Constraining decaying dark matter with BOSS data and the effective field theory of large-scale structures

TL;DR

This paper updates constraints on decaying dark matter models using BOSS-DR12 data analyzed with the EFTofLSS formalism, combining multiple datasets to refine limits on dark matter decay parameters and exploring implications for the $S_8$ tension.

Contribution

It provides the first calculation of mildly non-linear power spectra for decaying dark matter models using EFTofLSS and improves constraints on decay parameters with BOSS and Planck data.

Findings

Constraints on dark matter decay fraction: $f_{dcdm} \,\lesssim 0.022$ (95 ext{\%} C.L.)

Planck data dominates constraints, EFTofBOSS offers marginal improvement

EFTofBOSS data significantly refines constraints in models with WDM decay products

Abstract

We update cosmological constraints on two decaying dark matter models in light of BOSS-DR12 data analyzed under the Effective Field Theory of Large-Scale Structures (EFTofLSS) formalism, together with Planck, Pantheon and other BOSS measurements of the baryonic acoustic oscillation (BAO). In the first model, a fraction $f_{\rm dcdm}$ of cold dark matter (CDM) decays into dark radiation (DR) with a lifetime $\tau$. In the second model (recently suggested as a potential resolution to the $S_8$ tension), all the CDM decays with a lifetime $\tau$ into DR and a massive warm dark matter (WDM) particle, with a fraction $\varepsilon$ of the CDM rest mass energy transferred to the DR. Using numerical codes from the recent literature, we perform the first calculation of the mildly non-linear (matter and galaxy) power spectra with the EFTofLSS for these two models. In the case of DR products, we obtain the constraints $f_{\rm dcdm}\lesssim0.022$ (95\% C.L.) for lifetimes shorter than the age of the universe, and $\tau/f_{\rm dcdm} \gtrsim 250$ Gyr in the long-lived regime assuming $f_{\rm dcdm}\to1$. We show that Planck data contributes the most to these constraints, with EFTofBOSS providing a marginal improvement over conventional BAO and redshift space distortions ($f\sigma_8$) data. In the case of DR and WDM decay products, we find that EFTofBOSS data significantly improves the constraints at 68\% C.L. on the CDM lifetime with a $S_8$ prior from KiDS-1000. We show that, in order to fit EFTofBOSS data while lowering $S_8$ to match KiDS-1000, the best-fit model has a longer lifetime $\tau = 120$ Gyr, with a larger kick velocity $v_{\rm kick}/c \simeq \varepsilon \simeq 1.2\%$, than that without EFTofBOSS ($\tau = 43$ Gyr, $\varepsilon =0.6\%$). We anticipate that future surveys will provide exquisite constraints on such models.