The First Distance Constraint on the Renegade High Velocity Cloud Complex WD

J. E. G. Peek, Rongmon Bordoloi, Hugues Sana, Julia Roman-Duval, Jason Tumlinson, Yong Zheng


We present medium-resolution, near-ultraviolet VLT/FLAMES observations of the star USNO-A0600-15865535. We adapt a standard method of stellar typing to our measurement of the shape of the Balmer \(\epsilon\) absorption line to demonstrate that USNO-A0600-15865535 is a blue horizontal branch star, residing in the lower stellar halo at a distance of 4.4 kpc from the Sun. We measure the H & K lines of singly-ionized calcium and find two isolated velocity components, one originating in the disk, and one associated with high-velocity cloud complex WD. This detection demonstrated that complex WD is closer than 4.4 kpc and is the first distance constraint on the +100 km/s Galactic complex of clouds. We find that Complex WD is not in corotation with the Galactic disk as has been assumed for decades. We examine a number of scenarios, and find that the most likely is that Complex WD was ejected from the solar neighborhood and is only a few kpc from the Sun.


High-velocity clouds (HVCs) provide a unique window into the coolest component of the circumgalactic medium and the processes of Galactic inflow and outflow. HVCs, and the complexes into which they are arranged, are found by their emission in HI or absorption in numerous metal lines, and have radial velocities inconsistent with Galactic rotation (Wakker 1997). The precise origin of most HVCs is unknown, and some mix of Galactic fountain (e.g. Bregman, 1980), multiphase accretion (e.g. Fernández et al., 2012), and gas stripping from satellites is typically invoked (Putman 2012). The exception is the Magellanic stream, which was stripped from the large and small Magellenic clouds, and which we will exclude from our discussion in this work. HVCs with negative radial velocities, which are metal enriched in the range of 10% to 30% of the solar metallicity, are likely a tracer of the process by which material accretes onto the Galaxy, though the total rate of this accretion is very uncertain. Less explored are the HVCs with positive radial velocities, most of which are in the inner two quadrants of the Galactic sky. These include the Wannier complexes WA, WB, WD, WE, and the Smith Cloud (Wakker 1991). The Smith cloud has received significant attention of late, for its strongly cometary appearance which provides enough information to infer past trajectories, and make some inference as to its origin (Lockman 2008, Fox 2015).

Complex WD is the largest-area positive velocity HVC Complex, covering 310 square degrees with a total HI flux of 1.2 \(\times 10^7\) K km/s arcmin\(^2\), and a maximum HI column density of \(\sim 1.2 \times 10^{20}\) cm\(^{-2}\). It is by far the largest complex that exists in the inner two Galactic quadrants, where a small fraction of HVC flux is detected. With a range of velocities between +90 and +130 km/s, it is consistent with cylindrical rotation on the far side of the inner Galaxy, 20 kpc from the sun with a mass of 6 \(\times 10^7 M_\odot\). This would make it very similar in mass, Galactocentric radius, and height to Complex C, the largest area and brightest HVC complex (Thom 2008). One major issue in gaining a better physical understanding of these enigmatic clouds is their unknown distance. Since there are no objects of fixed luminosity in HVCs, there are effectively no intrinsic distance measures. HI emission or optical and UV absorption lines toward extragalactic background sources only provide distance-independent column densities. HVC distances would not only give us a masses for these structures, but also a context; the spatial relationship between the cloud and the nearby spatial and kinematic structure of the disk gives us insight as to its origin.

There are a number of indirect methods for measuring the distance to an HVC complex, including H\(\alpha\) emission and kinematic structure (Putman 2003, Peek 2007), but the only proven direct distance measure is stellar absorption. By observing stars with measured distances at medium or high spectral resolution, one can look for absorption lines in Na I, Ca II H & K, Ti II, and numerous ultraviolet absorption lines at the velocity of HI emission from HVCs (Schwarz 1995). By finding detections and non-detections of these absorption lines along lines of sight toward HI emitting HVCs, distances can be robustly measured. A number of clouds have well-measured distances using this method, but complex WD is not among them (Wakker 2001, Wakker 2007, Thom 2008, Wakker 2008).

In this work we report the first distance upper limit on Complex WD using medium resolution absorption line spectroscopy toward a blue horizontal branch star. We extend the methods of Sirko et al. (2004) to find the spectral type of a blue horizontal branch star, and thus put a precise distance limit. We use this to make some inferences as to the possible origin of Complex WD, and how it fits into the structure of Galactic HVCs as a whole.


New Observations

The observations of our target, USNO-A0600-15865535, were obtained at the ESO Very Large Telescope (VLT) at Cerro Paranal, Chile on the nights of May 6th and May 7th, 2016. The target was observed as a part of our program “Mapping the Cool Circumgalactic Medium with Calcium II” (097.A-0552, PI: Peek) that uses the FLAMES/GIRAFFE spectrograph, a fiber-fed multiobject spectrograph mounted on the Nasmyth focus of UT2 (Pasquini 2003). USNO-A0600-15865535 is a bright (\(g= 14.2\)) very blue (\(g-r = -0.21\)) source, unresolved in Pan-STARRS imaging (Magnier et al., 2013; Schlafly et al., 2012; Tonry et al., 2012). Through a combination of overoptimism and clerical error, it was targeted as a quasar candidate behind the circumgalactic medium of M83.

The GIRAFFE High Resolution mode was used in the H395.8 setup (HR02), which gives access to the 385.4 to 404.9nm wavelength range in the near UV at a spectral resolving power \(\lambda / \delta \lambda\) of 22,700. A total of 4.5 h of integration were obtained, split in 3 exposures of 90 min. Given that most of the program targets were faint, the non-standard 50 khz,1×1, high gain readout mode was used