SIMBAD references

It is important to determine whether the observed bow shocks in the working surfaces of Herbig-Haro (HH) outflows have lead to a destruction of dust grains and consequently to a change in the gas-phase metal abundances (say, of Fe) in the cooling regions of HH bow shocks. Detailed studies are currently available for only five HH outflows. This small number is due to the large observational and theoretical effort required to determine metal abundances in HH objects.

 Information about metal abundances in more HH objects is badly needed. We therefore use a very approximate method. We introduce a ``characteristic number'', A_me, whose definition is based only on the often observed line fluxes of [Fe II] λ5159, [Fe II] λ7155, [Ca II] λ7291, as well as Hα and Hβ. These fluxes can easily be determined from existing observations. We find a good correlation between A_me and the Fe abundance for the five well-studied HH objects. We use this correlation to determine approximate values of the gas-phase Fe abundance in 13 additional high-excitation and three additional low-excitation HH objects.

 The results are the following: Of the 16 high-excitation HH objects studied, there are six which approximately agree with normal Population I abundance (i.e., no depletion due to dust formation). The remaining 10 show some (very modest) Fe gas-phase depletion which, however, never gets larger than by a factor of 2.5. This result is in agreement with our qualitative expectations that fast shocks efficiently destroy dust grains. Of the five low-excitation HH objects studied, there are four which show a normal Population I abundance (strictly speaking, an even slightly higher abundance than this). This is completely unexpected. In low-excitation objects, one might expect strong gas-phase Fe depletion (showing the unchanged molecular cloud composition), unless the matter has previously gone through shocks of much higher shock velocities. We discuss this possible explanation and the question of whether low-excitation HH objects have a different ``history'' than usually assumed.