The ISO-LWS map of the Serpens cloud core^*

II. The line spectra

¹ Stockholm Observatory, SCFAB, Roslagstullsbacken 21, 106 91 Stockholm, Sweden e-mail: bem@astro.su.se; rene@astro.su.se
² Max-Planck-Institut für Radioastronomie, Auf dem Hügel, Bonn, Germany e-mail: sasha@mpifr-bonn.mpg.de

Corresponding author: B. Larsson, bem@astro.su.se

Received: 30 November 2001
Accepted: 29 January 2002

Abstract

We present spectrophotometric ISO imaging with the LWS and the CAM-CVF of the Serpens molecular cloud core. The LWS map is centred on the far infrared and submillimetre source FIRS 1/SMM 1 and its size is 8′   8′. The fine structure line emission in [O i] 63 μm and [C ii] 157 μm is extended on the arcminute scale and can be successfully modelled to originate in a PDR with and n(H₂) in the range of  cm^-3. Extended emission might also be observed in the rotational line emission of H₂O and high-J CO. However, lack of sufficient angular resolution prevents us from excluding the possibility that the emssion regions of these lines are point like, which could be linked to the embedded objects SMM 9/S 68 and SMM 4. Toward the Class 0 source SMM 1, the LWS observations reveal, in addition to fine structure line emission, a rich spectrum of molecular lines, superposed onto a strong, optically thick dust continuum (Larsson et al. [CITE]). The sub-thermally excited and optically thick CO, H₂O and OH lines are tracing an about 10³ AU source with temperatures higher than 300 K and densities above 10⁶ cm^-3 (  ). The molecular abundances, (H₂), are for CO, H₂O, OH and ¹³CO, respectively. Our data are consistent with an ortho-to-para ratio of 3 for H₂O. OH appears highly overabundant, which we tentatively ascribe to an enhanced (X-ray) ionisation rate in the  (). We show that geometry is of concern for the correct interpretation of the data and based on 2D-radiative transfer modelling of the disk/torus around SMM 1, which successfully reproduces the entire observed SED and the observed line profiles of low-to-mid-J CO isotopomers, we can exclude the disk to be the source of the LWS-molecular line emission. The same conclusion applies to models of dynamical collapse (“inside-out” infall). The 6 pixel resolution of the CAM-CVF permits us to see that the region of rotational H₂ emission is offset from SMM 1 by 30, at position angle 340°, which is along the known jet flow from the Class 0 object. This H₂ gas is extinguished by A_V  = 4.5 mag and at a temperature of 10³ K, which suggests that the heating of the gas is achieved through relatively slow shocks. Although we are not able to establish any firm conclusion regarding the detailed nature of the shock waves, our observations of the molecular line emission from SMM 1 are to a limited extent explainable in terms of an admixture of J-shocks and of C-shocks, the latter with speeds of about (15–20) km s^-1, whereas dynamical infall is not directly revealed by our data.