The innermost parts of the Milky Way (MW) are very difficult to observe due to the high extinction along the line of sight, especially close to the disc mid-plane. However, this region contains the most massive complex stellar component of the MW, the bulge, primarily composed of disc stars whose structure is (re-)shaped by the evolution of the bar. In this work, we extend the application of the orbit superposition method to explore the present-day 3D structure, orbital composition, chemical abundance trends and kinematics of the MW bulge. Thanks to our approach, we are able to transfer astrometry from Gaia and stellar parameters from APOGEE DR 17 to map the inner MW without obscuration by the survey footprint and selection function. We demonstrate that the MW bulge is made of two main populations originating from a metal-poor, high- α thick disc and a metal-rich, low- α thin disc, with a mass ratio of 4:3, seen as two major components in the metallicity distribution function (MDF). Finer MDF structures hint at multiple sub-populations associated with different orbital families of the bulge, which, however, have broad MDFs themselves. Decomposition using 2D Gaussian Mixture Models in the Fe/H-Mg/Fe plane identifies five components, including a population with ex-situ origin. Two dominant ones correspond to the thin and thick discs, and two in between trace the transition between them. We show that there is no universal metallicity gradient value that can characterise the MW bulge. The radial gradients closely trace the X-shaped bulge density structure, while the vertical gradient variations follow the boxy component. The MW bulge, while on average having subsolar metallicity, is more metal-rich compared to the surrounding disc populations, in agreement with extragalactic observations and state-of-the-art simulations, reinforcing its secular origin.
Khoperskov et al. (Fri,) studied this question.