spitzer survey of stellar structure
play

Spitzer Survey of Stellar Structure in Galaxies (S 4 G, PI Sheth ) : - PowerPoint PPT Presentation

Spitzer Survey of Stellar Structure in Galaxies (S 4 G, PI Sheth ) : structural decompositions Eija Laurikainen Univ. of Oulu; FINCA, Univ. of Turku Decomposition team: Heikki Salo Eija Laurikainen Jarkko Laine Sebastien Comer on Dimitri


  1. Spitzer Survey of Stellar Structure in Galaxies (S 4 G, PI Sheth ) : structural decompositions Eija Laurikainen Univ. of Oulu; FINCA, Univ. of Turku Decomposition team: Heikki Salo Eija Laurikainen Jarkko Laine Sebastien Comer´ on Dimitri Gadotti To appear in Salo et al. (2013)

  2. Sample: • Radial velocity in radio ν < 3000 km s − 1 (equivalent to D=41 Mpc) • Integrated blue magnitude m Bcorr < 15.5 mag • Angular diameter D 25 > 1’ • Galactic latitude > 30 0 ⇒ N=2352 galaxies (+ 62 in the same frames) Data: • IRAC images at 3.6 and 4.5 µm wavelengths • Map the galaxies up to 1.5 x D 25 • Surface density limit: µm (AB) 3 . 6 ∼ 27 mag arcsec − 2 • Pixel resolution (3.6 µm ) = 0.75” (for sampling 1.2”) 1

  3. Decomposition approach: Use GALFIT (Peng et al. 2010) • Levenberg-Marguadt algorithm: [ OBS ( x, y ) − MODEL ( x, y )] 2 χ 2 = 1 X X (1) σ ( x, y ) 2 N x y ⇒ minimize χ 2 between observed and model images ⇒ model image convolved with the image PSF • Use generalized ellipses for isophotes (C=0): 1 y ′ − y 0 C +2 ! ˛ ˛ C +2 | x ′ − x 0 | C +2 + r ( x ′ , y ′ ) = ˛ ˛ (2) ˛ ˛ q ˛ ˛ 2

  4. Fitting functions: • S´ ersic function (bulge) : ( r/r e ) 1 /n − 1 “ h i” Σ( r ) = Σ e exp − κ (3) • Exponential function (disk): Σ( r ) = Σ o exp( − r/h r ) (4) • Exponential function (edge-on disk): r x „ r x « sech 2 ( r z /h z ) , Σ( r x , r z ) = Σ o K 1 (5) h r h r • Modified Ferrers function (bar, β =0, α =2): 1 − ( r/r out ) 2 − β ˜ α  ˆ Σ 0 r < r out Σ( r ) = (6) 0 r ≥ r out • PSF-convolved point source (unresolved central source) 3

  5. What do we need: • Mosaicked science images (no none pixel values; PI) • PSF-image (oversampled, composite over rot. angles; Jarret) • Sigma-images (created from PI weight images) • Mask-images (created with SExtractor, edited by hand) • Orientation parameters, sky background, galaxy center ⇒ important to use those given with the decompositions Running the decompositions: • Use GALFIDL (Salo) ⇒ creation of σ images ⇒ semiautomatic filling of the input files ⇒ visualization of the outcome 4

  6. Sky background (P4): ⇒ Decompositions only for galaxies with large enough FOV 5

  7. Orientation parameters (P4): 6

  8. Comparison of LEDA and P4 orientations: LEDA P4 NGC7098 LEDA: pa,inc= 61.9 59.2 S4G: pa,inc= 72.2 55.0 200 200 200 100 100 100 0 0 0 -100 -100 -100 -200 -200 -200 -200 -100 0 100 200 -200 -100 0 100 200 -200 -100 0 100 200 /home/heikki/S4G_BARS.dir/s4g_barlen_flatimage_100511.dir/NGC7098_flat heikki@hiisi.oulu.fi Tue May 10 15:26:50 2011 250 250 ALL ALL BT<15 BT<15 200 200 150 150 Number Number LEDA b/a P4 b/a 100 100 More reasonable 50 50 7 0 0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 b/a (Leda) b/a (Pipeline 4)

  9. What kind of decompositions: (1) Automatic single S´ ersic fits (2) Automatic S´ ersic+expo fits (BD) (3) Individually checked bulge/disk/bar/nucleus fits (BDbarn): BD / BDn BDD / BDDn BDbar / BDbarn DDbar / DDbarn Dbar / Dbarn DD / DDn D / Dn Similar edge-on models 8

  10. How to show the profiles : an exponential disk 9

  11. Exponential disk + Ferrers bar ⇒ highlights non-axisymmetric structures; (Laurikainen et al. 2005) ⇒ important to look at: the profile and the model image Fitting based on χ 2 , but in final model selection not critical 10

  12. Examples of the decompositions: NGC 3938 [SA(s)c]: bulge/disk (BD) 11

  13. NGC 0628 [SA(s)c]: bulge/disk (BD) ⇒ PSF added 12

  14. NGC 1357 [(R’L)SA(rs)0/a]: bulge/disk/disk (BDD) 13

  15. NGC 2684 [SA(rs)b]: bulge/disk (Dn) ⇒ S´ ersic / PSF 14

  16. NGC 3266 [(RL)SB0 o ] (BDbar) Bars in bright galaxies: 15

  17. NGC 0936 [(L)SB(rs,bl)0 + ] (BDbar) Bars in bright galaxies: 16

  18. Bars in bright galaxies: NGC 5101 [(R 1 R 2 ’)SB(rs,nl?)0] (BDbar) 17

  19. Bars in very late-types: NGC 1688 [SB(s)dm] (Dbarn) 18

  20. Bars in very late-types: IC 3021 [dSB(s)0/a] (Dbarn) 19

  21. Bars in very late-types: NGC 3517 [SB(s)m] (dbar) 20

  22. Pure disk galaxies: NGC 3377A [Im] (D) 21

  23. Pure disk galaxies: NGC 0723 [(R’)SA(s)bc] (DD) 22

  24. Final multi-component fits 250 Edge-on’s excluded bd 200 More than 20 types of ’final models’ dbar d 150 (permutations of ’b’,’d’,’bar’,’n’...) 100 50 0 -5 0 5 10 1.0 Visual classifications from dbar from Buta et al. (2013) 0.8 bd d 0.6 Fraction 0.4 0.2 0.0 -5 0 5 10 Hubble T 23 /home/heikki/P4_JUTTU_PRO/fathi_s4g_compare2_bulge_fraction heikki@pc091079 Wed Oct 16 13:20:46 2013

  25. 1.2 1.0 Strong function of galaxy mass 0.8 d bd Fraction dbar 0.6 0.4 0.2 0.0 6 7 8 9 10 11 12 log 10 M* 1.2 1.0 Strong function of color 0.8 d bd Fraction dbar 0.6 0.4 0.2 0.0 -2 -1 0 1 2 3 m B -m 3.6 (AB) color 24 /home/heikki/P4_JUTTU_PRO/fathi_s4g_compare2_bulge_fraction_vs_mass heikki@pc091079 Wed Oct 16 13:23:51 2013

  26. Multi-component vs BD fits 6 2-Component BD ’Bulge’+’Disk’ galaxies: Final model L2007 4 Inclusion of bar-component Bulge Sersic n crucial in order to get correct 2 ← bulge S´ ersic index 0 -5 0 5 10 Hubble T 2-Component BD ← Bulge effective radius/ Final model L2007 disk scale length 1.0 R e /h r Multi-component fits in agreement with our NIRS0S /OSUBSGS fits 0.1 (Laurikainen et al. 2007) 25 -5 0 5 10 Hubble T /home/heikki/P4_JUTTU/P4_FIGURES.dir/p4fig_model_comparisons_talk_re_hr heikki@pc091079 Tue Mar 5 01:55:37 2013

  27. An other way of showing the effect of bars: non-barred ( 320) barred ( 213) Final: bar/nonbar 8 8 8 BD BD bar Final Final non-bar 6 6 6 4 4 4 n n n 2 2 2 0 0 0 -5 0 5 10 -5 0 5 10 -5 0 5 10 Hubble T Hubble T Hubble T non-barred barred Final: bar/nonbar 10.0 10.0 10.0 BD BD bar Final Final non-bar R e (kpc) R e (kpc) R e (kpc) 1.0 1.0 1.0 0.1 0.1 0.1 -5 0 5 10 -5 0 5 10 -5 0 5 10 Hubble T Hubble T Hubble T /home/heikki/P4_JUTTU_PRO/p4fig_finalsample_model_comparisons_bar_nobar heikki@pc091079 Mon Aug 5 12:05:42 2013 ⇒ automatic BD-fitting: does not work for barred galaxies ⇒ final models: similar in barred/non-barred galaxies 26

  28. Bars in decompositions/classifications 1.2 Bar in decomposition SB SB + SAB 1.0 IB IB + IAB 0.8 Fraction 0.6 0.4 0.2 0.0 -5 0 5 10 T (Buta) /home/heikki/P4_JUTTU_PRO/p4fig_finalsample_model_fractions_korea_bar_tbuta heikki@pc091079 Wed Oct 16 12:34:07 2013 ⇒ similar distributions among Hubble types ⇒ more bars in classification: very faint bars not fitted 27

  29. Bar fractions and galaxy mass: 1.2 Bar in decomposition SB SB + SAB 1.0 IB IB + IAB Barazza et al. OPT 0.8 Fraction 0.6 0.4 0.2 0.0 6 7 8 9 10 11 12 log 10 M* /home/heikki/P4_JUTTU_PRO/p4fig_finalsample_model_fractions_korea_bar_mstar heikki@pc091079 Wed Oct 16 12:34:07 2013 ⇒ f bar strong function of galaxy mass ⇒ but f bar significant also in faint galaxies ⇒ consistent with Barrazza et al. (2008) for log 10 M > 9.5 28

  30. IRSA IMPORTANT MANY P4 MODELS DONE BEFORE FINAL P1 IMAGES ⇒ need to be sure to use correct mask-files etc. All input data needed in decompositions: (“IDE” = galaxy designation) IDE.phot.1_nonan.fits IDE.1.finmask_nonan.fits IDE.phot.1_sigma.fits_ns PSF-1.composite.fits Output files from 3 different decompositions: “MODEL ” IDE_MODEL.outgal IDE_MODEL.outgal_subcomps.fits IDE_MODEL.outgal_profile.pdf IDE_MODEL.outgal_residual.pdf 29

  31. Pictorial web-database Main page List of 2331 S4G galaxies http://www.oulu.fi/astronomy/S4G P4/MAIN101112.dir/main101112.html 30

  32. INDEX-pages for 100 galaxies near NGC1415 Iconic view of available data 1 2.cm 2 3 4 5 6 7 1) P1 image 2) clipped mask image 3) clipped image 4) Deprojected image 5) Elliptical isophotes 6) Sky background 7) Decomposition 31 Clicking on galaxy name (’NGC1415’) ⇒ opens a decomposition page

  33. Conclusions: Suitable: • Properties of bulges and disks in a large galaxy mass range • Easy to do new decompositions Limitations: • Pixel resolution (pix=0.75”, final FWHM ∼ 2”) ⇒ small bulges not always possible to separate from PSF ⇒ not very good for nuclear bars/rings/lenses What kind of decompositions depends on the science goal! IRSA: now 200 gal., the rest Dec 2013 32

Recommend


More recommend