Astron. Astrophys. 364, 763-768 (2000)

2. The parameters for the Serpens jet

Very Large Array (VLA) 3.6 cm radio continuum maps of the NW lobe of the Serpens "triple source" jet obtained at three different epochs are shown in Fig. 1. These radio continuum observations were carried out with the 27 antennae of the Very Large Array (VLA) of the National Radio Astronomy Observatory (NRAO) ¹ in its A-configuration at 3.6 cm in 1990 (June 1), 1995 (July 13), and 1998 (May 9). These data are part of a long term monitoring program in progress of this remarkable radio continuum jet, and the technical details of the observations will be given in Curiel et al. (2000). The data were edited and calibrated following standard VLA procedures with AIPS. The maps of the three different epochs were made by weighting the () data with the Briggs (1995) "robustness" parameter set to 4 to enhance weak knots and faint extended emission along the radio jet, which yielded a synthesized beam of about and an rms noise of about 17 µJy, with small variations at each epoch. The maps were rotated 48.55 degrees counterclockwise in order to align the NW radio lobe with the vertical axis.

Fig. 1. 3.6 cm VLA radio continuum maps of the NW lobe of the Serpens radio jet obtained at three different epochs. The ordinate is aligned with a NW direction (PA). The proper motions and intensity variations of the knots along the jet can be clearly seen. The axes are labeled in arcesconds, with the zero point corresponding to the position of the outflow source. The identification of the radio knots was taken from Curiel et al. (1993).

This system actually has two lobes (the southern lobe not being shown in Fig. 1), both having proper motion velocities of  km s^-1 away from the central source (Curiel et al. 1993). As there are no radial velocity measurements for this purely continuum jet, we are forced to assume that the outflow axis lies on the plane of the sky, and that the measured proper motions therefore indicate that the jet has a velocity of  km s^-1.

The NW lobe shows 5 well defined knots (see Fig. 1), with separations of , corresponding to a projected distance of  cm (at a distance of 300 pc for the Serpens star formation region). If we want to model these knots as the result of an ejection velocity time-variability, the period of this variability has to have a value of  yr. Similar periods have been determined for the moving knots in HH 80-81 (Martí et al. 1995) and HH 1-2 (Reipurth et al. 2000).

The amplitude of the time-variability can be constrained as follows. From the analytic theory for the formation of working surfaces, one obtains that for an ejection velocity variability of the form:

the working surfaces first appear at a distance

from the source. This simple expression is strictly valid for , with the exact value for being given by Raga & Noriega-Crespo (1998).

Now, the first well defined knot in the NW lobe of the Serpens jet (see Fig. 1) is at a distance of from the source. However, there is an extension of the source emission of to the NW, which appears to represent the next knot forming along this outflow. This implies that the knots are formed at a distance of  cm from the source. If we take this value as an estimate for , we can use our estimates of  km s^-1 and  yr to derive an amplitude of  km s^-1.

In this way, we have selected parameters for a sinusoidal ejection velocity variability (Eq. 1), which are consistent with the knot spacings and proper motions observed for the NW lobe of the Serpens jet. The most uncertain parameter is the amplitude of the variability, as this amplitude depends on the distance from the source at which the knots first appear, which cannot be reliably determined from the observations.

In the 3.6 cm maps (Fig. 1) it is clearly seen that the knots in the NW lobe appear to trace a precession cone. The full opening angle of this cone has a value of (Curiel et al. 1993, 1996). From the morphology of the jet, one sees that the precession period has to be similar to the dynamical timescale of the leading condensation (see Fig. 1). Therefore, the precession period has a value  yr.

It is difficult to estimate the density of the jet and the environment directly downstream of the leading condensations of the two lobes of the Serpens outflow. Given the fact that these condensations have high proper motions, similar to those of the inner condensations (see Curiel et al. 1993), one would conclude that either the environment has a rather low density compared to the jet (for example, the jet could be moving into a cavity which was partially evacuated by previous outflow episodes), or that the jet is moving into (undetected) moving material which was previously ejected from the source. Molecular observations of the dense core indicate that the ambient gas density is probably several times  cm^-3, or even higher than  cm^-3 (e,g., Curiel et al. 1996; McMullin et al. 1994). Assuming that the radio emission comes from  km s^-1 shocks inside the jet beam, Curiel et al. (1993) estimate a jet density - cm^-3. To summarize, it appears that the radio continuum jet is moving into a low density cavity, or into previously ejected material, and that the jet density is in the - cm^-3 range.

Finally, the knots along the NW lobe of the Serpens outflow are at most marginally resolved across the outflow axis. This leads to an estimate of  cm for the jet radius.

© European Southern Observatory (ESO) 2000

Online publication: January 29, 2001
helpdesk.link@springer.de