2014MNRAS.443..828C

We present a detailed analysis of high-resolution, high signal-to-noise ratio spectra for five Aquarius stream stars observed with the Magellan Inamori Kyocera Echelle spectrograph on the Magellan Clay telescope. Our sample represents one-third of the 15 known members in the stream. We find the stream is not monometallic: the metallicity ranges from [Fe/H] = -0.63 to -1.58. No anticorrelation in Na-O abundances is present, and we find a strong positive Mg-Al relationship, similar to that observed in the thick disc. We find no evidence that the stream is a result of a disrupted classical globular cluster, contrary to a previously published claim. High [(Na, Ni, α)/Fe] and low [Ba/Y] abundance ratios in the stream suggest that it is not a tidal tail from a disrupted dwarf galaxy, either. The stream is chemically indistinguishable from Milky Way field stars with the exception of one candidate, C222531-145437. From its position, velocity, and detailed chemical abundances, C222531-145437 is likely a star that was tidally disrupted from ω-Centauri. We propose that the Aquarius stream is Galactic in origin, and could be the result of a disc-satellite perturbation in the Milky Way thick disc of the order of a few Gyr ago: derived orbits, UVW velocities, and angular momenta of the Aquarius members offer qualitative support for our hypothesis. Assuming that C222531-145437 is a tidally disrupted member of ω-Centauri, this system is the most likely disc perturber. In the absence of compelling chemical and/or dynamical evidence that the Aquarius stream is the tidal tail of a disrupted satellite, we advocate the `Aquarius group' as a more appropriate description. Like the Canis Major overdensity, as well as the Hercules and Monoceros groups, the Aquarius group joins the list of kinematically identified substructures that are not actually accreted material: they are simply part of the rich complexity of the Milky Way structure.