Vertical motion in the Galactic disc: unwinding the Snail

TL;DR

This paper models the spiral structure called the Snail in the Milky Way's vertical stellar distribution, interpreting it as a winding perturbation whose properties vary across the galaxy, revealing insights into its dynamical history.

Contribution

The study introduces a simple, physically interpretable model of the Snail spiral, linking its morphology to perturbation amplitude and age, and applies it to Gaia data across different galactic regions.

Findings

The perturbation amplitude is smaller in the inner galaxy and larger in the outer disc.

The estimated perturbation age varies between 0.2 and 0.6 Gyr across different regions.

Systematic residuals suggest the need for more complex models.

Abstract

The distribution of stars in the Milky Way disc shows a spiral structure--the Snail--in the space of velocity and position normal to the Galactic mid-plane. The Snail appears as straight lines in the vertical frequency--vertical phase plane when effects from sample selection are removed. Their slope has the dimension of inverse time, with the simplest interpretation being the inverse age of the Snail. Here, we devise and fit a simple model in which the spiral starts as a lopsided perturbation from steady state, that winds up into the present-day morphology. The winding occurs because the vertical frequency decreases with vertical action. We use data from stars in Gaia EDR3 that have measured radial velocities, pruned by simple distance and photometric selection functions. We divide the data into boxels of dynamical invariants (radial action, angular momentum); our model fits the data well in many of the boxels. The model parameters have physical interpretations: one, $A$, is a perturbation amplitude, and one, $t$, is interpretable in the simplest models as the time since the event that caused the Snail. We find trends relating the strength and age to angular momentum: (i) the amplitude $A$ is small at low angular momentum ($<1\,600\mathrm{\,kpc\,km\,s}^{-1}$ or guiding-centre radius $< 7.3\,$kpc), and over a factor of three larger, with strong variations, in the outer disc; (ii) there is no single well-defined perturbation time, with $t$ varying between 0.2 and 0.6 Gyr. Residuals between the data and the model display systematic trends, implying that the data call for more complex models.