A score-based diffusion generative model on deep infrared galaxy photometry yields a star formation rate density peaking at z=1.3 and shows distinct non-parametric star formation histories plus AGN activity peaking during the quenching transition of massive galaxies.
Parametrising Star Formation Histories
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
We examine the star formation histories (SFHs) of galaxies in smoothed particle hydrodynamics (SPH) simulations, compare them to parametric models that are commonly used in fitting observed galaxy spectral energy distributions, and examine the efficacy of these parametric models as practical tools for recovering the physical parameters of galaxies. The commonly used tau-model, with SFR ~ exp(-t/tau), provides a poor match to the SFH of our SPH galaxies, with a mismatch between early and late star formation that leads to systematic errors in predicting colours and stellar mass-to-light ratios. A one-parameter lin-exp model, with SFR ~ t*exp(-t/tau), is much more successful on average, but it fails to match the late-time behavior of the bluest, most actively star-forming galaxies and the passive, "red and dead" galaxies. We introduce a 4-parameter model, which transitions from lin-exp to a linear ramp after a transition time, which describes our simulated galaxies very well. We test the ability of these parametrised models to recover (at z=0, 0.5, and 1) the stellar mass-to-light ratios, specific star formation rates, and stellar population ages from the galaxy colours, computed from the full SPH star formation histories using the FSPS code of Conroy et al. (2009). Fits with tau-models systematically overestimate M/L by ~ 0.2 dex, overestimate population ages by ~ 1-2 Gyr, and underestimate sSFR by ~ 0.05 dex. Fits with lin-exp are less biased on average, but the 4-parameter model yields the best results for the full range of galaxies. Marginalizing over the free parameters of the 4-parameter model leads to slightly larger statistical errors than 1-parameter fits but essentially removes all systematic biases, so this is our recommended procedure for fitting real galaxies.
verdicts
UNVERDICTED 4representative citing papers
Optical imaging and BAGPIPES SED fitting of eight FXTs yields candidate hosts consistent with WD-IMBH TDEs or BNS mergers for most events, with one reclassified as a Galactic flare and evidence for diverse origins.
Panchromatic SED modeling yields SFRs with smaller offset and scatter than optical-only fits for starburst to post-starburst galaxies, while Prospector AGN torus models distinguish AGN but underpredict luminosities by an order of magnitude.
Prospector is a flexible code for Bayesian inference of stellar population parameters from multi-wavelength photometry and spectroscopy via forward modeling and posterior sampling.
citing papers explorer
-
pop-cosmos: Star formation over 12 Gyr from generative modelling of a deep infrared-selected galaxy catalogue
A score-based diffusion generative model on deep infrared galaxy photometry yields a star formation rate density peaking at z=1.3 and shows distinct non-parametric star formation histories plus AGN activity peaking during the quenching transition of massive galaxies.
-
Optical observations of candidate host galaxies of eight fast X-ray transients
Optical imaging and BAGPIPES SED fitting of eight FXTs yields candidate hosts consistent with WD-IMBH TDEs or BNS mergers for most events, with one reclassified as a Galactic flare and evidence for diverse origins.
-
Sparks II: Panchromatic SED modeling and galaxy physical properties across the starburst to post-starburst sequence
Panchromatic SED modeling yields SFRs with smaller offset and scatter than optical-only fits for starburst to post-starburst galaxies, while Prospector AGN torus models distinguish AGN but underpredict luminosities by an order of magnitude.
-
Stellar Population Inference with Prospector
Prospector is a flexible code for Bayesian inference of stellar population parameters from multi-wavelength photometry and spectroscopy via forward modeling and posterior sampling.