We present Herschel PACS mapping observations of the [O I] 63 μm line toward protostellar outflows in the L1448, NGC 1333-IRAS4, HH 46, BHR 71, and VLA 1623 star-forming regions. We detect emission spatially resolved along the outflow direction, which can be associated with a low-excitation atomic jet. In the L1448-C, HH 46 IRS, and BHR 71 IRS1 outflows this emission is kinematically resolved into blue-and redshifted jet lobes, having radial velocities up to 200 km s −1 . In the L1448-C atomic jet the velocity increases with the distance from the protostar, similarly to what is observed in the SiO jet associated with this source. This suggests that [O I] and molecular gas are kinematically connected and that thelatter could represent the colder cocoon of a jet at higher excitation. Mass flux rates (Ṁ jet (O I)) have been measured from the [O I] 63 μm luminosity adopting two independent methods. We find values in the range (1-4) × 10 −7  M yr −1 for all sources except HH 46, for which an order of magnitude higher value is estimated. Ṁ jet (O I) are compared with mass accretion rates (Ṁ acc ) onto the protostar and with Ṁ jet derived from ground-based CO observations. Ṁ jet (O I)/Ṁ acc ratios are in the range 0.05-0.5, similar to the values for more evolved sources. Ṁ jet (O I) in HH 46 IRS and IRAS4A are comparable to Ṁ jet (CO), while those of the remaining sources are significantly lower than the corresponding Ṁ jet (CO). We speculate that for these three sources most of the mass flux is carried out by a molecular jet, while the warm atomic gas does not significantly contribute to the dynamics of the system.

Context. Owing to the paucity of sub-arcsecond (sub)mm observations required to probe the innermost regions of newly forming protostars, several fundamental questions are still being debated, such as the existence and coevality of close multiple systems. Aims. We study the physical and chemical properties of the jets and protostellar sources in the NGC 1333-IRAS4A proto-binary system using continuum emission and molecular tracers of shocked gas. Methods. We observed NGC 1333-IRAS4A in the SiO(6−5), SO(6 5 −5 4 ), and CO(2−1) lines and the continuum emission at 1.3, 1.4, and 3 mm using the IRAM Plateau de Bure Interferometer in the framework of the CALYPSO large program. Results. We clearly disentangle for the first time the outflow emission from the two sources A1 and A2. The two protostellar jets have very different properties: the A1 jet is faster, has a short dynamical timescale ( 10 3 yr), and is associated with H 2 shocked emission, whereas the A2 jet, which dominates the large-scale emission, is associated with diffuse emission, bends, and emits at slower velocities. The observed bending of the A2 jet is consistent with the change of propagation direction observed at large scale and suggests jet precession on very short timescales (∼200−600 yr). In addition, a chemically rich spectrum with emission from several complex organic molecules (e.g. HCOOH, CH 3 OCHO, CH 3 OCH 3 ) is only detected towards A2. Finally, very high-velocity shocked emission (∼50 km s −1 ) is observed along the A1 jet. An LTE analysis shows that SiO, SO, and H 2 CO abundances in the gas phase are enhanced up to (3−4) × 10 −7 , (1.4−1.7) × 10 −6 , and (3−7.9) × 10 −7 , respectively. Conclusions. The intrinsic different properties of the jets and driving sources in NGC 1333-IRAS4A suggest different evolutionary stages for the two protostars, with A1 being younger than A2, in a very early stage of star formation previous to the hot-corino phase.

Aims. As part of the WISH (Water In Star-forming regions with Herschel) key project, systematic observations of H 2 O transitions in young outflows are being carried out with the aim of understanding the role of water in shock chemistry and its physical and dynamical properties. We report on the observations of several ortho-and para-H 2 O lines performed with the HIFI instrument toward two bright shock spots (R4 and B2) along the outflow driven by the L1448 low-mass proto-stellar system, located in the Perseus cloud. These data are used to identify the physical conditions giving rise to the H 2 O emission and to infer any dependence on velocity. Methods. We used a large velocity gradient (LVG) analysis to derive the main physical parameters of the emitting regions, namely n(H 2 ), T kin , N(H 2 O) and emitting-region size. We compared these with other main shock tracers, such as CO, SiO and H 2 and with shock models available in the literature. Results. These observations provide evidence that the observed water lines probe a warm (T kin ∼ 400−600 K) and very dense (n ∼ 10 6 −10 7 cm −3 ) gas that is not traced by other molecules, such as low-J CO and SiO, but is traced by mid-IR H 2 emission. In particular, H 2 O shows strong differences with SiO in the excitation conditions and in the line profiles in the two observed shocked positions, pointing to chemical variations across the various velocity regimes and chemical evolution in the different shock spots. Physical and kinematical differences can be seen at the two shocked positions. At the R4 position, two velocity components with different excitation can be distinguished, of which the component at higher velocity (R4-HV) is less extended and less dense than the low velocity component (R4-LV). H 2 O column densities of about 2 × 10 13 and 4 × 10 14 cm −2 were derived for the R4-LV and the R4-HV components, respectively. The conditions inferred for the B2 position are similar to those of the R4-HV component, with H 2 O column density in the range 10 14 −5 × 10 14 cm −2 , corresponding to H 2 O/H 2 abundances in the range 0.5−1 × 10 −5 . The observed line ratios and the derived physical conditions seem to be more consistent with excitation in a low-velocity J-type shock with strong compression rather than in a stationary C-shock, although none of these stationary models seems able to reproduce the whole characteristics of the observed emission.