The Hubble Space Telescope Cluster Supernova Survey: V. Improving the Dark Energy Constraints Above z>1 and Building an Early-Type-Hosted Supernova Sample

TL;DR

The study leverages the HST Cluster SN Survey to assemble a high-z SN Ia sample, including 10 events beyond z=1, to sharpen dark-energy constraints and investigate time variation in w. By augmenting Union2 with new HST NICMOS/ACS photometry and applying host-mass corrections, the authors derive precise cosmological parameters, finding a nearly flat ΛCDM universe with ΩΛ ≈ 0.72 and w ≈ -1.01, while constraining curvature to the percent level. The work demonstrates the added value of z>1 SNe for probing w(z) and ρDE, and outlines practical paths for future improvements using WFC3/HST, robust calibration networks, and cluster-targeted surveys to build larger, balanced high-z SN samples. Overall, the results reinforce the robustness of current dark-energy constraints and highlight strategies to further reduce systematics and exploit high-redshift SNe for time-varying dark-energy tests.

Abstract

We present ACS, NICMOS, and Keck AO-assisted photometry of 20 Type Ia supernovae SNe Ia from the HST Cluster Supernova Survey. The SNe Ia were discovered over the redshift interval 0.623 < z < 1.415. Fourteen of these SNe Ia pass our strict selection cuts and are used in combination with the world's sample of SNe Ia to derive the best current constraints on dark energy. Ten of our new SNe Ia are beyond redshift $z=1$, thereby nearly doubling the statistical weight of HST-discovered SNe Ia beyond this redshift. Our detailed analysis corrects for the recently identified correlation between SN Ia luminosity and host galaxy mass and corrects the NICMOS zeropoint at the count rates appropriate for very distant SNe Ia. Adding these supernovae improves the best combined constraint on the dark energy density ρ_{DE}(z) at redshifts 1.0 < z < 1.6 by 18% (including systematic errors). For a LambdaCDM universe, we find Ω_Λ= 0.724 +0.015/-0.016 (68% CL including systematic errors). For a flat wCDM model, we measure a constant dark energy equation-of-state parameter w = -0.985 +0.071/-0.077 (68% CL). Curvature is constrained to ~0.7% in the owCDM model and to ~2% in a model in which dark energy is allowed to vary with parameters w_0 and w_a. Tightening further the constraints on the time evolution of dark energy will require several improvements, including high-quality multi-passband photometry of a sample of several dozen z>1 SNe Ia. We describe how such a sample could be efficiently obtained by targeting cluster fields with WFC3 on HST.

Figures (11)

• Figure 1: Composite color ($i_{\mathrm 775}$ and $z_{\mathrm 850}$) images of 20 SNe Ia from the HST Cluster Supernova Survey. Each SN Ia is shown in a box of $3.2\arcsec\times3.3\arcsec$ (North up and East left). Note the redshift of SCP06E12 is uncertain, and we use the cluster redshift as a guide.
• Figure 2: 15 SNe Ia light curve fits by SALT2. Flux is normalized to the $z_{\mathrm 850}$-band zeropoint magnitude. ACS $i_{\mathrm 775}$, ACS $z_{\mathrm 850}$ and NICMOS F110W data is color coded in blue, green and red respectively. Note that SCP 05D06 (z=1.314) has $H$-band data from Keck AO system (orange) melbourne07a and that this data is consistent with the HST/ACS and HST/NICMOS light curve data.
• Figure 3: Diagnostics plot for the individual data sets. From left to right: irreducible sample dispersion (filled circles) and variance-weighted RMS about the best-fit model (open circles); the average sample residual from the best-fit model ($\mu_{\mathrm{measured}} - \mu_{\mathrm{model}}$) excluding and including systematic errors; and the best-fit slope of the Hubble residual (in magnitudes) versus redshift --- $\partial \mu_\mathrm{residual}/\partial z$. Note that the errors on the sample dispersion include only statistical errors and do not include possible systematic errors. The confidence intervals on the weighed RMS are obtained with Monte-Carlo simulations. The triangles in the sample residual plot show the effect of including the filter shifts discussed in Section \ref{['sec:diagnostics']}.
• Figure 4: Hubble diagram for the Union2.1 compilation. The solid line represents the best-fit cosmology for a flat $\Lambda$CDM Universe for supernovae alone. SN SCP06U4 falls outside the allowed $x_1$ range and is excluded from the current analysis. When fit with a newer version of SALT2, this supernova passes the cut and would be included, so we plot it on the Hubble diagram, but with a red triangle symbol.
• Figure 5: $\Lambda$CDM model: $68.3\%$, $95.4\%$, and $99.7\%$ confidence regions of the $(\hbox{$\Omega_m$},\hbox{$\Omega_{\Lambda}$})$ plane from SNe Ia combined with the constraints from BAO and CMB. The left panel shows the SN Ia confidence region only including statistical errors while the right panel shows the SN Ia confidence region with both statistical and systematic errors.