FLASH Talks: Improving Cosmological Utility of Type Ia Supernovae through Physics and Big Data (Jeffrey Andrews) & It Takes Two to Tango: Modeling Binary Stellar Populations in the Gravitational Wave Era (Matt Siebert)
Friday, 15 October 2021 noon — 1 p.m. MST
FLASH Talks are scientific talks for the staff at NOIRLab and the University of Arizona's Steward Observatory. For the Zoom link, please email Parker Fagrelius.
Matt Siebert, UCSC
Improving Cosmological Utility of Type Ia Supernovae through Physics and Big Data
After correcting for their light-curve shape and color, Type Ia supernovae (SNe Ia) are precise cosmological distance indicators. However, there remains a non-zero intrinsic scatter in the differences between measured distance and distances inferred from a cosmological model (i.e., Hubble residuals). We have found that Hubble residuals are correlated with different properties of SN Ia host galaxies (e.g., mass, star-formation rate, and metallicity). Cosmological analyses attempt to correct for these effects empirically without any physical knowledge of the progenitor system, potentially introducing a bias. I will present two projects designed to understand the dominant physical sources of intrinsic scatter. First, I will present Kaepora, a relational database for SN Ia observations containing 4975 public spectra of 777 SNe Ia. Using this large sample, we produce composite spectra that have been precisely controlled for phase and light-curve shape, and show that SN Ia distances can potentially be improved with the knowledge of SN ejecta velocities. Additionally, we have obtained host-galaxy spectroscopy for the Foundation and Swope combined cosmological SN sample. The Foundation/Swope survey is the largest low-z SN sample, and we have observed 363 out of 523 total SN host galaxies. The mass-metallicity relationship of our preliminary sample is consistent with the results from Pan et al. 2014 (containing a similar sample of 82 SN Ia host-galaxy spectra) and contains some of the lowest mass galaxies studied in a cosmological sample. Our sample also shows evidence for a "metallicity-step" using measurements from both the galaxy nucleus and the supernova location. With the full sample, we hope to constrain the functional form of this relationship and better understand the underlying physical cause of the empirical host-galaxy correction.
Jeffrey Andrews, Northwestern
It Takes Two to Tango: Modeling Binary Stellar Populations in the Gravitational Wave Era
With the discovery of dozens of merging black holes and neutron stars by gravitational wave observatories over the past five years, the study of the complexities of binary stellar evolution - including mass transfer, tides, and r-process nucleosynthesis - has taken on a new urgency. In this talk, I will describe the current status of modeling binary star populations as well as several critical shortcomings that my collaborators and I are working to systematically address. In particular, I will focus on how we are using modern statistical and machine learning methods using dedicated supercomputers to improve our physical models of binary star populations.