2024MNRAS.529.3806K

The environmental dependence of Type Ia supernova (SN Ia) luminosities is well established, and efforts are being made to find its origin. Previous studies typically use the currently observed status of the host galaxy. However, given the delay time between the birth of the progenitor star and the SN Ia explosion, the currently observed status may differ from the birth environment of the SN Ia progenitor star. In this paper, employing the chemical evolution and accurately determined stellar population properties of 44 early-type host galaxies, we, for the first time, estimate the SN Ia progenitor star birth environment, specifically [Fe/H]_Birth and [α/Fe]_Birth. We show that [α/Fe]_Birth has a $30.4^{\text{+10.6}}_{-10.1}{{\ \rm per\ cent}}$ wider range than the currently observed [α/Fe]_Current, while the range of [Fe/H]_Birth is not statistically different ($17.9^{\text{+26.0}}_{-27.1}{{\ \rm per\ cent}}$) to that of [Fe/H]_Current. The birth and current environments of [Fe/H] and [α/Fe] are sampled from different populations (p-values of the Kolmogorov-Smirnov test <0.01). We find that light-curve fit parameters are insensitive to [Fe/H]_Birth (<0.9σ for the non-zero slope), while a linear trend is observed with Hubble residuals (HRs) at the 2.4σ significance level. With [α/Fe]_Birth, no linear trends (<1.1σ) are observed. Interestingly, we find that [α/Fe]_Birth clearly splits the SN Ia sample into two groups: SN Ia exploded in [α/Fe]_Birth-rich or [α/Fe]_Birth-poor environments. SNe Ia exploded in different [α/Fe]_Birth groups have different weighted-means of light-curve shape parameters: 0.81 ± 0.33 (2.5σ). They are thought to be drawn from different populations (p-value = 0.01). Regarding SN Ia colour and HRs, there is no difference (<1.0σ) in the weighted-means and distribution (p-value > 0.27) of each [α/Fe]_Birth group.