DOES THE MILKY WAY HAVE AN ACCRETED DISK COMPONENT? Gregory Ruchti Lund Observatory!!!! in collaboration with: Justin Read, Sofia Feltzing, Thomas Bensby, Antonio Pipino
ACCRETED STARS Our current cosmology requires the merging and associated accretion of stars and dust to form large-scale structure. The halo is most sensitive to small substructures accreted halo stars The disk is more sensitive to massive mergers that reach higher metallicity and suffer from dynamical friction and disk plane dragging accreted disk stars Accreted disk stars probe late, massive mergers and the dark matter disk
THE CHEMO- Ruchti et al. (2014) DYNAMICAL TEMPLATE [Mg/Fe] Ez/Ec Accreted Heated Accreted halo stars [Fe/H] Accreted disk stars Accreted disk stars inhabit low Ez/Ec and Jz/Jc and low [α/fe], distinct from Galactic disk stars. max Z above plane Ez/Ec eccentricity simulations from Read et al. (2008) Jz/Jc
SEARCHING FOR THE ACCRETED DISK Ruchti et al. (2014) 522 G. R. Ruchti et al. [Mg/Fe] [Fe/H] Figure 5. The high-resolution spectroscopic sample, R13, plotted in the [[Mg/Fe],[Fe/H]] plane. Stars are separated into high- and low-α regimes, with Sample a cut at 0.3 dex in from [Mg/Fe] (as defined in Section 3.1). The data were divided into different metallicity bins shown by the vertical dashed lines. TheRuchti uncertainty in [Fe/H] et al. and [Mg/Fe] (2011;2013) are less than 0.1 dex. Subsequent plots are colour-coded in the same manner. ~300 stars initially selected to investigate metal-poor GM disk. disc and the Miyamoto & Nagai (1975) potentialforthedisc: disc = R 2 + (a +. (10) z 2 + b 2 ) 2 Finally, we assumed a NFW dark matter profile (Navarro, Frenk & Ez/Ec Accreted Stars Jz/Jc Figure 6. The vertical component of the specific orbital angular momentum versus the vertical component of the orbital energy for the R13 data set.
SEARCHING FOR 8 G. R. Ruchti et al. THE ACCRETED DISK 518 G. R. Ruchti et al. Ruchti et al. (2014) No accreted disk. [Mg/Fe] Satellite mass ure 1. Left: location of MW stars (red); the LMC (orange); and giant stars in the Sagittarius, Carina, and Fornax dwarf galaxies (blue) in the [[Fe/H], g/fe]] plane. Data compiled from Bensby, Feltzing & Oey (2014), Ruchti et al. (2013), Nissen & Schuster (2010), Van der Swaelmen et al. (2013), Carretta l. (2010), Lemasle et al. (2012)andLetarteetal.(2010). For [Fe/H] > 1.3(verticaldashedlines)thesatellitegalaxieshave[Mg/Fe]ratiostypicallyless n 0.3 dex (horizontal dashed line), and are lower [Fe/H] than the typical MW star at similar metallicities. Right: a simple chemical evolution model to explain Figure 1. Left: location of MW stars (red); the LMC (orange); and giant stars in the Sagittarius, Carina, and Fornax dwarf galaxies (blue) in the [[Fe/H], se data (see Section 3.1.1 for details). We show results for stars in the [[Fe/H], [Mg/Fe]] plane after 8 Gyr of star formation with an assumed exponentially [Mg/Fe]] plane. Data compiled from Bensby, Feltzing & Oey (2014), Ruchti et al. (2013), Nissen & Schuster (2010), Van der Swaelmen et al. (2013), Carretta lining gas inflow. We plot two models of different mass: 10 et al. (2010), Lemasle et al. (2012)andLetarteetal.(2010). For [Fe/H] > 1.3(verticaldashedlines)thesatellitegalaxieshave[Mg/Fe]ratiostypicallyless thanmore 0.3 dex (horizontal massive dashed line), andsatellites 9 M (Dwarf; red); and 1012 M (MW; black). The dwarf model assumes a low star formation ciency ν = 0.1 Gyr 1 ; we consider both ν = 0.1 1 Gyr are lower 1 for the MW model. In both cases, we plot results for two different SNIa rates (solid and than the typical MW star at similar metallicities. Right: a simple chemical evolution model to explain ted lines, as marked). these data (see Section 3.1.1 for details). We show results for stars in the [[Fe/H], [Mg/Fe]] plane after 8 Gyr of star formation with an assumed exponentially declining should gas inflow. Wereach plot two modelshigher of different mass: [Fe/H], 10 9 M (Dwarf; red); and 1012 M (MW; black). The dwarf model assumes a low star formation mano 2013; efficiency Micali,Matteucci&Romano ν = 0.1 Gyr and we 1 ; we consider 2013, bothto ν mention = 0.1 andsome 1 Gyr see none at the 1 for[mg/fe] the MW model. tail of aindistribution both cases, we of stars plot results formed forattwo a single different epoch; SNIathey rates (solid and dotted lines, as marked). the most recent versions) and of local dwarfs (e.g. Lanfranchi, could have formed at the same location as the majority of thin disc tteucci & Cescutti 2006; Lanfranchi & Matteucci 2010). stars but at a time when the surface density was lower; or they could n Fig. 1(b), Romano highest we show 2013; results Micali,Matteucci&Romano for metallicities! two models with a2013, fixedtodark mention some have formed [Mg/Fe] at radii tail R of > ar distribution (where the of stars surface formed density at a single is alsoepoch; they tter potential of theofmost 10 9 M recent (Dwarf) versions) and 10 and 12 of M local (MW). dwarfs We model (e.g. Lanfranchi, lower) andcould later migrated have formed inwards at the(e.g. same Sellwood location & asbinney the majority 2002; of thin disc low mass Matteucci case with& acescutti star formation 2006; efficiency Lanfranchi of& ν Matteucci = 0.1 Gyr 1 2010). ; Roškar et al. stars 2008a, but at 2008b; a timeminchev when theet surface al. 2012). density was lower; or they could adopt bothinν Fig. = 0.1 1(b), andwe 1 Gyr show 1 results for thefor higher two mass models case. withfor a fixed dark Recent observational have formedevidence at radii R suggests > R that (where starsthe in the surface outer disc density is also h models, matter we plot potential two curves of 10 9 showing M (Dwarf) results and for 10low- 12 M and(mw). high- We model of the MWlower) do indeed and later have migrated lower [Mg/Fe] inwards ratios (e.g. than Sellwood those in & the Binney 2002; Ia rates, the varying low mass the parameter case with aa star (seeformation equation efficiency 2 in PM04)from of ν = 0.1 Gyr inner Jz/Jc 1 ; disc (e.g. Roškar Bensby et al. et 2008a, al. 2011; 2008b; Anders Minchev et al. et 2014; al. 2012). Bergemann 5 to 0.18, werespectively. adopt both ν = 0.1 and 1 Gyr 1 for the higher mass case. et For al. 2014). Recent ThisFigure could observational 6. imply The vertical radial evidence migration. component suggests Alternatively, ofthat thestars specific in the orbital outerangular disc momentum MW asversus ado difference indeed the vertical have in the lower component radial [Mg/Fe] scalelength of theratios orbital between than energy those forinthethe R13 data he figure bothclearly models, shows we plot that twothe curves [Mg/Fe] showing ratio results slowly for low- de- and high- could be explained of the set. Ez/Ec Downloaded from http://mnras.oxfordjournals.org/ at Lu Downloaded from http://mnras.oxfordjournal
COPING WITH SAMPLE BIAS = + Ruchti et al. (2014) With biases there should be at least two at [Fe/H] > -0.8, and we see none
THE GAIA-ESO SURVEY Five year survey using ESO VLT to obtain ~100,000 spectra in the Milky Way (see Gilmore et al. 2012). kinematically unbiased! DR2 just released, much larger sample to work with. Z (kpc) Z [α/fe] [Alpha/Fe] 0.6 0.4 0.2 0.0 0.2 2 1 0 1 2 ~3000 stars 2.0 1.5 1.0 0.5 0.0 0.5 [Fe/H] 3 2 4 6 8 10 12 14 R R (kpc)
CONCLUSIONS We built a chemo-dynamical template to identify an accreted disk component detritus from late, massive mergers. Current evidence suggests the Milky Way had a quiescent merger history and a correspondingly light dark matter disk. BUT! Possible signs of an accreted disk in the Gaia-ESO Survey. Stay tuned