Evidence of a Disk-wind Origin for Fluorescent H${}_2$ in Classical T Tauri Stars

Abstract

We use FUV spectra of 36 T Tauri stars, predominately from $\text{Hubble Space Telescope}$'s ULLYSES program, to examine the kinematic properties of fluorescent H${}_2$ emission lines for evidence of disk outflows. Leveraging improvements to the $\text{HST}$-COS wavelength solution, we co-add isolated lines within four fluorescent progressions ([$\text{v’}$,$\text{J’}$] = [1,4], [1,7], [0,2], and [3,16]) to improve signal-to-noise, and we fit each co-added line profile with one or two Gaussian components. Of the high S/N line profiles (S/N $\ge$ 12 at the peak of the profile), over half are best fit with a combination of a broad and narrow Gaussian component. For profiles of the [1,4] and [1,7] progressions, we find a systematic blue-shift of a few km s${}^{-1}$ between the broad and narrow centroid velocities and stellar radial velocities. For the [0,2] progression, we find centroid velocities consistently blueshifted with respect to stellar radial velocities on the order of -5 km s${}^{-1}$ for the single and narrow components, and -10 km s${}^{-1}$ for the broad components. Overall, the blueshifts observed in our sample suggest that the molecular gas traces an outflow from a disk wind in some sources, and not solely disk gas in Keplerian rotation. The low-velocity systematic blue-shifts, and emitting radii as inferred from line FWHMs, observed in our sample are similar to those observed with optical [O I] surveys of T Tauri stars. We estimate H${}_2$ mass-loss rates of 10${}^{-9}$ to 10${}^{-11}$ $M_{\odot }$ yr${}^{-1}$, but incomplete knowledge of wind parameters limits comparisons to global models.

30 pages, 15 figures, accepted to $\text{AJ}$