Modeling the Milky Way

The understanding of our Galaxy is the main goal of our team. For this reason we have many opened studies facing different aspect of the populations present in the Milky Way (MW).


Research lines

N-body and hydrodynamical simulations: diagnosing the nature of spiral arms

Pure N-body simulations performed with the ART Tree N-body code have been conducted for the kinematic analysis of the stellar disc component. Our first results, obtained considering a high number of disc particles and low time step and spatial resolution reveals two clearly different behaviours for the Spiral Arm Pattern speed (Roca-Fàbrega et al. 2013). Whereas subdominant disc cases present transient spiral arm features co-rotating with particles, Milky Way like simulations with high disk/halo ratio develops transient spiral compatible with a pattern speed constant in radius. Both cases significantly depart from the spiral arm density wave theory.
In parallel, to facilitate the analysis of the density function on the phase space of the simulated galactic discs – test particle and N-body – we have implemented statistical tools clustering algorithms, and moments of the distribution function (vertex deviation and third order moments) (Roca-Fàbrega et al., 2010a; Master Thesis; Roca-Fàbrega et al. 2010b; Roca-Fàbrega et al. 2011). This study reveals that the vertex deviation of the velocity ellipsoid is a good tracer of Spiral Arms’ density perturbation and resonance radius (Roca-Fàbrega et al. 2013, to be submitted).
New simulations, now using ART code + hidrodynamics (HART), in a cosmological context, are under study. We plan to use them to study how big is the impact of the galactic environment (gas, galactic neighbours, supernovae feedback, …) in the kinematics and large structure formation of galactic disks. Our first results suggest that systems with more than one misaligned bar are present in the universe due to the existence of different mechanisms that can generate them (i.e. secular evolution, interactions, primordial monolithic collapse, …).

Manifolds model to model the Galactic spiral arms and edge-on discs

In 2006-2010 we proposed a new theory to explain the formation of both spirals and rings in barred galaxies using a common dynamical framework (Athanassoula, Romero-Gómez & Masdemont, 2009) based on the orbital motion driven by the unstable equilibrium points of the rotating bar potential. In Romero-Gómez et al. (2011) , for the first time, we used this method to study the bar-driven dynamics in the inner part of our Milky Way. Our model reproduces well the relevant over-densities found in the galactic longitude-velocity CO diagram. Our model states that to reproduce the morphology of the outer spiral arms would necessitate either more massive and more rapidly rotating bars, or including in the potential an extra component describing the spiral arms.
Currently, we study the morphology given by the invariant manifolds in edge-on 3D galactic models obtained by imposing a tilt of the galactic bar. Both the orbits trapped by the invariant manifolds and test particle simulations under this imposed bar potential make a warped disc similar to the ones seen in external galaxies (Sanchez-Martin, Romero-Gomez, Masdemont 2013, in preparation).

The Red Clump population as seen by Gaia. First scientific applications

We have created a catalogue with the aim of mimicking the characteristics of the Red Clump population in the Milky Way. The catalogue is obtained by generating a set of initial conditions with the number of particles matching the number of Red Clump stars in the Solar Neighbourhood and having kinematic characteristics of the Red Clump (K0-1III stars). Then we have integrated these particles in a 3D Galactic model composed of an axisymmetric component and a combination of bulge and Long bar. Finally, using the photometry characteristic of the Red Clump and a realistic Drimmel extinction, we compute the Gaia G magnitude and create a first catalogue, that is, all particles with G<20. This catalogue is useful to make a first statistical study of what Gaia will see of the Galactic disc. A second catalogue, selecting only particles with an error in Radial Velocity better than 10km/s is also analysed in statistical terms in Romero-Gomez et al (2013, to be submitted). However, this second catalogue is also useful to perform a first scientific application, that is, the study of the moments of the velocity distribution in the Galactic disc to study the non-axisymmetric component of the potential (Romero-Gomez et al, 2014, in preparation).

How can Gaia help us with detecting and characterising the Galactic warp?

We use several stellar populations, in particular the red clump stars mentioned above to trace the kinematics of the Galactic warp. We use again 3D test particle simulations, and modify adiabatically the Galactic disc to make it warped. Using the final configuration of the warped population, we study the effects of the warped potential in the components of the velocity vector. We observe that the kinematic signature of the warp is more significant in the vertical direction (either in the W component of velocity vector or the \mu_b component of Galactic proper motion). We also compare the theoretical results with the ones given by observations (UCAC4). Finally, we implement another method to better assess the characteristics of the Galactic warp based on great circle cell count (mgc3 method). We also add the Gaia errors to the observables from our warp simulations and check how these errors can affect the kinematic signature of the warp (Abedi et al, 2014, in preparation).

Preparing the Besançon Galaxy Model for the comparison with Gaia

In 2010, in the context of the Gaia CU2 coordination unit, we started the optimization the Besançon Population Synthesis model (Robin et al. 2003), a powerful model that provides a different approach to Galactic Dynamics for a comprehensive understanding of the formation and evolution of the MW. The model has been optimized to handle variations of the star formation rate, initial mass function and evolutionary tracks. Multivariate analysis tools have been implemented to tune the free parameters of the model to the full sky Tycho-2 catalogue (Czekaj et al., 2011). High dependence on the interstellar extinction models has been tested. First results indicates that a decreasing SFR and an IMF combination of Kroupa and Haywood are the most plausible scenarios for the solar neighbourhood (Czekaj, 2012,PhD, and Czekaj et al, 2013, submitted).

Local kinematic groups in the Milky Way

There are groups of stars that present common motions in the Galaxy. They are called kinematic groups. We have deeply evaluated the dynamics that favours the appearance of kinematic groups, such as the ones observed in the solar neighbourhood (Antoja et al 2011). One of the most plausible explanations for the origin of the moving groups is the orbital and resonant regions related to the large scale structure (bar and spiral arms) of the Milky Way (Antoja et al, 2010). Our test particle simulations predict a stellar kinematic response to the spiral arms and bar strongly dependent on disc position. The kinematics of the solar neighbourhood can be used as a constraint to understand the dynamics of the bar(s) and spiral structure of our galactic disc (Antoja et al 2009). Spiral arms induce some imprints on the velocity at solar regions and could be used to infer some properties of these arms, such as pattern speed, strength, orientation and lifetime (Antoja et al. 2001). The constraint of the MW spiral structure properties, however, is presently not straightforward. Our test-particle simulations have shown that the parameter space is still too large to obtain a unique fitting to the observed velocity distribution. Gaia capabilities have been deeply evaluated in this context (Antoja et al. 2010). A recent study of the kinematic groups observed by RAVE survey has allowed us, for the first time to study the dependence on Galactic position of the (thin and thick) disc moving groups, indicating that they are large scale features, and pointing out, once more, that they are indeed dynamical effects (Antoja et al.2012 , Figueras et al. 2011).

Related thesis