Abstract
The lives of galaxies are shaped by the interplay of their mass components, and hence galaxy mass dissections represent a vital contribution to the field of galaxy evolution. To determine what sets the growth rate of galaxies in terms of their stellar mass build-up, I first cross-calibrate two methods of measuring the fuel for star formation held in cold molecular gas reservoirs (constraining dust masses and dynamical masses in intermediate/additional steps). I proceed to present an application in mapping out the gas scaling relations across the galaxy population (including the efficiency with which cold gas is converted into stars), and their dependencies on galaxy properties. Using a full spectral fitting approach, I then constrain the history of star formation, and thus stellar mass evolution, over longer timescales, complemented with a careful exploration of the robustness of the procedure.In more detail, I first compare molecular gas masses derived by using as an indirect tracer either CO emission at sub-mm wavelengths, or infrared dust continuum emission. Using an updated scaling with a secondary dependency on metallicity, I find that both methodologies can be brought into agreement without any systematic offset and with a reduced scatter of ~0.1 dex, based on IRAM observations of ~80 galaxies in Stripe82 with IR coverage from WISE 22 micron and Herschel-SPIRE (250, 350, 500 micron). Using the kinematic information in the CO line emission, I further find a very encouraging agreement between the dynamical mass within a half-light radius and the sum of estimated enclosed mass components (stellar mass, dark matter content, atomic and molecular H gas), despite the simplicity of the assumed dark matter profile and halo concentration.
My dust-based calibration is then applied to the full set of ~10k galaxies with overlapping WISE+SPIRE data. We find that enhanced star-forming activity cannot solely be attributed to a larger amount of fuel, but also to a higher star-forming efficiency, consistent with literature findings based on CO-inferred gas masses. Further, we show that the efficiency increases for galaxies of smaller size with more extended bulges (at fixes mass and star formation rate). Investigating trends in the Kennicutt-Schmidt relation (linking the surface densities of molecular gas and star formation), we find that changes in structural properties (bulge fraction, size) or morphological factors (spiral arms, bars, mergers) primarily move galaxies along the relation.
Next to the more instantaneous factors impacting present-day star-forming properties, I also aim to investigate if the latter are in any way connected to the early star formation history (SFH), by using a galaxy-integrated spectro-photometric fitting procedure. To evaluate the robustness of the method, I first analyse changes in the parameter values and SFHs recovered as a function of slight variations in the settings of the procedure (MCMC fitting parameters, spectral properties), as well as adopting a galaxy-integrated vs resolved strategy. I demonstrate that, even if the exact shape of the recovered SFH can vary to some extent, several key parameters are constrained robustly in any case, including: stellar mass, present-day star formation rate, mass-weighted stellar age, as well as how far along a galaxy is into the declining phase of its SFH.
Finally, I apply the galaxy-integrated fitting routine (with the most recent episode of star formation decoupled from the early SFH) to a set of ~440 massive star-forming galaxies selected from the SDSS-IV MaNGA survey. I show that, at fixed mass, the majority of galaxies with enhanced star formation at present have already shown enhanced activity throughout most of cosmic time. Compared to their slower-growing peers, they have started forming only later in cosmic time, such that both groups reach a similar mass today. Using the recovered SFHs, I derive the scatter in the "Main Sequence" (MS) relation between stellar mass and star formation rate recovered at different lookback times. I find that stochastic fluctuations reflected in the recent burst of star formation (due to, e.g., variations in gas inflow, minor mergers, stellar feedback cycles) only play a modest part in the observed scatter (~0.12 dex out of ~0.35 dex), and the dominant drivers are instead found in Hubble timescale processes (with distinct halo properties, or a long-term differentiation in bulge formation histories as plausible candidates).
| Date of Award | 28 Apr 2021 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Stijn Wuyts (Supervisor) & Carole Mundell (Supervisor) |