Quantifying Secular Evolution Through Structural Decomposition

Transcription

1 Quantifying Secular Evolution Through Structural Decomposition University of St Andrews / ICRAR (UWA)

2 o How do structures form? bulge disk bar pseudo-bulge disk bulge??? o Are ellipticals and bulges essentially the same? o How does environment shape galaxy structure? o How is stellar mass distributed between structure? o Can structure be used to trace evolution?

3 Single-Sérsic Photometry Structural Decomposition o Single-Sérsic modelling o Multi-component modelling o Total size, mag, index o Bulge-Disk decompositions o Wavelength dep. on o Stellar mass/light breakdown structural measurements (see Kelvin et al. 2012)

4 Galaxy and Mass Assembly ~340,000 galaxies r < 19.8 mag ~310 deg2 "study structure on scales of 1 kpc to 1 Mpc" galaxy... clusters groups mergers structure

5 Volume-Limited Sample Limits: < z < 0.06 Mr < galaxies

6 Half-Light Radius RGB:Hig

7 Eyeball Classification

8 Eyeball Classification

9 Eyeball Classifications Sd SBbcd Sbc SB0a S0a E LBS z

10 Multi-Component Models M01: Single-Sérsic M02: De Vaucouleurs bulge + exponential disk M03: Sérsic bulge + exponential disk M04: Sérsic bulge + Sérsic disk

11 Elliptical: G M01: Single-Sérsic

12 Elliptical: G M02: De Vaucouleurs bulge + exponential disk

13 Elliptical: G M03: Sérsic bulge + exponential disk

14 Elliptical: G M04: Sérsic bulge + Sérsic disk

15 S0a: G M01: Single-Sérsic

16 S0a: G M02: De Vaucouleurs bulge + exponential disk

17 S0a: G M03: Sérsic bulge + exponential disk

18 S0a: G M04: Sérsic bulge + Sérsic disk

19 SBbc: G M01: Single-Sérsic

20 SBbc: G M02: De Vaucouleurs bulge + exponential disk

21 SBbc: G M03: Sérsic bulge + exponential disk

22 SBbc: G M04: Sérsic bulge + Sérsic disk

23 Model Choice How do we select the 'best' model?

24 Model Choice How do we select the 'best' model? Bayesian Information Criterion: 2 k n total goodness of fit number of free parameters number of contributing pixels

25 Model Choice How do we select the 'best' model? Bayesian Information Criterion: 2 k n total goodness of fit number of free parameters number of contributing pixels Use visual classifications as a guide: E/Sd S0a/Sbc Single-Sérsic Multi Component Lowest BIC

26 Structural Results Sérsic index

27 Structural Results Half-light radius

28 Early/Late type bulges Global Measurements Component Measurements

29 Component Mass Stellar Masses: Taylor log(stellar Mass) log(stellar Mass) Ellipticals dominate at high-mass, disks at low-mass Late-type bulges share more in common with disks than early-type bulges

30 Stellar Mass Breakdown Mass in the local Universe: Hierarchical merging Gas accretion Secular evolution ~45.8% ~47.7% ~6.5%

31 Summary Automated, fast and robust structural decomposition is essential in order to model increasingly large galaxy datasets to a high level of accuracy. Early-type bulges are well described by the Kormendy relation, whereas late-type bulges do not follow this relation early-type bulges ~ classical bulge, late-type bulges ~ pseudo-bulge The evolutionary processes of monolithic collapse/merging and gas accretion contribute roughly equal measures of stellar mass in the local universe. Secular evolutionary processes contribute ~6.5% of the total stellar mass at z < 0.06 through the creation of pseudo-bulges. Future Work Bulge-Disk-Bar decomposition (ring, secondary disk, AGN, ) Extension of the redshift baseline and imaging quality/depth HST, VST KIDS, VISTA VIKING

32 High-Redshift Modelling Bulge-Disk decomposition essential for a full understanding of galaxy structure and mass breakdown

33 'Little Blue Spheroids'

34 'Little Blue Spheroids' Initially classified as early-type single-component Closer inspection: star-forming blue Also noted as 'Little Blue Fuzzies' in Brough et al. (2011) Reassigned disk

35 2MASS - UKIDSS - VIKING Significant improvements in structural measurements when moving from previous-generation to current-generation to next-generation survey data

36 Half-Light Radius RGB:Hig

37 Half-Light Radius K-band Image

38 Half-Light Radius Model

39 Half-Light Radius Residual

40 SIGMA: Structural Pipeline

41 SIGMA: Structural Pipeline Galaxy Modelling Output Input - Catalogue - Value-added Chi2 tests Truncations μ, μe, μ0 Data 60 GB mosaics 3 regions 9 bands weight maps = ~5 TB Catalogue 167,600 objects Cutout 400'' 400'' cutout GALFIT 3 Robust self-check Object Detection Source Extractor Primary + Secondary (neighbours) PSF Creation PSFEx (Empirical 2D PSF) ~50 stars used (at least 10) R Wrapper: Source Extractor PSF Extractor CFITSIO GALFIT 3 Speed: 15 seconds per galaxy per passband per processor

42 SIGMA Structural Investigation of Galaxies via Model Analysis Imaging & Pointing Data 400'' x 400'' cutout Star identification Empirical PSF Galaxy detection Sérsic modelling Model self-check Value added results SExtractor Bertin PSFEx Bertin 2011 GALFIT3 Peng Model Fit Parameters

43 ~340,000 galaxies r < 19.8 mag ~310 deg2