Hubble Space Telescope Evidence for an Intermediate-Mass Black Hole in the Globular Cluster M15. I. STIS Spectroscopy and WFPC2 Photometry

TL;DR: In this paper, the authors describe a project with the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) to measure the line-of-sight velocities of stars in the central few arcseconds of the dense globular cluster M15.

Abstract: In this series of two papers, we describe a project with the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) to measure the line-of-sight velocities of stars in the central few arcseconds of the dense globular cluster M15. The main goal of this project is to search for the possible presence of an intermediate-mass central black hole. This first paper focuses on the observations and reduction of the data. We scanned the central region of M15 spectroscopically by consecutively placing the 01 HST/STIS slit at 18 adjacent positions. The spectral pixel size exceeds the velocity dispersion of M15. This puts the project at the limit of what is feasible with STIS, and exceedingly careful and complicated data reduction and analysis were required. We applied corrections for the following effects: drifts in the STIS wavelength scale during an HST orbit; the orbital velocity component of HST along the line of sight to the cluster, and its variations during the HST orbit; and the apparent wavelength shift that is perceived for a star that is not centered in the slit. The latter correction is particularly complicated and requires many pieces of information: (1) the positions and magnitudes of all the stars near the center of M15; (2) accurate positionings of the STIS slits during the observations; (3) and the HST/STIS point-spread function (PSF) and line-spread function (LSF). To address the first issue we created a stellar catalog of M15 from the existing HST/WFPC2 data discussed previously by Guhathakurta et al., but with an improved astrometric and photometric calibration. The catalog is distributed electronically as part of this paper. It contains 31,983 stars with their positions and U, B, and V magnitudes. To address the second issue, we model the observed intensity profiles along the STIS slits to determine the slit positionings to 0007 accuracy in each coordinate. To address the third issue, we obtained observations of a bright field star to which we fitted multi-Gaussian PSF and LSF models. Upon reduction of the M15 spectroscopy we ultimately obtain 19,200 one-dimensional STIS spectra, each for a different aperture position in M15, with a velocity scale accurate to better than 2.5 km s-1. We develop an algorithm that co-adds the spectra for individual apertures and use it to extract spectra of individual stars with minimum blending and maximum signal-to-noise ratio. In Paper II we use these spectra to extract reliable line-of-sight velocities for 64 stars, half of which reside within R = 24 from the cluster center. These velocities constrain the central structure, dynamics, and mass distribution of the cluster.