Supernova Simulations and Strategies For the Dark Energy Survey

TL;DR

This paper analyzes supernova light curve simulations for the Dark Energy Survey, optimizing strategies to maximize supernova detection and redshift measurement accuracy for cosmological studies.

Contribution

It introduces a comprehensive simulation framework and evaluates survey strategies, leading to an optimized 30 sq. degree survey plan with high supernova sample purity.

Findings

Forecasts up to 4000 Type Ia supernovae in the redshift range 0.05<z<1.2

Significant improvement in high-redshift color measurements due to DES camera's red efficiency

Achieves up to 98% purity in Type Ia supernova sample with photometric identification

Abstract

We present an analysis of supernova light curves simulated for the upcoming Dark Energy Survey (DES) supernova search. The simulations employ a code suite that generates and fits realistic light curves in order to obtain distance modulus/redshift pairs that are passed to a cosmology fitter. We investigated several different survey strategies including field selection, supernova selection biases, and photometric redshift measurements. Using the results of this study, we chose a 30 square degree search area in the griz filter set. We forecast 1) that this survey will provide a homogeneous sample of up to 4000 Type Ia supernovae in the redshift range 0.05<z<1.2, and 2) that the increased red efficiency of the DES camera will significantly improve high-redshift color measurements. The redshift of each supernova with an identified host galaxy will be obtained from spectroscopic observations of the host. A supernova spectrum will be obtained for a subset of the sample, which will be utilized for control studies. In addition, we have investigated the use of combined photometric redshifts taking into account data from both the host and supernova. We have investigated and estimated the likely contamination from core-collapse supernovae based on photometric identification, and have found that a Type Ia supernova sample purity of up to 98% is obtainable given specific assumptions. Furthermore, we present systematic uncertainties due to sample purity, photometric calibration, dust extinction priors, filter-centroid shifts, and inter-calibration. We conclude by estimating the uncertainty on the cosmological parameters that will be measured from the DES supernova data.