4 The outflow activity of S140 IRS1

The combination of our H₂ data with previous observational results, allows us to draw a comprehensive picture of the outflow activity of S140 IRS1 on a wide range of spatial and temporal scales:

• Our high-resolution speckle interferometric observations provide information about the detailed morphology of the circumstellar material in the immediate vicinity of IRS1, on spatial scales from $\sim$100 AU to $\sim$10000 AU.
• The high-resolution radio continuum observations provide information about the ionized stellar winds and jets close to the central source.
• The H₂ line emission images show those regions where the outflowing material from the protostars impacts with the dense environment. Because of the relatively short radiative cooling time (on the order of a few years) of the H₂ $\nu=1{-}0$ S(1) transition, the molecular hydrogen emission illustrates the current shock-activity in the outflows.
• The radio molecular line emission observations provide information about the large scale structure of the outflows on spatial scales of $\sim$50000 AU and the long-term effects of the outflows on the ambient material

  4.1 The 160$^\circ$/340$^\circ$ outflow

  The presence of the large scale $\bf 160\hbox{$^\circ$ }/340\hbox{$^\circ$ }$ outflow was already well established from radio line observations. In the radio maps, outflowing gas can be seen at least up to distances of 1' (54000 AU) from IRS1 (e.g. Minchin et al. 1993). Our high-resolution speckle observations (Schertl et al. 2000; Weigelt et al. 2002) revealed the detailed structure of the cavity through which the material in the $160\hbox{$^\circ$ }$ outflow lobe is flowing on scales of 90-1800 AU. Our molecular hydrogen emission line images show only two relatively weak H₂ knots associated with this outflow, knots 3 and 9. This suggests that there are currently not many strong shocks in this outflow, perhaps because the path of the outflow is already well cleared and the material is flowing away without producing strong shocks. However, the detection of H₂ shock emission in this outflow is aggravated due to the inclination of the flow and projection effects: In the south-eastern part, where the outflow is directed towards us, H₂ emission very close to ( 2'' away from) IRS1 would be projected onto the very bright emission of the cavity. In the north-western part of this outflow, H₂ shock emission might be present, but hidden behind the dense circumstellar material around IRS1. The fact that the high resolution radio continuum observations presented by Schwartz (1989) and Hoare & Muxlow (1996) do not show any sign of the $160\hbox{$^\circ$ }/340\hbox{$^\circ$ }$ outflow, supports our assumption that this outflow is probably currently not very active.

  A rough estimate of the kinematic age of this outflow can also be made: from the projected distance of knot 9, an assumed velocity of $\sim$100 kms^-1, and assuming an inclination angle of , we find a lower limit of years.

  4.2 The 20$^\circ$/200$^\circ$ outflow

  The existence of a $\bf 20\hbox{$^\circ$ }/200\hbox{$^\circ$ }$ outflow has not so far been well established. Radio line observations had shown a relatively weak low-velocity wing in the CO and HCO⁺ emission that indicated outflowing material in the $20\hbox{$^\circ$ }$ direction, but no reliable evidence for an outflow in the $200\hbox{$^\circ$ }$ direction. The high-resolution radio continuum maps of S140 IRS1 presented by Schwartz (1989) revealed a strongly elongated jet-like appendage extending from IRS1 in the south-west direction (position angle $225\hbox{$^\circ$ }$), and a diffuse radio source (VLA 4) south-west of IRS1 (position angle $217\hbox{$^\circ$ }$) which may be a radio Herbig-Haro object. These features were interpreted as evidence for ejection of material from IRS1 in the south-western direction. The high-resolution radio observations by Hoare & Muxlow (1996) showed a strongly elongated structure oriented in the $44\hbox{$^\circ$ }/224\hbox{$^\circ$ }$ direction. It was interpreted as a highly collimated jet, probably being driven by a disk wind. The 8.4 GHz VLA observations of S140 presented by Tofani et al. (1995) revealed an interesting S-shape for the emission within 1'' of IRS1. While the core of the emission is oriented in the $44\hbox{$^\circ$ }/224\hbox{$^\circ$ }$ direction, the position angle of the north-eastern tip of the 8.4 GHz emission is $20\hbox{$^\circ$ }$, in very good agreement with the position angle of the outflow cavities we found in the speckle images.

  Our high-resolution speckle observations (Weigelt et al. 2002) revealed three arc-like structures with position angles of $10\hbox{$^\circ$ }$, $20\hbox{$^\circ$ }$, and $25\hbox{$^\circ$ }$ north-east of IRS1. The shapes of these arcs could be well reproduced by a semi-analytical model of jet-driven flows, in which prompt entrainment occurs at the head of the traveling bow shock. This strongly suggested a flow of material in this direction, but could not be considered as a clear proof. The result that the eastern edge of the supposed cavities is actually associated with strong H₂ emission, confirms our previous interpretation that these cavities have been created by outflowing material from IRS1 in the $\sim$20$^\circ$ direction.

  Our H₂ line emission image reveals numerous knots along the paths of the $\bf 20\hbox{$^\circ$ }/200\hbox{$^\circ$ }$ outflow: knots 1, 2, 10, 11, and 12 are related to the $20\hbox{$^\circ$ }$ flow, knots 4 and 5 to the $200\hbox{$^\circ$ }$ flow. These knots span a range of position angles from $20\hbox{$^\circ$ }$ to $45\hbox{$^\circ$ }$, which is consistent with the orientation of the elongated radio emission immediately around IRS1.

  The numerous H₂ shocks in these directions indicate that the outflow currently interacts very strongly with its surroundings. Perhaps the outflow is currently in the process of clearing its way through the ambient material. The rather wide range of position angles of the individual H₂ knots north-east of IRS1 suggest a pulsed nature for the outflow and directional variability. A wiggling jet in the NE/SW direction is also suggested by the radio continuum and maser emission on the scale of $0.5\hbox{$^{\prime\prime}$ }$ (Tofani et al. 1995), which extends in the 45 $\hbox{$^\circ$ }$ direction i.e. towards H₂ knot 2. The temporal variability of the outflow direction may be caused by the precession of the circumstellar disk around the outflow source which is a member of a non-coplanar binary system (see Weigelt et al. 2002 for details; for the theoretical background of this model see e.g. Papaloizou & Terquem 1995; Terquem 1998; Bate et al. 2000).

  A rough estimate of the kinematic age of this outflow, based on the distance of knot 10, an assumed inclination angle of $i \sim 0\hbox{$^\circ$ }$ and a velocity $\sim$100 kms^-1, yields $\tau \sim 2300$ years for the $\bf 20\hbox{$^\circ$ }/200\hbox{$^\circ$ }$ outflow. This is nearly three times younger than the lower limit to the kinematic age of the $\bf 160\hbox{$^\circ$ }/340\hbox{$^\circ$ }$ outflow.