Variability of the DG Tau Forbidden Emission Line Low Velocity Component

TL;DR

This work investigates the origin of the Low Velocity Component (LVC) in the optical forbidden emission lines of the young star DG Tau by tracking spectral and spatial changes over three epochs (~2003–2021) using kinematic fitting and spectro-astrometry. The authors identify up to six blue-shifted jet components and a red wing, confirm a jet slowdown of about $\approx 100\,\mathrm{km\,s^{-1}}$, and resolve three LVC sub-components (LVC-H, LVC-M, LVC-L) whose behavior is more stable in velocity than the jet but shows significant profile evolution. The LVC-M and LVC-L are interpreted as a disk wind and the upper disk atmosphere, respectively, with a minimum de-projected height $\geq 2$ au for LVC-M in [O I] 5577, favoring an MHD disk-wind origin although a photoevaporative wind cannot be ruled out. The results highlight the utility of combining kinematic fitting with spectro-astrometry to constrain wind origins in YSOs, while emphasising the need for higher spectral resolution and denser temporal sampling to resolve blending and potential jet–wind time lags.

Abstract

Optical Forbidden Emission Lines (FELs) come from transitions with long radiative decay times needing low density gas where collisions between atoms are rare. They are produced in the outflows driven by young stellar objects. These lines trace distinct velocity components, including a Low Velocity Component (LVC), which may be tracing a magneto hydrodynamic (MHD) or photoevaporative (PE) wind. We study the jet and LVC of the star DG Tau, whose jet velocity has decreased since 2006. We aim to investigate a link between the high velocity jet and the LVC and clarify the LVC origin as an MHD or PE wind by studying spectral \& spatial changes over time. Using kinematic fitting \& spectro-astrometry, we analyse three epochs of spectra spanning ~18 years. A ~100 km/s decrease in velocity from 2003 to 2021 aligns with known slowing of the jet. Fitting of the [O I] λ6300, [O I] λ5577, and [S II] λ6731 lines reveal complex FEL profiles, with up to six blue-shifted components and a redshifted wing, in agreement with Chou et al. (2025). We see three LVC sub-components (LVC-H, LVC-M, and LVC-L) that are consistent with entrained jet material, disk wind, and dense upper disk atmosphere respectively. While jet components vary in time, the LVC remains quite stable, with changes in the relative brightness of each sub-component. The results cannot distinguish between a MHD or PE wind origin for the LVC. A limit of less than 2 au is put on the de projected height of the LVC-M in [O I] λ5577, where there is no jet contribution. This supports a disk wind and may favor an MHD wind origin. The near constant peak velocity of LVC-M needs further study in context of a shared origin for jets and MHD winds. Future work needs observations with higher spectral resolution and time cadence to resolve blended components and examine a possible time lag between changes in the jet and LVC.

Figures (4)

• Figure 1: Position velocity diagrams of the [O I] $\lambda$5577 (left) [O I] $\lambda$6300 (middle) and [S II] $\lambda$6731 (right) FELs in 2003, 2010 and 2021 from DG Tau. Contours start at $3\sigma$ (2003) and $7\sigma$ (2010, 2021) above the background rms noise and increase by factors of $\sqrt{2}$ (2003) and $\sqrt{3}$ (2010, 2021). The kinematic components identified by Chou2024 in the 2003 data (HVC1, HVC2, LVC) are marked. The assumed distance to DG Tau is 125 pc (see Table \ref{['properties_table']}).
• Figure 2: Kinematic fitting comparison of the [O I] $\lambda$5577, [O I] $\lambda$6300 and [S II] $\lambda$6731 line profiles via Gaussian decomposition in 2003, 2010 and 2021. The line profiles are shown in the stellocentric frame and have been normalised to the continuum level. These profiles are extracted over the full spatial range of the PV diagrams shown in Fig.\ref{['DGTau_pvplots']}.
• Figure 3: Top row: Comparison of the line profiles of [O I] $\lambda$6300 and [S II] $\lambda$6731 for each epoch of data. The line profiles have been normalised to max peak height to emphasise changes in the line shape over time. Bottom row: Position spectra for the [O I] $\lambda$6300 and [S II] $\lambda$6731 lines in 2003 (red), 2010 (blue) and 2021 (black). The analysis of the [O I] $\lambda$5577 line is not presented here as it does not show an extended jet component due to its high critical density.
• Figure 4: Top row: Comparison of the line profiles in the LVC region of each FEL in each epoch of data. 2003 = red, 2010 = blue and 2021 = black. These panels show a zoom of the line profiles shown in Fig. \ref{['DGTau_SA_HVC']} from -90 to 50 kms$^{-1}$. This is done to emphasise the change of line shape in the LVC region over time. The line profiles have been normalised to max peak height. Bottom row: The position spectra of the LVC region for each FEL in each epoch of observation. 2003 = red, 2010 = blue and 2021 = black. The 1-$\sigma$ uncertainties range from 1 to 5 au where 5 au is measured for the lowest signal to noise emission i.e. [O I] $\lambda$5577 in 2021.