Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 321, 497-512 (1997)

Previous Section Next Section Title Page Table of Contents

7. Conclusions

We can summarize the results of our phenomenological study of the spectrum variations in these highly-veiled cTTS as follows:

  • We derive a spectral class of K6V for DG Tau from with lower-resolution data which is independent of the excitation potentials of the weak photospheric lines. The high-resolution data suggest a spectral type of G8IV for DI Cep.
  • The veiling appears to be caused by an additional, effectively featureless continuum source rather than by the filling in of photospheric absorption lines by faint emission features. Its strength is weakly anti-correlated with EW (H [FORMULA]) in DR Tau and DG Tau as one would naively expect.
  • The variations in the equivalent widths of a wide range of emission lines during each night are almost solely due to line flux rather than veiling continuum variations. This is true for a strongly veiled system like DG Tau as well as for a weakly veiled TTS like DI Cep. The veiling continuum is variable in all three stars, but apparently on a timescale which is roughly an order-of-magnitude longer.
  • The variations of the equivalent widths of the Ca II [FORMULA] 8662 and He I lines are correlated with those of H [FORMULA] in all three stars (additionally, the variations of the Ca II H line is well correlated with H [FORMULA] in DG Tau and DR Tau), despite the fact that these lines have very different excitation potentials.
  • All emission lines in DG Tau and DR Tau show the same quasi-periodic variations - which may or may not be the same as the rotation period of the stars - and no phase-shift greater than about [FORMULA] can be seen between lines of different excitation potentials. These variations do not show up the in veiling continuum, undoubtably causing the weakness of the veiling-EW correlations. However, the suggested quasi-period for DG Tau is much shorter than the published photometric period.
  • The variations of the line profiles in the few high-resolution spectra obtained are predominantly in the blue wings of the H [FORMULA] and H [FORMULA] profiles in DR Tau, and in the red wing in DI Cep.
  • The short-timescale variations are too short to be caused by quasi-spherical wind variations, suggesting that the observed line variations on all timescales are cause either by "flare" activity of unknown origin or by rotational modulation. In either case,
  • The regions responsible for the H [FORMULA] line emission must be much larger than the central star by factors greater than 2-16 in area and are thus probably distinct from that which produces the veiling continuum. This situation explains why the short timescale variations in the lines are not seen in the veiling continuum.

Given these constraints, we are forced to conclude that the short timescale variations in the extended emission line region is due to some sort of "flare" activity in the disk or associated with the "hole" at the disk-stellar magnetic field interface. Though we have neither a theory for magnetospheric disk accretion which is detailed enough to be compared with such extensive observations nor a "smoking gun" which can definitively distinguish between this and a relatively complicated but possible disk plus stellar wind scenario (e.g. magnetic activity of the accretion disk, or asymmetric chromospheric disk emission excited by weak and/or high-order magnetic fields on the star), the burden of proof has clearly shifted to disk models for classical T Tauri stars.

Both the integration of the empirical information into detailed models for the accretion regions (Guenther & Hessman 1996) and further long-term studies of this type with simultaneous BVRI photometry and high-resolution spectroscopy are needed.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998