English

Towards optimal and robust $f_{\rm NL}$ constraints with multi-tracer analyses

Cosmology and Nongalactic Astrophysics 2023-09-20 v2

Abstract

We discuss the potential of the multi-tracer technique to improve observational constraints of the local primordial non-Gaussianity (PNG) parameter fNLf_{\rm NL} from the galaxy power spectrum. For two galaxy samples AA and BB, the constraining power is b1BbϕAb1AbϕB\propto |b_1^B b_\phi^A - b_1^Ab_\phi^B|, where b1b_1 and bϕb_\phi are the linear and PNG galaxy bias parameters. We show this allows for significantly improved constraints compared to the traditional expectation b1Ab1B\propto |b_1^A - b_1^B| based on naive universality-like relations where bϕb1b_\phi \propto b_1. Using IllustrisTNG galaxy simulation data, we find that different equal galaxy number splits of the full sample lead to different b1BbϕAb1AbϕB|b_1^B b_\phi^A - b_1^Ab_\phi^B|, and thus have different constraining power. Of all of the strategies explored, splitting by grg-r color is the most promising, more than doubling the significance of detecting fNLbϕ0f_{\rm NL}b_\phi \neq 0. Importantly, since these are constraints on fNLbϕf_{\rm NL}b_\phi and not fNLf_{\rm NL}, they do not require priors on the bϕ(b1)b_\phi(b_1) relation. For direct constraints on fNLf_{\rm NL}, we show that multi-tracer constraints can be significantly more robust than single-tracer to bϕb_\phi misspecifications and uncertainties; this relaxes the precision and accuracy requirements for bϕb_\phi priors. Our results present new opportunities to improve our chances to detect and robustly constrain fNLf_{\rm NL}, and strongly motivate galaxy formation simulation campaigns to calibrate the bϕ(b1)b_\phi(b_1) relation.

Keywords

Cite

@article{arxiv.2302.09066,
  title  = {Towards optimal and robust $f_{\rm NL}$ constraints with multi-tracer analyses},
  author = {Alexandre Barreira and Elisabeth Krause},
  journal= {arXiv preprint arXiv:2302.09066},
  year   = {2023}
}

Comments

9 pages, 4 figures, 1 table. Comments welcomed! v2 matches version published in JCAP