Age- Abundance Trends in the Solar Neighborhood Diane Feuillet MPIA April 25, 2017 IAUS 330 Nice, France
GalacNc Chemical EvoluNon Alpha elements Fe type elements [O/Fe] [Na/Fe] [Mg/Fe] [Cr/Fe] [Ca/Fe] [Ni/Fe] [Ti/Fe] [Zn/Fe] [Fe/H] [Fe/H] Bensby+ 2014
How do we interpret chemical abundances? Observed chemical abundance??? History of the Milky Way Haywood+ 2013
How do we interpret chemical abundances? Observed chemical abundance??? History of the Milky Way Timing depends on local star forma:on rate Haywood+ 2013
Need to describe the full disk star Uniform forma:on rate?? APOGEE Hayden+ 2015
Need to describe the full disk star Uniform forma:on rate?? APOGEE Hayden+ 2015 Stellar Ages
Stellar Ages Ages cannot be directly measured Empirical Gyrochronology Chromospheric acnvity CN abundances Model- dependent Isochrone model matching Asteroseismic mass Mass- age relanon Requirements for APOGEE stars Works for giant stars Applicable to a large sample (> 100,000) Preserves ability to examine detailed chemical abundance pa_erns with Nme Small uncertainnes
Stellar Ages Ages cannot be directly measured Empirical Gyrochronology Chromospheric acnvity CN abundances Model- dependent Isochrone model matching Asteroseismic mass Mass- age relanon Age [Gyr] Age [Gyr] log(g) 0 1 2 3 4 1.0 2.7 4.3 6.0 7.7 9.3 11.0 1.0 2.7 4.3 6.0 7.7 9.3 11.0 6000 5500 5000 4500 4000 3500 3000 Temperature [K] V T [mag] 2 1 0 1 2 3 4 PARSEC 5 6000 5500 5000 4500 4000 3500 3000 Temperature [K]
Feuillet+ 2016 Local Sample Parallax measurements 1m+APOGEE 700 stars within 400 pc of the sun Hipparcos parallax uncertainty < 10 % (J- K) > 0.5, M H > 2 [Fe/H], Teff, log(g), M V Test sample for Gaia+APOGEE log(g) 0 1 2 1.00 0.73 7 0 7 0.33 0 3 4 5500 5000 4500 4000 3500 Temperature [K]
Feuillet+ 2016 Bayesian Likelihood f(τ, ζ) ψ(τ)φ(ζ) L(τ, ζ, m)s(τ, ζ,m)ξ(m)dm. ψ(τ) )ξ(m) )φ(ζ) Assume flat SFH in age Chabrier IMF MDF flat within σ Input parameters: [Fe/H], T eff, M V, log g Take mean of age PDF true age mean age SelecNon funcnon
Feuillet+ 2016 Hierarchical Modeling of SFH Find a more informed prior for the SFH of a sample of stars - - propose a model σ Full sample Use the full age PDFs to constrain the model parameter(s) pa ( data ) pa ( ) Li ( ) Na ( ) ( a) d. ψ(τ µ, σ) = i Gaussian + uniform SFH ( ) ( ) (1 A) (τ µ) 2 σ 2π exp + A C 2σ 2 µ) A outlier fracnon C normalized constant μ mean age σ age dispersion τ age
Model the SFH for stars with similar abundance Results in the mean age of giants currently in the solar neighborhood with the same alpha abundance Feuillet+ 2016 Hierarchical Modeling of Local Sample 0.3 ψ(τ µ, σ) = (1 A) σ 2π exp Bayesian ages ( ) (τ µ) 2 + A C 2σ 2 0.3 Hierarchical modeling [α/m] 0.1 [α/m] 0.1 0.1 1 9.5 9.0 8.5 8.0 0.1 1 9.5 9.0 8.5 8.0
Feuillet+ in prep Hierarchical Modeling of Local Sample 0.3 [α/m] 0.1 0.3 0.1 1 9.5 9.0 8.5 8.0 [α/m] 0.1 0.5 0.1 1.0 0.5 0.5 1.0 0.5 1.0 1 9.5 9.0 8.5 8.0
Feuillet+ in prep C and N mass dependence Driven by mass(temperature)- dependence of CNO cycle and internal mixing from first dredge- up [C/M] C [C/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Supports methods of MarNg+ 2016 N Must be tested outside solar neighborhood [N/M] [N/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
Feuillet+ in prep Alpha elements Individual alpha elements agree with average alpha O Consistent with previous work on solar- like stars and local dwarfs [O/M] 1 9.5 9.0 8.5 8.0 [O/M] 1.0 0.5 0.5 Difference in light and heavy alphas [Mg/M] Mg [Mg/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 [S/M] [S/M]
Feuillet+ in prep Alpha elements Individual alpha elements agree with average alpha Si Consistent with previous work on solar- like stars and local dwarfs [Si/M] 1 9.5 9.0 8.5 8.0 [Si/M] 1.0 0.5 0.5 Difference in light and heavy alphas [Ca/M] Ca [Ca/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 [Ti/M] [Ti/M]
Feuillet+ in prep Na and Al Smiljanic+ 2016 suggest internal mixing in red giant stars enhances Na in massive stars [Na/M] Na [Na/M] Al is consistent with results from solar- like stars. Na is not, perhaps there is some effect of mixing [Al/M] 1 9.5 9.0 8.5 8.0 Al [Al/M] 1.0 0.5 0.5 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 [P/M] [P/M]
Feuillet+ in prep Mn and Ni Most theorencal yields predict similar behavior for Mn and Ni [Mn/M] Mn [Mn/M] Could be some NLTE effects (see BarsNni & Bensby 2015) 1 9.5 9.0 8.5 8.0 Ni 1.0 0.5 0.5 Kobayashi+ 2006 solves Mn vs Ni difference with winds [Ni/M] [Ni/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 [Co/M] [Co/M]
Conclusions [O/M] Hierarchical modeling is a powerful tool for examining age trends for large samples of stars Clear relanon between C & N and age in the solar neighborhood Different behavior of light and heavy alphas Gaia will allow for individual age- abundance trends as a funcnon of posinon through the disk [Si/M] [Mn/M] 1 9.5 9.0 8.5 8.0 1 9.5 9.0 8.5 8.0 1 9.5 9.0 8.5 8.0
Ques:ons?
O [O/M] [O/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Mg [Mg/M] [Mg/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 S [S/M] [S/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
Si [Si/M] [Si/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Ca [Ca/M] [Ca/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Ti [Ti/M] [Ti/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
Na [Na/M] [Na/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Al [Al/M] [Al/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 P [P/M] [P/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
K [K/M] [K/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 V [V/M] [V/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Cr [Cr/M] [Cr/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
Mn [Mn/M] [Mn/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Ni [Ni/M] [Ni/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5 Co [Co/M] [Co/M] 1 9.5 9.0 8.5 8.0 1.0 0.5 0.5
GalacNc Archaeology! Stars are good tracers of GalacNc evolunon! Elemental abundances in stellar atmospheres reflect the composinon of the ISM! The composinon of the ISM changes with Nme! IN GENERAL the ISM is enriched with Nme " [Fe/H] increases! Different elements are enriched differently