Probing the evolution of Type Ia supernovae with their ejecta velocities

There is growing evidence that Type Ia supernovae (SNe Ia) are likely produced via multiple explosion channels. Understanding how different channels evolve with redshift is critical in determining their precision in measuring cosmological parameters. Previous studies indicated that SN Ia ejecta velocity is one powerful tool to differentiate between different channels. It was also suspected that the tight correlation with the host-galaxy environment could result in the evolution of SN ejecta velocities. In this work, we measure the Si II 6355 velocities from ~400 confirmed SNe Ia discovered by the Pan-STARRS1 Medium Deep Survey (PS1-MDS), and combine them with the SNe discovered by different surveys to form a large compilation of velocity measurements. For the first time, we find that the SNe Ia with faster Si II 6355 have a significantly different redshift distribution from their slower counterparts (with a p-value of 0.00008 from the K-S test), in the sense that HV SNe Ia are more likely to be found at lower redshift. The trend may suggest a strong evolution of SN Ia ejecta velocity, or imply that the SN Ia demographics (as distinguished by their ejecta velocities) are likely to vary with time. Our results also imply that the progenitor system of HV SNe Ia (and possibly some NV SNe Ia) may favor a metal-rich environment and/or scenarios of long delay time. However, we do not see a significant difference (in ~2 sigma) in Hubble residuals when splitting our sample based on the Si II 6355 velocity.
Comment: 21 pages, 14 figures, 4 tables, submitted to ApJ