The aim of this project is to carry out a comprehensive analysis of all the recently completed Supernova Type Ia (SN Ia) surveys and derive robust cosmological constraints from them. The investigators will collaborate with theorists developing state-of-the-art 3-d simulations of SN Ia explosions to test the models against observational data, with the long-term goal of developing theoretically-informed SN distance estimates. In addition, the project will optimize the design of and develop new analysis tools for future large SN surveys, in particular developing methods for SN cosmology analysis without spectroscopic identification or redshifts. All these tools will be part of a flexible, public software package, SNANA, that the proposers have and will continue to develop and share with the community.
Broader impacts of the work include training of undergraduate and graduate students, incorporating supernova data into Google Sky, release of software for use by the community, and development of undergraduate astronomy labs.
This NSF grant period was focussed on two main topics: 1) combine the world's best Type Ia supernova (SNIa) data samples to improve measurements of the dark energy equation of state parameter w, and 2) start a new imaging survey, called the Dark Energy Survey, that includes a high-redshift Supernova program expected to collect a sample that is several times larger than the current world sample, and is of much higher quality. For the first topic, two well known and independent SN-collaborations joined forces: the Sloan Digital Sky Survey-II SN group (SDSS-II) with low redshift SNIa (redshifts z < 0.4), and the Supernova Legacy Survey (SNLS) with SNIa at higher redshifts up to about 1. Since current measurements are limited by systematic uncertainties, our focus has been to improve our understanding of subtle measurement biases.We have significantly improved the accuracy of measuring the SN brightness as described in a recent paper (Betoule et al. A&A 552, 124, 2013). In another paper (Kessler et al., ApJ, 764, 48, 2013) we have evaluated a new kind of uncertainty related to the fact that we don't know exactly how or why the SN brightness varies. In a 3rd paper (in preparation), we use detailed simulations to improve our estimateof key analysis uncertainties, mainly those related to modeling the SNIa brightness as a function of wavelength and time since explosion. For the second topic (DES), we ran a very successful science verification program that ran from Oct 2012 through Feb 2013. During this time we implemented a complicated set of programs that analyze the data immediately in order to discover new supernova and estimate their type (Ia or core collapse) using broadband (griz) photometry. With 62 CCDs per field, four griz filters and up to ten fields exposed in one night, there are up to 2480 analysis jobs needed to process a single night of SN data! To process this enormous amount of data in much less than a day, the analysis jobs are run at "NCSA," the National Center for Supercomputing Applications at the University of Illinois.This immediate analysis is needed in order to schedule and take additional spectroscopic observations while the SNe are still bright. The spectra are used to precisely classify the SN type ... it is only the Type Ia (SNIa) that are used to measure cosmological parameters. We anticipate such a large sample that only a small fraction of the SNe will be spectroscopically classified; the remaining SNe will be classified in the post-survey analysis using only griz photometry.To ensure that the discovery software is working properly, we monitor the detection efficiency of fake supernova that are overlaid on real galaxies in the real images.The official start of the DES recently began onAug 31 2013. A summary of the DES contributions from this NSF grant are 1) provide software framework (SNANA) for simulations and analysis used to design and optimize the DES-SN survey (see Bernstein et al., APJ, 753, 152, 2012), 2) rigorously test the SN-discovery pipeline before sending it to NCSA, and provide a framework to reliably test future updates, 3) generate fake SNe and provide code to overlay these fakes onto real galaxies as part of the SN-data processing and monitoring, 4) install a high-precision GPS antenna & receiver at CTIO to measure precipitable water vapor (PWV) in the atmosphere. The goal of the PWV measurements is to improve the calibration in the z band centered at 9200 A. PWV results are posted each day at www.suominet.ucar.edu/data/pwvGlobalDaily : (goto 2013 and search for CTIO).
