The proposed research will focus on two primary science goals: (1) extracting higher-order statistics (e.g., the bispectrum) from TSZ maps produced using planck data and inferring cosmological…

This $460,835 federal grant was awarded by the National Aeronautics and Space Administration (NASA) to The Trustees of Columbia University in the City of New York, doing business as Columbia University, to conduct research on extracting higher-order statistics from Sunyaev-Zel'dovich (SZ) effect maps and applying novel machine learning methods to analyze SZ data. The research aims to improve constraints on dark energy, neutrino mass, and other cosmological parameters by leveraging SZ data from the Planck satellite and ground-based experiments like the Atacama Cosmology Telescope. The award, which has a period of performance through February 2026, is not associated with a set-aside designation. Columbia University is the prime recipient and will serve as the lead research institution on this NASA-funded astrophysics project.

The proposed research will focus on two primary science goals: (1) extracting higher-order statistics (e.g., the bispectrum) from TSZ maps produced using planck data and inferring cosmological parameters in combination with the TSZ power spectrum, which will allow degeneracies between cosmological and baryonic-physics parameters to be broken, and (2) applying novel machine-learning methods to directly extract the maximum amount of cosmological information in the aforementioned TSZ maps, without relying on any particular summary statistic. These science goals will be further enhanced by analyzing TSZ maps made from a combination of planck data and data from high-resolution, ground based CMB experiments, in particular the atacama cosmology telescope (act). The planck-only TSZ maps are limited to 10-arcminute angular resolution due to the resolution of the mid-frequency planck channels, which are crucial for TSZ extraction by combining planck and act data, the act team has constructed publicly available TSZ maps with roughly 2-arcminute angular resolution, which we will analyze. The outcomes of this research program will be new measurements of TSZ higher-order statistics on an unprecedentedly broad range of angular scales, new constraints on dark energy and the neutrino mass derived from these statistics, and the first application of a novel machine-learning framework to directly extract cosmological information from wide-area TSZ maps. This work is directly related to two key NASA astrophysics science objectives: "how does the universe work?" and "how did we get here?" the proposed research addresses the first question by yielding new constraints on the behavior of dark energy and the properties of the fundamental constituents of the universe, as well as by applying new analysis methods to enable more sensitive searches for new physics in the cosmos. The investigation addresses the second question by mapping the distribution of thermal gas pressure throughout the universe, which is a direct probe of cosmic structure formation and feedback processes associated with galaxy formation on small scales.