Astron. Astrophys. 330, 327-335 (1998)

3. Results

3.1. Velocity structure

In Figs. 2, 3 and 4, we present examples of our data, clearly showing the quality of the observations.

Fig. 2. Position velocity plots of the H  spectra. They are: Top row - slits 1, 2 and 3; Middle row - slits 4, 5 and 6; Bottom row - slits 7, 8 and 8 shown at a different scale. We refer to Fig. 1 for the slit orientation. The velocity scale is barycentric. East is towards the bottom in the top row, while South is down in the rest of the panels. The lowest contour is 5 times the rms noise and each successive step is a factor 2 . The last panel displays the spectrum obtained through slit 8 on an expanded velocity scale. The upper left panel (slit 1) has the position marked where we have determined the maximum level of a possible atmospheric H  contribution

Fig. 3. The full H  profiles from the a component of HH 29. The top row is from slit 2 followed in succession downwards by slits 1 and 3 (see Fig. 1). The scale in the EW direction is in arc seconds and indicated on the top. The resolution is thus 1 2 (corresponding to 140AU  280AU). IRS5 is located 2 3 towards the upper left (NE). The sudden onset of the shock is obvious, as is the radical change in the profiles over small spatial elements

Fig. 4. Position velocity plots of the [SII] 6717Å (left) & 6731Å (right) spectra obtained through slit 1. The contours orientation and scales etc as in each panel of Fig. 2.

The lines are broad and show a very complex structure both with respect to position and velocity. Referring now to Fig. 1  and Fig. 2, both of which displays H  emission, attention is immediately drawn to a few features. In slits 1, 2 & 3 which pass through HH 29a, and immediately N and S of this knot, we see that the intensity gradient is steeper in the direction of IRS5, thus suggesting that the shock is located on this side. We also see, eastward of HH 29 `proper' (compare with Fig. 1), two fainter and well separated components. They have velocities of  -130 km s^-1 and -15 km s^-1 (barycentric). It is conceivable that this emission could be due to recombination from a precursor to the shock, such as has been detected in HH34 by Heathcote & Reipurth (1992). The lower velocity component is found essentially at rest with respect

to the telescope (and the molecular cloud L1551 within which HH 29 is located, V_LSR  = +7 km s^-1). Although usually not present, atmospheric H  emission has been observed previously (L. Pasquini, private communication). It can thus not be excluded that part of this feature is caused by emission in the atmosphere of the Earth. Its structure can be discerned from Fig. 2 (slits 1, 2 & 3). The FWHM of this feature is, however, 1 pixel (  4.5 km s^-1) wider than the atmospheric lines of equivalent brightness found within the same and adjacent orders of the Echelle spectrogram. The brightness of the low velocity feature, also varies as a function of position along the slit. If we also refer to the last two panels in Fig. 2 (both corresponding to slit 8), it can be seen that the velocity of what we above called the -15 km s^-1  component

varies with an amplitude of several tens of km s^-1  along the N-S direction of this slit. Taken together, all this indicates that at least some - if not most or all - of the emission is intrinsic to the region surrounding HH 29, although we can not exclude that part of the emission is of telluric origin. For the purposes of this paper we have taken all emission above the faintest level detected in this feature to be intrinsic. In Fig. 2 we have marked at what position in the structure this level is to be taken.

With regard to the component found at -130 km s^-1  , comparison with Fig. 1  show that slits 1, 2 & 3 will admit emission from the feature HH 29f (compare slit 8 - two last panels of Fig. 2) at the spatial coordinates in question. We also see that the amount of H  emission that is actually observed in the various slits is consistent with the faint high velocity component (the -130 km s^-1  component) is being dominated by emission from HH 29f.

The full line width at zero intensity, FWZI_S, corrected for the instrumental profile can be calculated according to , which corresponds to the analytical solution of the deconvolution integral for Gaussian line profiles. Since a simple inspection of our data tells us that the profiles are not Gaussian, we have investigated how great a discrepancy is introduced if a simple measurement of at what velocity the line intensity goes to zero is performed instead. The discrepancy is found to be  a few % and consequently we will use the FWZI_M in the following.

In Table 1  we have listed the FWZI_M of the detected lines at the position of HH 29a, and in Table 2 we list the same quantity for H  for the other knots identified in HH 29.

Table 1. Ions observed in HH 29a

Table 2. FWZI of each knot in H . Ratios of high excitation and low excitation lines to Balmer lines

We can use our high spectral resolution data in order to compare, qualitatively, with the models of HRH. In Fig. 3, we have displayed the full line profiles of the H  line, over an area 7  by 6, centered on the (0,0) coordinate in Figs. 1, i.e. on the intensity maximum of the HH 29a component. Inspection of Fig. 3 and comparison with the Fig. 1 and the figures in FLP, show immediately the steep gradients in the profiles associated with features a & d respectively on opposite sides of the object. The line profiles of these two features, located 2  or less than 300AU apart, have the same FWZI_M, while the peak emission is separated by  90 km s^-1.

3.2. Line fluxes and line ratios

Following Raga et al.  (1996), we have formed the line ratio [NII]6584Å/H  for tracing high excitation conditions, and the ratios [OI]6300Å/H , [SII] /H  and [CaII]7291Å/H  - all of which trace zones of low level excitation. Further, we have taken the [OII] /H  and the [OIII]5007Å/H  ratios from the results of FLP as indicators of levels of high excitation. The latter ratio is a tracer of very high excitation. Note that all these ratios are for practical purposes to be considered reddening independent. The effect of a standard reddening law on the [CaII]7291Å/H  ratio (the most affected) is 20%, which for our purposes is unimportant. The result is shown in Table 2, where we present the ratios for the features HH 29a, b, c, d, e and f.

In Table 3 we present the integrated flux of the H  line within 2  bins along the slits. In Table 4  we then display the [SII]/H  as well as the 6717Å/6731Å ratios. Here the data is thus presented in 2  bins over all of HH 29 and not only for the distinct features as in Table 1 and Table 2. The former of these ratios is mapping out the level of excitation over the projected surface of HH 29, while the latter ratio is a good measure of the electron density, n_e.

Table 3. Observed H  line fluxes. Units of 10^-15  erg cm^-2 s^-1

Table 4. The ratios of the sum of the 6717,6731Å lines to the H  line (designated R₁) as well as the 6717/6731 ratio (designated R₂)

Since we are aiming for a three dimensional representation of HH 29, we have calculated the flux in H  and both of the [SII] lines as a function of velocity, in 20 km s^-1  bins, and also spatially in steps of 1  along the 2  wide slit. Selected parts of this data is presented in Table form and interpreted in Sect. 5.

© European Southern Observatory (ESO) 1998

Online publication: January 8, 1998
helpdesk.link@springer.de