(Abridged) We analyzed the stellar parameters and radial velocities of ~1200 stars in five bulge fields as determined from the Gaia-ESO survey data (iDR1). We use VISTA Variables in The Via Lactea (VVV) photometry to obtain reddening values by using
a semi-empirical T_eff-color calibration. From a Gaussian decomposition of the metallicity distribution functions, we unveil a clear bimodality in all fields, with the relative size of components depending of the specific position on the sky. In agreement with some previous studies, we find a mild gradient along the minor axis (-0.05 dex/deg between b=-6 and b=-10) that arises from the varying proportion of metal-rich and metal-poor components. The number of metal-rich stars fades in favor of the metal-poor stars with increasing b. The K-magnitude distribution of the metal-rich population splits into two peaks for two of the analyzed fields that intersects the near and far branches of the X-shaped bulge structure. In addition, two lateral fields at (l,b)=(7,-9) and (l,b)=(-10,-8) present contrasting characteristics. In the former, the metallicity distribution is dominated by metal-rich stars, while in the latter it presents a mix of a metal-poor population and and a metal-intermediate one, of nearly equal sizes. Finally, we find systematic differences in the velocity dispersion between the metal-rich and the metal-poor components of each field. Our chemo-kinematical analysis is consistent with a varying field-to-field proportion of stars belonging to (i) a metal-rich boxy/peanut X-shaped component, with bar-like kinematics, and (ii) a metal-poor more extended rotating structure with a higher velocity dispersion that dominates far from the Galactic plane. These first GES data allow studying the detailed spatial dependence of the Galactic bulge populations, thanks to the analysis of individual fields with relatively high statistics.
(Abridged) We have used the atmospheric parameters, [alpha/Fe] abundances and radial velocities, determined from the Gaia-ESO Survey GIRAFFE spectra of FGK-type stars (iDR1), to provide a chemo-kinematical characterisation of the disc stellar populat
ions. We focuss on a subsample of 1016 stars with high quality parameters, covering the volume |Z|<4.5kpc and R in the range 2-13kpc. We have identified a thin to thick disc separation in the [alpha/Fe] vs [M/H] plane, thanks to the presence of a low-density region in the number density distribution. The thick disc stars seem to lie in progressively thinner layers above the Galactic plane, as metallicity increases and [alpha/Fe] decreases. The thin disc population presents a constant value of the mean distance to the plane at all metallicities. Our data confirm the already known correlations between V_phi and [M/H] for the two discs. For the thick disc sequence, a study of the possible contamination by thin disc stars suggests a gradient up to 64km/s/dex. The distributions of V_phi, V_Z, and orbital parameters are analysed for the chemically separated samples. Concerning the gradients with galactocentric radius, we find for the thin disc a flat behaviour of V_phi, a [M/H] gradient of -0.058dex/kpc and a small positive [alpha/Fe] gradient. For the thick disc, flat gradients in [M/H] and [alpha/Fe] are derived. Our chemo-kinematical analysis suggests a picture in which the thick disc seems to have experienced a settling process, during which its rotation increased progressively, and, possibly, the V_phi dispersion decreased. At [M/H]-0.25dex and [alpha/Fe]0.1dex, the mean characteristics of the thick disc in distance to the Galactic plane, V_phi, V_phi dispersion and eccentricity agree with those of the thin disc stars, suggesting a possible connection between these populations at a certain epoch of the disc evolution.
In spite of many observational efforts aiming to characterize the chemical evolution of our Galaxy, not much is known about the origin of fluorine (F). Models suggest that the F found in the Galaxy might have been produced mainly in three different w
ays, namely, Type II supernovae, asymptotic giant branch nucleosynthesis, or in the core of Wolf-Rayet stars. Only a few observational measurements of F abundances are available in the literature and mostly for objects whose characteristics might hamper an accurate determination of fluorine abundance (e.g.,complex mixing and nucleosynthesis processes, external/internal contamination). We derive the F abundances for a set of nine cool main-sequence dwarfs in the solar neighbourhood, based on an unblended line of the HF molecule at 2.3 microns. In addition, we study the s-process elements of five of these stars. We acquire data using the high-resolution IR-spectrograph CRIRES and gather FEROS data from the European Southern Observatory archive. Several of the analysed stars seem to be slightly fluorine enhanced with respect to the Sun, although no correlation is found between the F abundance and the iron content. In addition, the most fluorine enriched stars are also yttrium and zirconium enriched, which suggests that AGB fluorine nucleosynthesis is the dominant source of fluorine production for the observed stars. Nevertheless, the correlation between [F/Fe] and the s-elements is rather weak and possibly masked by the uncertainties in the F abundance measurements. Finally, we compare our derived F abundances to previous measurements of alpha-element and iron-peak element abundances. Type II core collapse Supernovae do not appear to be the main site of F production for our targets, as no correlation seems to exist between the [F/Fe] and the [alpha/Fe] ratios.
The parameter fit from a model grid is limited by our capability to reduce the number of models, taking into account the number of parameters and the non linear variation of the models with the parameters. The Local MultiLinear Regression (LMLR) algo
rithms allow one to fit linearly the data in a local environment. The MATISSE algorithm, developed in the context of the estimation of stellar parameters from the Gaia RVS spectra, is connected to this class of estimators. A two-steps procedure was introduced. A raw parameter estimation is first done in order to localize the parameter environment. The parameters are then estimated by projection on specific vectors computed for an optimal estimation. The MATISSE method is compared to the estimation using the objective analysis. In this framework, the kernel choice plays an important role. The environment needed for the parameter estimation can result from it. The determination of a first parameter set can be also avoided for this analysis. These procedures based on a local projection can be fruitfully applied to non linear parameter estimation if the number of data sets to be fitted is greater than the number of models.
