The Impacts of Modeling Choices on the Inference of the Circumgalactic Medium Properties from Sunyaev-Zeldovich Observations

TL;DR

This paper examines how different modeling choices affect the inference of circumgalactic medium properties from Sunyaev-Zeldovich observations, emphasizing the importance of sample selection and fitting models.

Contribution

It systematically investigates the impact of modeling choices on SZ-based CGM property inference using simulations and forecasts for upcoming observations.

Findings

Two-halo term inclusion significantly alters inferred parameters.

Matching by the sample's mass distribution is crucial for accurate modeling.

Number of free parameters affects the robustness of fit results.

Abstract

As the signal-to-noise of Sunyaev-Zeldovich (SZ) cross-correlation measurements of galaxies improves our ability to infer properties about the circumgalactic medium (CGM), we will transition from being limited by statistical uncertainties to systematic uncertainties. Using thermodynamic profiles of the CGM created from the IllustrisTNG (The Next Generation) simulations we investigate the importance of specific choices in modeling the galaxy sample. These choices include different sample selections in the simulation (stellar versus halo mass, color selections) and different fitting models (matching by the shape of the mass distribution, inclusion of a two-halo term). We forward model a mock galaxy sample into projected SZ observable profiles and fit these profiles to a generalized Navarro-Frenk-White profile using forecasted errors of the upcoming Simons Observatory experiment. We test the number of free parameters in the fits and show that this is another modeling choice that yields different results. Finally, we show how different fitting models can reproduce parameters of a fiducial profile, and show that the addition of a two-halo term and matching by the mass distribution of the sample are extremely important modeling choices to consider.