• Corpus ID: 249017919

Unifying Sunyaev-Zel'dovich and X-ray predictions from clusters to galaxy groups: the impact of X-ray mass estimates on the $Y-M$ scaling relation

@inproceedings{Pop2022UnifyingSA,
  title={Unifying Sunyaev-Zel'dovich and X-ray predictions from clusters to galaxy groups: the impact of X-ray mass estimates on the \$Y-M\$ scaling relation},
  author={Ana-Roxana Pop and Lars Hernquist and Daisuke Nagai and Rahul Kannan and Rainer Weinberger and Volker Springel and Mark Vogelsberger and Dylan Nelson and R{\"u}diger Pakmor and Paul Torrey},
  year={2022}
}

• Ana-Roxana Pop, L. Hernquist, +7 authors P. Torrey
• Published 23 May 2022
• Physics

One of the main limitations in precision cluster cosmology arises from systematic errors and uncertainties in estimating cluster masses. Using the Mock-X pipeline, we produce synthetic X-ray images and derive cluster and galaxy group X-ray properties for a sample of over 30,000 simulated galaxy groups and clusters with 𝑀 500crit between 10 12 and 2 × 10 15 M (cid:12) in IllustrisTNG. We explore the similarities and differences between IllustrisTNG predictions of the Sunyaev-Zel’dovich and X-ray scaling relations with mass. We find a median hydrostatic mass bias 𝑏 = 0 . 125 ± 0 . 003 for 𝑀 500crit > 10 13 M (cid:12) . The bias increases to 𝑏 = 0 . 17 ± 0 . 004 when masses are derived from synthetic X-ray observations. We model how different underlying assumptions about the dependence of 𝑌 X on halo mass can generate biases in the observed 𝑌 SZ – 𝑀 𝑌 X scaling relation. In particular, the simplifying assumption that 𝑌 X – 𝑀 tot is self-similar at all mass scales largely hides the break in 𝑌 SZ – 𝑀 tot and overestimates 𝑌 SZ at galaxy and groups scales. We show that calibrating the 𝑌 X –mass proxy using a new model for a smoothly broken power law reproduces the true underlying 𝑌 SZ – 𝑀 tot scaling relation with high accuracy. Moreover, 𝑀 𝑌 X estimates calibrated with this method lead to 𝑌 SZ – 𝑀 𝑌 X predictions that are not biased by the presence of lower mass clusters or galaxy groups in the sample. Finally, we show that our smoothly broken power law model provides a robust way to derive the 𝑌 X –mass proxy, significantly reducing the level of mass bias for clusters, groups, and galaxies.