Improved Dark Energy Constraints from ~100 New CfA Supernova Type Ia Light Curves

TL;DR

This work expands the nearby Type Ia supernova sample with the CfA3 set to tighten constraints on the dark energy EOS parameter $w$ within a flat universe when combined with BAO priors. It tests four light-curve fitters (SALT, SALT2, MLCS2k2 with $R_V=3.1$, and MLCS2k2 with $R_V=1.7$) to diagnose systematic differences and environment-dependent luminosity. The Constitution set, formed by Union+CfA3, yields $1+w=0.013^{+0.066}_{-0.068}$ (with a $0.11$ mag systematic), consistent with a cosmological constant, though systematics now dominate the error budget. The paper emphasizes the need for improved photometric accuracy, retraining fitters with CfA3, incorporating near-infrared data to disentangle host reddening, and considering host-galaxy environments to reduce systematic uncertainties in future dark energy constraints.

Abstract

We combine the CfA3 supernova Type Ia (SN Ia) sample with samples from the literature to calculate improved constraints on the dark energy equation of state parameter, w. The CfA3 sample is added to the Union set of Kowalski et al. (2008) to form the Constitution set and, combined with a BAO prior, produces 1+w=0.013 +0.066/-0.068 (0.11 syst), consistent with the cosmological constant. The CfA3 addition makes the cosmologically-useful sample of nearby SN Ia between 2.6 and 2.9 times larger than before, reducing the statistical uncertainty to the point where systematics play the largest role. We use four light curve fitters to test for systematic differences: SALT, SALT2, MLCS2k2 (R_V=3.1), and MLCS2k2 (R_V=1.7). SALT produces high-redshift Hubble residuals with systematic trends versus color and larger scatter than MLCS2k2. MLCS2k2 overestimates the intrinsic luminosity of SN Ia with 0.7 < Delta < 1.2. MLCS2k2 with R_V=3.1 overestimates host-galaxy extinction while R_V=1.7 does not. Our investigation is consistent with no Hubble bubble. We also find that, after light-curve correction, SN Ia in Scd/Sd/Irr hosts are intrinsically fainter than those in E/S0 hosts by 2 sigma, suggesting that they may come from different populations. We also find that SN Ia in Scd/Sd/Irr hosts have low scatter (0.1 mag) and reddening. Current systematic errors can be reduced by improving SN Ia photometric accuracy, by including the CfA3 sample to retrain light-curve fitters, by combining optical SN Ia photometry with near-infrared photometry to understand host-galaxy extinction, and by determining if different environments give rise to different intrinsic SN Ia luminosity after correction for light-curve shape and color.

Figures (21)

• Figure 1: Hubble diagram and residuals for the Constitution sample fit by SALT. The bottom panel shows the new CfA3 SN Ia in red and the Union sample in black. The residuals are with respect to a universe without dark energy, $\Omega_M$$=0.27$ and $\Omega_\Lambda$$=0$. The best-fit cosmology is plotted in the residuals panels. The large scatter at high redshift is one of the main weaknesses of the conventional approach to calculating distance moduli from the SALT light-curve fits.
• Figure 2: Hubble diagram and residuals for MLCS17. The new CfA3 points are shown in red and the OLD and High-z points are in black. MLCS17 (and MLCS31) has a smaller dispersion at high redshift than SALT (and SALT2). The nearby MLCS17 distances are larger than in SALT, making the High-z distances smaller relative to a matter-only universe and resulting in a greater value of $1+w$. This effect is seen in how the MLCS17 best-fit cosmology line is closer to the axis.
• Figure 3: Contour plots of $\Omega_\Lambda$ vs. $\Omega_M$ for $1+w=0$ for SALT, with no assumptions about flatness. The concordance cosmology ($\Omega_\Lambda$$=0.73$, $\Omega_M$$=0.27$) is shown as a dot. The top panel shows how adding the CfA3 sample considerably narrows the contours along the $\Omega_\Lambda$ axis. The bottom panel shows the combination of the SN contours with the BAO prior, with the flat-universe straight line overplotted for reference.
• Figure 4: Contour plots of $w$ vs. $\Omega_M$ in a flat universe. The concordance cosmology ($w=-1$, $\Omega_\Lambda$$=0.73$, $\Omega_M$$=0.27$) is shown as a dot in each plot. The left two plots are for SALT while the right two plots are for MLCS17. The top row of plots show how adding the CfA3 sample considerably narrows the contours along the $w$ axis. The second row of plots show the combination of the SN contours with the BAO prior.
• Figure 5: The distribution of $\Delta$ versus $z$. Faint SN Ia (high $\Delta$) are not found at higher redshifts because of magnitude-limited searches. The CfA3 sample has an effective limiting magnitude of $\sim$18.5 mag.