Spatial and kinematic properties of the forbidden emission line region of T Tauri stars

We have carried out a long-slit spectroscopic survey for 38 T Tauri stars (TTSs) to study the spatial and kinematic properties of their forbidden emission line (FEL) regions. With these observations we hope to provide more insight into the complex physical structure of the outflows from young stars on the smallest spatial scales observable by long-slit spectroscopy. Due to the differential nature of the observational method, information on the spatial properties (offset from the stellar continuum and spatial width) on sub-arcsec (sub-seeing) scales can be obtained. For most TTSs the [OI] \lambda\lambda 6300, 6363, [NII] \lambda 6583 and [SII] \lambda\lambda 6716, 6731 lines have been investigated at a typical spatial resolution of 1.5'' and a velocity resolution of 20 - 50 km s^{-1}. A sub-sample of 9 closeby stars (Haro 6-10, XZ Tau, UZ Tau E, HN Tau, DO Tau, DP Tau, UY Aur, RW Aur and V536 Aql) has been extensively studied and the direction of their outflows has been approximately determined by taking spectra at several slit position angles, if not known from emission-line CCD imaging. The spatial and kinematic properties of the FEL regions of these 9 TTSs are described in detail. Together with 3 additional stars discussed in the literature, a sample of 12 stars provides the basis for the following main results of our survey: The so-called high-velocity component (HVC) of the FELs (or gas of high velocity which presumably represents in many cases a HVC being blended with emission of lower velocity) is generally spatially more extended than the so-called low-velocity component (LVC, or gas near the stellar velocity). In the [SII] \lambda 6731 line the centroid of the high-velocity gas is located typically at distances of 0.6'' from the TTS while for the low-velocity gas this value is smaller on average by more than a factor of 3. Comparing the spatial properties of the high-velocity gas among the investigated FELs, it turns out that the largest spatial width and the largest offset of the centroid from the star is usually observed in the [NII] \lambda 6583 line, while the emission region is most compact in the [OI] \lambda 6300 line. In [OI] the centroid of the high-velocity gas is typically offset by only 0.2'' from the star whereas in [SII] and [NII] 3 and 3.5 times larger average values have been measured, respectively. In the case of the low-velocity gas, the smallest offset of the emission centroid is also observed in [OI] (typically 0.1'' in [OI] and 0.2'' in [SII]). Our data provide additional support for the model of \cite[Kwan \& Tademaru (1988, 1995)]{ref31} according to which the HVC observed in the FELs of many TTSs is formed in a well-collimated jet, while the LVC represents gas from a physically distinct flow component (possibly a disk wind or a disk corona). The larger spatial extent of the high-velocity gas in [SII] and [NII] compared to that in [OI] is most probably the result of a jet decreasing in density with increasing distance from the source combined with an increase in excitation. The decrease of the electron density with distance is rather obvious for a jet with diverging stream lines, but why the electron temperature increases is unclear.