Constraining viscous-fluid models in $f(Q)$ gravity using cosmic measurements and large-scale structure data

Abstract: This paper investigates the impact of the effects of bulk viscosity on the accelerating expansion and large-scale structure formation of a universe in which the underlying gravitational interaction is described by $f(Q)$ gravity. Various toy models of the $f(Q)$ gravity theory, including power-law ($f_{1\rm{CDM}}$), exponential ($f_{2\rm{CDM}}$), and logarithmic ($f_{3\rm{CDM}}$) are considered. To test the cosmological viability of these models, we use 57 Hubble parameter data points (OHD), 1048 supernovae distance modulus (SNIa), their combined analysis (OHD+SNIa), 14 growth rate ($f$-data), and 30 redshift-space distortions ($f\sigma_8$) datasets. We compute the best-fit values $\Omega_m, H_0$ and exponents ${n, p, \gamma}$ including the bulk viscosity coefficient $\zeta_0$, through a detailed statistical analysis. Moreover, we study linear cosmological perturbations and compute the density contrast (\delta(z)), growth factor $D(z)$, growth rate $(f(z)$, and redshift-space distortion $f\sigma_8(z)$. Based on the Akaike Information Criterion (AIC) and Bayesian / Schwartz Information Criterion (BIC), a statistical comparison of the $f(Q)$ gravity models with \lcdm is made. From our statistical analysis of cosmic measurements, we found an underestimation of all models on the OHD data; therefore, statistical viability led us to weigh it in favour of SNIa data which resulted in the exponential ($f_{2\rm{CDM}}$) model without bulk viscosity being the most plausible alternative model, while on average the $f_{1\rm{CDM}}$ model performed statistically the weakest. We also found that adding bulk viscosity did not improve the fit of the models to all observational data.