Dynamically hot recent star bursts in the Galactic disc

Andreas Just 1 , Kseniia Sysoliatina 1

  • 1 ARI/ZAH, Heidelberg University, Heidelberg

Abstract

The JJ-model is a semi-analytic model of the Milky Way disc based on the star formation history (SFR), an age-velocity dispersion relation (AVR), an age-metallicity relation (AMR), and the initial mass function (IMF) for thin and thick disc. For solving the Poisson equation gas and Dark matter are included. The current version of the model is mainly based on Gaia DR2 and APOGEE data. It reproduces the solar cylinder and provides age distributions, vertical density profiles and velocity distribution functions for different stellar types. For star count predictions the stellar halo is included by a simple model.
The high quality and large sample size of the Gaia DR2 data enabled us to identify two star formation peaks on top of the smooth SFR of the thin disc. The older burst with an age of 3 Gyr is mainly visible in F main sequence stars, whereas the younger burst visible in A stars is about 0.5 Gyr old and may still be active at present day. The vertical velocity dispersions are 26 km/s and 12 km/s, respectively, which is significantly lower than the velocity dispersion of thick disc stars, but about a factor of two larger than the corresponding values of the AVR for the thin disc. This is a strong hint that these enhanced SFR was triggered by external perturbations.

Milky Way Model

The JJ-model of the Galactic disc is a semianalytic model based on analytic input functions for the star formation history (SFR), the IMF, the age-velocity dispersion relation (AVR), and the age-metallicty relation (AMR). In the model the vertical density profiles are derived by solving the Poisson and Jeans equations including the gravitational potential of the ISM and the dark matter halo. The comparison with observational data in order to optimize the input parameters is done in a forward procedure. After calculating the model for a chosen parameter set the stellar populations are derived based on PARSEC or MIST stellar libraries. Then an extinction model is applied, the limits and errors of the data as well as the completeness functions of the data samples are applied. The comparison is then done directly in the observational space. Therefore the density and velocity profiles shown here are profiles of the observational samples and not the intrinsic (complete) profiles of the model. For the basic model see Just & Jahreiss (MNRAS 402, 461, 2010) The most updated version is described in detail in Sysoliatina & Just (A&A 647, A39, 2021).

The four fundamental input functions of the disc model. JJ10: Just& Jahreiss 2010, RJ15: Rybizki & Just 2015, R17: Rybizki 2017.

The AMR is derived by fitting the observed metallicity distribution function (MDF) of red clump stars from APOGEE.

MDF comparison of APOGEE data with the disc model for thin (right histograms) and thick (left histograms) disc (discs are separated by the alpha-enhancement).

The identification of the observed cumulative metallicity distribution with the cumulative age distribution of the corresponding populations in the model results in the AMR for thin and thick disc.

Gaia DR2 Data

We selected three nearly complete samples in apparent magnitudes:

  • full 6-D information including radial velocities G=7-12mag
  • 5-D astrometric solution G=7-17mag
  • cone sample |b|>80°: photometry only

The incompleteness of the observed samples in the local volume as function of z due to the magnitude limits are taken into account for each subsample accordingly. 

Gaia sources with coloured areas showing the selected magnitude ranges.

Local cylinder with |z|<600pc and maximum and minimum distances in the plane selected by the magnitude range of the samples. The galactocentric radial range is additionally cut in order to stay local.

Subsamples without (top panels) and with (bottom panels) radial velocities.

The stellar subsamples are chosen across the CMD in order to cover a wide range of age distributions.

Age distributions of the different samples predicted by the model. Results with the PARSEC isochrones are more noisy due to a simpler interpolation scheme.

Results

The Gaia DR2 data show for the first time that the vertical density profiles of A and F stars are shallower than expected. This is inconsistent with the low velocity dispersion in the solar neighbourhood based on simple models. We find that additional SFR peaks with enhanced velocity dispersions can resolve this discrepancy.

Velocity dispersion of the two SFR peaks compared to the Age-Velocity-Dispersion relation (AVR) of local disc (blue line).

Local star formation history (black dashed line). The contributions of thick disc (yellow) and thin disc incl. the SFR peaks (green) are shown. For comparison the old thin disc SFR (pink, Just et al. 2010) and the adapted standard model without SFR peaks (blue) are plotted.

Density profiles of observed A and F stars compared to the standard model (red) without SFR peaks and the best fit models with PARSEC and MIST isochrones (blue and green lines).

Based on the Becanson Galaxy model (BGM, Mor et al 2019) the earlier SFR burst with an age of approx. 3 Gyr was discovered. In our local model we have a better age resolution, which allows us to identify also the younger SFR peak. Addititionally, the velocity dispersions of the 2 SFR peaks are free parameters in our model. The higher velocity dispersions compared to the AVR of the disc results in the enhanced number densities of the young populations (A and F stars) at larger distances z>300pc as observed.

Introduction

All recent work on the JJ-model and all plots shown here were done by Kseniia Sysoliatina. The results are published in Sysoliatina & Just (A&A 647, A39, 2021). For more details about the SFR peaks see the contributed talk by Kseniia in SS21 on Friday, 9:45-10:00 (#825). Details on the extension of the JJ-model to a large range of Galactocentric radii are presented by Kseniia on her ePoster #999 with title 'Stellar population synthesis across the Galactic disk with JJ model. Possible implications of Gaia EDR3'.