would result in very long model computation times.

The results for [FORMULA] are fairly robust, even if we assume equal values for the effective temperature of both the magnetic and non-magnetic regions of the star ([FORMULA]). We can also separate B and f, finding filling factors [FORMULA]. These large filling factors indicate that there are very few field free regions on the star.

We showed that ignoring a magnetic field can, depending on the lines used, result in errors in effective temperature and underestimates of veiling.

The detection of fields of kG-strength in at least two of the classical T Tauri stars is encouraging. Together with the indirect evidence accumulated so far, it seems justifiable to conclude that strong magnetic fields are common in these stars. As pointed out in the introduction, the detection of fields of kG-strength is a a necessary but not sufficient condition for the magnetic accretion model. A coupling between disk and the star is required for the magnetic accretion model, implying the field has to be a low order multipole, which we could neither prove, nor disprove. Again, a low order multipole is a necessary but not a sufficient for the model to work, as pointed out by Safier (1998).

The detection of strong magnetic fields together with the perspective that X-rays of T Tauri stars are generated by stellar dynamo activity (Neuhäuser et al. 1995) might also be crucial for the understanding of the formation of planets from the circumstellar disk (Glassgold et al. 1997).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: December 16, 1998