Re-evaluation of the central velocity-dispersion profile in NGC 6388

¹ ESA, Space Science Department, Keplerlaan 1, 2200 AG Noordwijk The Netherlands
e-mail: nluetzge@rssd.esa.int
² Department of Astronomy, University of Texas at Austin, Austin, TX 78712, USA
³ School of Mathematics and Physics, University of Queensland, Brisbane, QLD 4072, Australia
⁴ Max-Planck-Institute for Astronomy, Königstuhl 17, 69117 Heidelberg, Germany
⁵ Gemini Observatory, Northern Operations Center, 670 N. A’ohoku Place Hilo, Hawaii 96720, USA
⁶ European Southern Observatory (ESO), Karl-Schwarzschild-Strasse 2, 85748 Garching, Germany
⁷ Sterrewacht Leiden, Leiden University, Postbus 9513, 2300 RA Leiden, The Netherlands

Received: 16 December 2014
Accepted: 8 July 2015

Abstract

Context. The globular cluster NGC 6388 is one of the most massive clusters in our Milky Way and has been the subject of many studies. Recently, two independent groups found very different results when measuring its central velocity-dispersion profile with different methods. While we found a rising profile and a high central velocity dispersion (23.3 km s^-1), measurements obtained by Lanzoni et al. (2013, ApJ, 769, 107) showed a value lower by 40%. The value of the central velocity dispersion has a serious effect on the mass and possible presence of an intermediate-mass black hole at the center of NGC 6388.

Aims. The goal of this paper is to quantify the biases arising from measuring velocity dispersions from individual extracted stellar velocities versus the line broadening measurements of the integrated light using new tools to simulate realistic observations made with integral field units (IFU).

Methods. We used a photometric catalog of NGC 6388 to extract the positions and magnitudes from the brightest stars in the central three arcseconds of NGC 6388 and created a simulated SINFONI and ARGUS dataset. The IFU data cube was constructed with different observing conditions (i.e., Strehl ratios and seeing) reproducing the conditions reported for the original observations as closely as possible. In addition, we produced an N-body realization of a ~10⁶ M_⊙ stellar cluster with the same photometric properties as NGC 6388 to account for unresolved stars.

Results. We find that the individual radial velocities, that is, the measurements from the simulated SINFONI data, are systematically biased towards lower velocity dispersions. The reason is that the velocities become biased toward the mean cluster velocity as a result of the wings in the point spread function of adaptive optics (AO) corrected data sets. This study shows that even with AO supported observations, individual radial velocities in crowded fields do not reproduce the true velocity distribution. The ARGUS observations do not show this kind of bias, but they were found to have larger uncertainties than previously obtained. We find a bias toward higher velocity dispersions in the ARGUS pointing when fixing the extreme velocities of the three brightest stars, but these variations are within the determined uncertainties. We reran Jeans models and fit them to the kinematic profile with the new uncertainties. This yielded a black-hole mass of M_• = (2.8 ± 0.4) × 10⁴ M_⊙ and M/L ratio M/L = (1.6 ± 0.1) M_⊙/L_⊙, consistent with our previous results.