McLure, Ross

Dunlop, James

Carnall, Adam Christopher

Science and Technology Facilities Council (STFC)

2019-10-08T09:39:47Z

2019-10-08T09:39:47Z

2019-11-26

This thesis presents several related analyses designed to understand the star-formation histories (SFHs) and quenching mechanisms of massive quiescent galaxies across cosmic time. More generally, it contains research directed at sophisticated modelling and Bayesian fitting of galaxy spectra. I firstly present Bayesian Analysis of Galaxies for Physical Inference and Parameter EStimation, or Bagpipes, a new, publicly available Python code that can be used to rapidly generate complex model galaxy spectra and to fit these to arbitrary combinations of spectroscopic and photometric data. I then perform a detailed analysis of the SFHs of a sample of 9289 quiescent galaxies from UltraVISTA with stellar masses, M∗> 1010M⊙ and observed redshifts from 0:25 < z < 3:75. The majority of these galaxies exhibit SFHs that rise gradually then quench relatively rapidly, over 1-2 Gyr. This behaviour is consistent with recent cosmological hydrodynamic simulations, where AGN-driven feedback in the low-accretion (jet) mode is the dominant quenching mechanism. At z > 1, I also find a class of objects with SFHs that rise and fall very rapidly, with quenching timescales of < 1 Gyr, consistent with quasar-mode AGN feedback. Finally, at z < 1, I find a population with SFHs that quench more slowly than they rise, over > 3 Gyr, consistent with other such analyses in the local Universe. I confirm the trend towards earlier formation with increasing stellar mass (downsizing) at fixed observed redshift, and a trend towards more rapid quenching at higher stellar masses. I then present a general investigation of the use of parametric SFH models in spectral fitting analyses. Parametric models for galaxy SFHs are widely used, though they are known to impose strong priors on physical parameters, with consequences for measurements of the galaxy stellar-mass function, star-formation- rate density (SFRD) and star-forming main sequence (SFMS). I investigate the effects of the exponentially declining, delayed exponentially declining, lognormal and double power law SFH models. I demonstrate that each of these models imposes strong priors on specific star-formation rates (sSFRs), potentially biasing the SFMS, and also imposes a strong prior preference for young stellar populations. I show that stellar mass, SFR and mass-weighted age inferences from high-quality mock photometry vary with the choice of SFH model by at least 0.1, 0.3 and 0.2 dex respectively. However the biases with respect to the true values depend more on the true SFH shape than the choice of model. I also demonstrate that photometric data cannot discriminate between SFH models, meaning it is important to perform independent tests to find well-motivated priors. In response to this I finally fit a low-redshift, volume-complete sample from the Galaxy and Mass Assembly (GAMA) Survey with each model. I demonstrate that the inferred stellar masses and SFRs at redshift, z ~ 0:05 are consistent with other analyses. However, the inferred cosmic SFRDs peak at z ~ 0:4, approximately 6 Gyr later than direct observations suggest, meaning that mass-weighted ages are significantly underestimated. This makes the use of parametric SFH models for understanding mass assembly in galaxies challenging. I finally present a Bayesian full-spectral-fitting analysis of 75 massive (M∗> 1010:3M⊙) UVJ-selected galaxies at redshifts of 1:0 < z < 1:3, combining extremely deep rest-frame ultraviolet spectroscopy from VANDELS with multi-wavelength photometry by the use of a sophisticated physical plus systematic uncertainties model. I constrain the stellar mass vs stellar age relationship, finding a strong trend towards earlier formation with increasing stellar mass (downsizing) of 1:48+0:34 ≲0:39 Gyr per decade in mass. I show that this is consistent with other spectroscopic studies from 0 < z < 2. This places strong constraints on the AGN-feedback models used in cosmological simulations. I demonstrate that, although the relationships predicted by the Simba and IllustrisTNG simulations agree well with observations at z = 0:1, they are too shallow at z = 1, predicting an evolution of . 0:5 Gyr per decade in mass. The majority of the lowest-mass galaxies in the sample (M∗~ 1010:5M⊙) are consistent with formation in recent (z < 2), intense starburst events, with timescales of ≲ 500 Myr. A second class of objects experience extended star-formation epochs before rapidly quenching, passing through both green-valley and post-starburst phases. The most massive galaxies in the sample are extreme systems: already old by z = 1, they formed at z ~ 5 and quenched by z = 3. However, I find evidence for their continued evolution through both AGN and rejuvenated star-formation activity. To understand the detailed SFHs of these objects, similar studies must be extended to the highest redshifts.

http://hdl.handle.net/1842/36209

en

The University of Edinburgh

Carnall A. C., 2017, preprint, (arXiv:1705.05165)

Carnall A. C., et al., 2015, MNRAS, 451, L16

Carnall A. C., McLure R. J., Dunlop J. S., Dave R., 2018, MNRAS,

Carnall A. C., Leja J., Johnson B. D., McLure R. J., Dunlop J. S., Conroy C., 2019, ApJ, 873, 44

Carnall A. C., et al., 2019, preprint, 1903.11082 (arXiv:1903.11082)

Kemp T. W., Dunlop J. S., McLure R. J., Schreiber C., Carnall A. C., Cullen F., 2019, preprint, (arXiv:1903.08169)

Leja J., Carnall A. C., Johnson B. D., Conroy C., Speagle J. S., 2018, preprint, (arXiv:1811.03637)

star-formation

quenching

Bayesian Analysis of Galaxies for Physical Inference and Parameter EStimation

Bagpipes

Star-formation histories of massive quiescent galaxies

Thesis or Dissertation

Doctoral

PhD Doctor of Philosophy