JWST study of the DG Tau B disk-wind candidate

Context. The origin of outflows and their impact on protoplanetary disk evolution and planet-formation processes are still crucial open questions. DG Tau B is a Class I protostar associated with a structured disk and a rotating conical CO outflow and was recently identified by ALMA as one of the best CO disk wind candidate. It is therefore a perfect target for studying these questions. Aims. We aim to map and study any outflow component intermediate between the axial jet and the CO outflow in order to constrain the origin (irradiated or shocked disk wind or swept-up material) of the redshifted molecular outflows in DG Tau B. Methods. We analyzed observations obtained with James Webb Space Telescope NIRSpec-IFU and NIRCam, supplemented with IFU data from the SINFONI/VLT instrument. We investigated the morphology, kinematics, and excitation conditions of the ro-vibrational H 2 emission lines and their relation with the atomic jet and CO outflow. We focus our analysis on the redshifted outflow lobe. Results. We observe a global layered structure of the redshifted outflows in DG Tau B, with the atomic jet inside the H 2 cavity, which in turn is nested inside the CO conical outflow. We also find temperature, velocity, and collimation to be increasing towards the axis of the flow. The redshifted H 2 emission traces a narrow conical cavity (semi-opening angle of 9.4°) that is wider than the axial jet but nested just inside the CO outflow. Both the jet and the H 2 cavity originate from the innermost regions of the disk ( r 0 < 6 au). The redshifted H 2 cavity flows with a constant vertical velocity of V z = 22.5 ± 0.8 km s −1 , twice faster than the conical CO flow. The excitation conditions imply a hot H 2 gas ( T ex ≃ 2200 K) with an average mass flux of Ṁ (H 2 ) = 3 × 10 −11 M ⊙ yr −1 , which is significantly lower than the jet and CO values. Conclusions. The global layered H 2 -CO structure in temperature, velocity, and collimation in the DG Tau B redshifted lobe is consistent with a magneto-hydrodynamic disk wind scenario. The hot H 2 could trace the inner, dense photodissociation layer in the wind. An H 2 launching region at disk radii of 0.2−0.4 au combined with a large ejection efficiency ( ξ ≃ 1) would account for the mass flux and kinematics. Alternatively, the near-IR ro-vibrational H 2 could be emitted in the interaction layer driven by successive jet bow shocks into an outer disk wind or envelope. Further constraints on both scenarios will be obtained from the analysis of MIRI observations.