The purpose of large scale structure (LSS) surveys, both space based such as euclid and wfirst, and ground based, such as desi and lsst, is to elucidate the questions of origins of structure in our…

The National Aeronautics and Space Administration (NASA) awarded a $531,823.42 grant to the University of California, Berkeley (UC Berkeley) to develop new theoretical and numerical modeling methods for extracting cosmological information from large-scale structure surveys, such as Euclid and WFIRST. The grant does not have a set-aside designation. The goal is to create improved techniques for analyzing data from space-based and ground-based surveys to better understand the origins of structure in the universe. The methods will be tested using realistic simulations of dark matter and galaxy populations before being applied to real observational data. The results may significantly benefit upcoming large-scale structure missions. As a public research university, UC Berkeley conducts a wide range of federally funded research across science, engineering, and other disciplines. The State of California Controllers Office serves as the parent organization administering the grant funding to UC Berkeley.

The purpose of large scale structure (LSS) surveys, both space based such as euclid and wfirst, and ground based, such as desi and lsst, is to elucidate the questions of origins of structure in our universe and its contents. The primary goal of this proposal is to develop a new theoretical and numerical modeling of how to extract the cosmological information from the large scale structure observations by these surveys. We propose to do this in the context of a recently developed framework which holds the promise of extracting all the cosmological information optimally. The method starts with an exact bayesian probabilistic model for the cosmological parameters given the data, which is then solved through a process that combines elements of numerical optimization in high dimensions and analytic marginalization to find the best solution for cosmological parameters and their covariance matrix. We propose to apply this method both to weak lensing and galaxy clustering observations, as well as to datasets that combine both. We will investigate cosmological parameters and astrophysical nuisance parameters that are to be marginalized over. The methods will be developed using realistic simulations of both dark matter and of galaxies populated in the dark matter simulations, before being applied to real data. Results of this study will provide new methods that may dramatically improve upon the existing methods and may be of direct and significant benefit to space and ground based LSS missions.