Search

The detection of intermediate-mass black holes in the centres of globular clusters is highly controversial, as complementary observational methods often deliver significantly different results. In order to understand these discrepancies, we develop a procedure to simulate integral field unit (IFU) observations of globular clusters: Simulating IFU Star Cluster Observations (sisco). The inputs of our software are realistic dynamical models of globular clusters that are then converted in a spectral data cube. We apply sisco to Monte Carlo cluster simulations with a realistic number of stars and concentrations. Using independent realizations of a given simu-lation we are able to quantify the stochasticity intrinsic to the problem of observing a partially resolved stellar population with integrated-light spectroscopy. We show that the luminosity-weighted IFU observations can be strongly biased by the presence of a few bright stars that introduce a scatter in the velocity dispersion measurements up to ≃40 per cent around the expected value, preventing any sound assessment of the central kinematic and a sensible inter-pretation of the presence/absence of an intermediate-mass black hole. Moreover, we illustrate that, in our mock IFU observations, the average kinematic tracer has a mass of ≃0.75 M⊙, only slightly lower than the mass of the typical stars examined in studies of resolved line-of-sight velocities of giant stars. Finally, in order to recover unbiased kinematic measurements we test different masking techniques that allow us to remove the spaxels dominated by bright stars, bringing the scatter down to a level of only a few per cent. The application of sisco will allow us to investigate state-of-the-art simulations as realistic observations. [ABSTRACT FROM AUTHOR]

Accurate determination of the local dark matter density is important for understanding the nature and distribution of dark matter in the Universe. This requires that the local velocity distribution is characterized correctly. Here, we present a kinematic study of 16,276 SDSS/SEGUE G-type dwarf stars in the solar neighbourhood, with which we determine the shape of the velocity ellipsoid in the meridional plane. We separate our G-dwarf stars based on their [Fe/H] and [a/Fe] abundances and infer the local velocity distribution independently for each subsample using a maximum-likelihood method that accounts for possible contaminants. We show by constructing vertical Jeans models that the different sub-samples yield consistent results only when we allow the velocity ellipsoid in the disc to be tilted, demonstrating that the common assumption of decoupled radial and vertical motions in the disc is incorrect. Further, we obtain that the tilt of the velocity ellipsoid is consistent among the different sub-samples. We find that increase in the tilt with height is well described by the relation αtilt = (-0.90 ± 0.04) arctan(|z|/R⊙) - (0.01 ± 0.005), which is close to alignment with the spherical coordinate system and hence a velocity ellipsoid pointing to the Galactic Centre. We also confirm earlier findings that the sub-samples behave almost isothermally with both radial and vertical velocity dispersion approximately constant with height. We conclude that the coupling between radial and vertical motion captured in the velocity ellipsoid tilt cannot be ignored when considering dynamical models of the solar neighbourhood. In a subsequent paper, we will develop a new modelling scheme informed by these results and make an improved determination of the local dark matter density. [ABSTRACT FROM AUTHOR]