2022MNRAS.509..844T

The Milky Way disc exhibits intricate orbit substructure of still debated dynamical origin. The angle variables (θ_ph, θ_R) - which are conjugates to the actions (L_z, J_R), and describe a star's location along its orbit - are a powerful diagnostic to identify l:m resonances via the orbit shape relation Δθ_R/Δθ_ph = -m/l. In the past, angle signatures have been hidden by survey selection effects (SEs). Using test particle simulations of a barred galaxy, we demonstrate that Gaia should allow us to identify the Galactic bar's outer Lindblad resonance (OLR;l = +1, m = 2) in angle space. We investigate strategies to overcome SEs. In the angle data of the Gaia Data Release 2 (DR2) Radial Velocity Spectrometer (RVS) sample, we independently identify four candidates for the OLR and therefore for the pattern speed Ω_bar. The strongest candidate, Ω_bar ∼ 1.4Ω₀, positions the OLR above the 'Sirius' moving group, agrees with measurements from the Galactic Centre, and might be supported by higher order resonances around the 'Hercules/Horn'. But it misses the classic orbit orientation flip, as discussed in the companion study on actions. The candidate Ω_bar ∼ 1.2Ω₀ was also suggested by the action-based study, has the OLR at the 'Hat', is consistent with slow bar models, but still affected by SEs. Weaker candidates are Ω_bar = 1.6Ω₀ and 1.74Ω₀. In addition, we show that the stellar angles do not support the 'Hercules/Horn' being created by the OLR of a fast bar. We conclude that - to resolve if 'Sirius' or 'Hat' is related to the bar's OLR - more complex dynamical explanations and more extended data with well-behaved SEs are required.