Reconstructing the star formation histories of galaxies with the dense basis method

The star formation histories (SFHs) of galaxies have long been approximated by simple parametric forms while estimating quantities like stellar masses, star formation rates and ages. However, there is considerable diversity seen among galaxy SFHs in cosmological simulations, and individual SFHs are variable across a range of timescales corresponding to the various physical processes that govern star formation. With higher S/N multiwavelength Spectral Energy Distributions (SEDs) from broadband photometry and sophisticated analysis tools, we develop the Dense Basis SED fitting method (Iyer & Gawiser 2017) to reconstruct the SFHs of individual galaxies with uncertainties. Applying this method to CANDELS data, we reconstruct the SFHs of nearly 50,000 galaxies across a wide range of redshifts. An updated version of the method (Iyer et al. 2019) uses Gaussian Processes to create smooth SFHs that are independent of any choice of functional form, with a flexible number of parameters that recover the maximum amount of information from individual SEDs. Using this method, we estimate the number and duration of major star formation episodes in a galaxy's past, in addition to quantifying the evolution of galaxy SFHs with mass, morphology, and redshift. The distribution of SFHs at a particular epoch constrains feedback and wind strengths modeled in simulations of galaxy evolution. (Iyer et al. 2018) uses SFHs as trajectories in SFR-M* space to probe the previously inaccessible low-mass, high-redshift regime of the SFR-M* correlation. This new technique of SFH reconstruction allows us to probe a wide range of quantities that were previously inaccessible through SED fitting, creating new possibilities of probing galaxy formation and evolution at high redshifts.Ph.D.Includes bibliographical reference