Dylan Lazarus

McMaster University

Dylan is a PhD student at McMaster university working with Dr. Laura Parker. He completed his BSc in Physics and Mathematics at McGill, and transitioned to studying astrophysics for his MSc, also completed at McMaster. Dylan’s primary research interest is studying the elusive population of rejuvenating galaxies using observations. It is widely assumed that the typical evolutionary pathway of a galaxy is the transition from star-forming to quenched. However, if a galaxy’s supply of cold gas is replenished after being quenched – due to internal or environmental processes – star formation can reignite in a process known as rejuvenation.

His MSc focused on studying the gas properties of rejuvenating galaxies using an existing selection method (Cleland & McGee, 2021). This method relies on comparing UV emission, which purportedly traces star formation on ~100 Myr timescales, to Hα line emission, which traces star formation on much shorter ~10 Myr timescales. However, through the past year of his PhD studies, he has found that these indicators are largely affected by dust extinction and trace overlapping timescales. He has now developed an improved method for identifying rejuvenating galaxies that is computationally inexpensive and uses only commonly observed regions of a galaxy’s spectral energy distribution.

Using stellar population synthesis codes, he studied the contribution of stars over 20 solar masses – capable of ionizing their surrounding media – to a galaxy’s UV emission. Short times (<10 Myr) after a burst of star formation, these ionizing stars account for over 90% of the integrated UV emission of the stellar population. While this contribution decreases over time, UV and Hα emission from rejuvenating galaxies does not trace star formation on distinct timescales. Dylan has derived a relation to convert a galaxy’s Hα emission to the UV emission produced by ionizing stars, to isolate the UV emission originating from stars below 20 solar masses. This results in two SFR indicators on distinct timescales, and coupled with dust corrections, offers a significantly more reliable classification diagnostic. A paper describing this method is currently in preparation to be submitted to MNRAS.