Probability of forming gaps in the GD-1 stream by close encounters of globular clusters

Abstract

One of the most intriguing properties of the GD-1 stellar stream is the existence of three gaps. If these gaps were formed by close encounters with dark matter subhalos, the GD-1 stream opens an exciting window through which we can see the size, mass, and velocity distributions of the dark matter subhalos in the Milky Way. However, in order to use the GD-1 stream as a probe of the dark matter substructure, we need to disprove that these gaps are not due to the perturbations from baryonic components of the Milky Way. Here we ran a large number of test particle simulations to investigate the probability that each of the known globular clusters (GCs) can form a GD-1-like gap, by using the kinematical data of the GD-1 stream and GCs from Gaia EDR3 and by fully taking into account the observational uncertainty. We found that the probability that all of the three gaps were formed by GCs is as low as 1.2 × 10 −5 and the expected number of gaps formed by GCs is only 0.057 in our fiducial model. Our result highly disfavors a scenario in which GCs form the gaps. Given that other baryonic perturbers (e.g., giant molecular clouds) are even less likely to form a gap in the retrograde-moving GD-1 stream, we conclude that at least one of the gaps in the GD-1 stream was formed by dark matter subhalos if the gaps were formed by flyby perturbations.

A Greedy and Optimistic Approach to Clustering with a Specified Uncertainty of Covariates

“…(Note that stars that are too far away from the Sun are too faint to be observed.) The assumed position and velocity of the Sun is the same as those in Doke and Hattori (2022). We chose n k = 1 + {(k − 1) mod 10}, where "x mod a" denotes the residual of the division (x divided by a).…”