Combining Gaia DR, DR and Matthias Steinmetz (AIP) a preview on the full Gaia dataset Matthias Steinmetz (AIP)
a t a d h AS t 5 - TG E V & A R ase e l e r Ga lact ic D app y lica nam tion ics s Towards RAVE DR6
RAVE: 5th public data release Kunder, et al. 07 Intermediate resolution (R~7500) 57 588 stars, 50 78 spectra (DR: 8 0 stars) 9 <I< mag Database: Radial velocities Spectral morphological flags Teff, logg, [M/H] Mg, Al, Si, Ti, Ni, Fe Line-of-sight Distances > 50 km/s 0 50 km/s -0 0 km/s -50-0 km/s < -50 km/s Photometry: MASS, APASS Proper motions: UCAC5, PPMX, PPMXL, Tycho-, TGAS Selection function (Wojno et al 07) Distances with TGAS priors (McMillan et al 07) Nov 06 / Gaia Data Workshop Heidelberg
RAVE - some new developments major clean-up of the data base, recovered almost 50% of poor fields revised distances for low metallicity revised metallicity for supersolar abundances more calibration targets calibration based on astroseismology from K (red giants) IR flux method Teff distances with TGAS priors, twin distances
DR5 and TGAS Cross matches with TGAS Credits: Maarten Breddels, Kristin Riebe, RAVE team Visualisation tool: vaex Data: Gaia GDR, TGAS, full catalogue and RAVE DR5 RAVE DR5: 5,600 LAMOST DR:,00 5 GALAH DR: 8,500 APOGEE DR:,700
DR5 - complimentary catalogue: RAVE-on The Astrophysical Journal, 80:59 (9pp), 07 May The Astrophysical Journal, 80:59 (9pp), 07 May Casey et al. Casey et al. Figure. Detailed chemical abundances in the fourth internal data release from the Gaia-ESO survey compared to this work. The number of stars shown in each panel is indicated, and the bias and rms deviations are shown. Stars are colored by the S/N of the RAVE spectra. Figure 7. Effective temperature Teffg,and surface gravity logwith g fortherave starscalibration after combining labels the main sequence and and giantkunder star models. Only results Figure 5. Stellar parameter (Teff, log [Fe/H]) comparison literature sources usedfrom by Kordopatis et al. (0) et al. (07). Stars are meeting ours/n quality constraints are shown 5). The top panels show logarithmic andindication bins in the three panels are colored the colored by the of the RAVE spectra. Note(see thatsection this comparison is three for illustrative purposes only: density, it is not an of bottom independent agreement with thebyliterature median metallicity in each bin. because some metal-poor stars in this literature sample were used in the construction of our training set (see the text for details). before increasing at lower metallicities. If we consider all stars, the spectra is 50 pixel, at which point our abundance precision smallest abundance rms we see with respect to Gaia-ESO is is about 0.07 dex, varying a few tenths of a dex between 0.6 dex forelements. [Ca/H] and [Al/H]. The increasing rms at low different metallicity is likely a consequence of multiple factors, namely, inaccurate abundance labels for metal-poor stars (Section.);.. External Validation only weak, blended lines being available in RAVE, which cease Comparison with RAVE to be visible in hot... and/or metal-poor stars; anddr to a lesser extent, low S/N for those particular stars being compared. We cross-matched our results against the Unfortunately, official fourth not RAVE all of these represented quoted in point each data factors releaseare (Kordopatis et by al. the 0) as anerrors initial label. thesecomparison reasons, although it affects small number of of For external (Figure 0). In only ordera to provide a fair comparison, we only show when stars that meet a number of quality stars, we recommend caution using individual abundances flags in both samples. Our constraints require that the S/N... Comparison with the RAVE DR Calibration Sample significant effect on the detailed abundance or spectrothe fourthdistance RAVE determinations data release made use of a number photometric between these studies of (Binney et al. 0). high-resolution studies to verify the accuracy of their derived stellar atmospheric parameters. These samples include main... Comparisons with Reddy, Bensby, and Valentifocus & Fischer sequence stars and giant stars, with a particular to include metal-poor stars to identify (and correct) any deviations at low We searched the literature for studies that overlap with metallicities. We refer reader to Kordopatis et al. (0) RAVE, and which base the their analysis on high-resolution, high for the full compilation of four literature sources. the stellar S/N spectra. We found notable studiesalthough with a sufficient atmospheric parameters in this from multiple level of overlap: the Milky Waycompilation disk studiescome by Reddy et al. (heterogeneous) sources,et we find generally agreement (00, 006) and Bensby al. (0), as well as good the Valenti & Fischer (005) work(figure on exoplanet host star candidates. These with these works 5). However, we note that some studies perform a careful (manual; expert) analysis using 6 The Cannon based data driven model (Casey, RAVE et al, 07) Red giants matched to APOGEE stellar parameters subgiants & main sequence matched to K/EPIC Teff,, O, Mg, Al, Si, Ca, Fe and Ni Ansatz can be applied to full Gaia RVS spectral data set
DR5 distances & parallaxes with TGAS priors complement Binney-Burnett (00) distance pipeline with TGAS priors (McMillan & RAVE, 07) DR5 parallaxes are all giants dwarfs TGAS 50000 0000 overestimated for hot dwarfs (Teff > 5,500 K) 0000 0000 0000 0 000 000 0000 8000 6000 000 000 0 0.0 8 underestimated for giants with < DR5 0.5 P. J. McMillan et al..0 $ [mas].5.0 hot dwarfs can be improved by using temperatures derived using IR flux method 6 Figure. Histograms of the quoted random parallax uncertainties ( $) 0.5 0.6 0.6-0.099 Giants Cool Dwarfs Hot Dwarfs Mean Mean 0.067 from TGAS and those from RAVE DR5 for stars common to 0.0 the two cat0.5 0.5 59 stars 8586 stars 0 stars Std Dev 0.867 Std Dev 0.98 0.5 alogues. We show histograms of the uncertainties for all stars (solid), and 0. 0. 0.0 separately for giants ( <.5) and dwarfs (.5). The y-axis 0.5 0. the number of stars per bin, 0. gives and there are 0 bins in total in both cases. 0.0 The0. cut-o at mas for the TGAS0.parallaxes is due to a filter applied by the 0.5 Gaia consortium to their DR. For RAVE sources we make the standard 0.0 0. 0. cuts to the catalogue described in Kunder et al. (07). TGAS0.05 parallaxes 0.00 are0.0more for0.0dwarfs, for precise 0 than 0not RAVE s but necessarily giants. IRF M TGAS IRF M TGAS P. J. McMillan et al. 0.6 Mean 0.56 Std Dev 0.985 0.5 0.6 0.5 Giants Mean -0.05 6 stars Std Dev 0.868 0.5 0. Cool Dwarfs Mean 0.056 5 stars Std Dev 0.850 0.0 0.5 Hot Dwarfs Mean 0.07 7859 stars Std Dev 0.95 0.0 0. 0.5 0. 0. 0.0 0. 0. 0. 0. 0.0 0.0 0.5 0.0 0 IRF M TGAS 7 0.05 0 DR5 0 DR5 0.00 0 DR5
DR5 distances & parallaxes with TGAS priors combined DR5+TGAS distances more accurate than either determination in isolation P( s/s) All Giants Dwarfs RAVE improve DR5 distance uncertainty by factor, TGAS by. ( for giants) Distances to RAVE-TGAS stars 0 5 P( s/s) RAVE + TGAS 0.0 0. 0. 0. 0. 0.5 0.6 0.7 0.8 s/s RAVE 8000 7000 6000 5000 000 T e 0 5 RAVE + TGAS 8000 7000 6000 5000 000 T e 0.6 0. 0. 0.0 s/s P( $) [mas ] derive ages for RAVE stars, many.0 RAVE + TGAS with relative uncertainties of 0% 0.6 or less TGAS All Giants Dwarfs 0.0 0. 80. 0.6 0.8.0 $ [mas] cumulative fraction P( / ) 0.8 0. 0. RAVE 0.0.5 RAVE.0.5 RAVE+TGAS.0.5.0 0.5 0.0 0.0 0. 0. 0.6 0.8 /
Kinematics of low Z stars in t-sne projection to identify very metal-poor stars ([Fe/H]< dex) from Matijevic et al 07 select those 55 stars with TGAS RAVE DR5 Roederer et al 0 parallax uncertainties better than 0% continuation of Ruchti & RAVE (0) to lower Z About 5% of the RAVE- identified very low metallicity stars in the solar neighborhood have disk-like kinematics Kunder & RAVE, in prep 9
unction of < VR>/ R rms), with -poor bins l standard rs radially RAVE-TGAS (cf. Wojno et al. 07). We find negative gradients in < VR>/ R for both young and old stars, with metal-rich bins having the steepest gradient, and metal-poor bins having flatter trends. In Vφ, for young stars we find that metal-rich stars lag the LSR more than metal-poor stars. Age, kinematics, and chemical correlations in RAVE Sample selection: ~ 0 000 turnoff stars from RAVE DR5 Bin Bin Bin Bin rnoff wn by note for a unger than efore bed in [M/H], using a ut age, Bin 5 Bin 6 Bin 7 Bin 8 B. Orbital parameters Kinematic trends in young and old stars Orbits are integrated using the galpy python package (Bovy 05). For our young sample, we find they have mostly circular orbits. Young, metal-rich stars have slightly more eccentric orbits with smaller guiding radii than young, metal-poor stars. For old stars, we find they have larger eccentricities, with broader guiding radii distributions that do not depend on metallicity. 0 Jennifer Wojno & RAVE, in prep 5
Age, kinematics, and chemical correlations in RAVE old young Negative < V R >/ R, steeper than measured previously for RAVE (cf. Siebert et al. 0, Williams et al. 0), see also talk by Ismael Carrillo on Friday Indicates presence of bar (Monari et al. 0) and/or spiral arms (Faure et al. 0) Young, metal-rich stars lag LSR, brought to solar neighborhood by radial migration (blurring) Jennifer Wojno & RAVE, in prep
towards DR6 (scheduled for summer/fall 08) Some house keeping publication of spectra improved chemical elements Stellar parameters Credits: Maarten Breddels, Kristin Riebe, RAVE team Visualisation tool: vaex Data: Gaia GDR, TGAS, full catalogue and RAVE DR5 using Gaia DR priors enhanced calibrations
towards DR6: stellar parameters with Gaia priors stellar parameters from the reverse distance pipeline (McMillan & RAVE 07) to be extended to Gaia DR distances as soon as they are available DR5 TGAS priors [Fe/H] SNR > 0 58,70 stars [Fe/H] Teff Teff
towards DR6: stellar parameters with Gaia priors stellar parameters from the reverse distance pipeline (McMillan & RAVE 07) to be extended to Gaia DR distances as soon as they are available DR5 TGAS priors [Fe/H] SNR > 0,58 stars [Fe/H] Teff Teff
towards DR6: stellar parameters with Gaia priors stellar parameters from the reverse distance pipeline (McMillan & RAVE 07) to be extended to Gaia DR distances as soon as they are available DR5 [Fe/H] SNR > 70,55 stars [Fe/H] TGAS priors Teff 5 Teff
towards DR6: stellar parameters with Gaia priors stellar parameters from the reverse distance pipeline (McMillan & RAVE 07) to be extended to Gaia DR distances as soon as they are available DR5 TGAS priors [Fe/H] SNR > 00 5, stars [Fe/H] Teff 6 Teff
Summary 6 P. J. McMillan et al. RAVE DR5: currently the largest overlap with TGAS revised DR5+TGAS distances low metallicity stars with disk-like kinematics structure Figure. The location of stars with small fractional age uncertainties in the c deeper insight on origin of velocity in the J band, M, have been corrected for extinction using the most likely age uncertainties less than 0 percent, the central figure those with age uncert in the local disk The number density indicated by the colour bar corresponds to the numbers o J (J K s )0. Unsurprisingly, the smallest fractional age uncertainties are for st better understanding of systematic effects in various catalogues young population stronger affected than old ones many techniques developed for RAVE can be applied to low S/N Gaia data DR6 in the making 7