ABSTRACT
We compute the one-point probability distribution function (PDF) of an initially Gaussian dark matter density field using spherical collapse (SC). We compare the results to other forms available in the literature and also compare the PDFs in the Λ-cold dark matter model with an early dark energy (EDE) model. We find that the skewed lognormal distribution provides the best fit to the non-linear PDF from SC for both cosmologies, from a = 0.1 to 1 and for scales characterized by the comoving width of the Gaussian: σG = 0.5, 1, and 2. To elucidate the effect of cosmology, we examine the linear and non-linear growth rates through test cases. For overdensities, when the two models have the same initial density contrast, the differences due to cosmology are amplified in the non-linear regime, whereas, if the two models have the same linear density contrast today, then the differences in cosmology are damped in the non-linear regime. This behaviour is in contrast with voids, where the non-linear growth becomes ‘self-regulatory’ and is less sensitive to cosmology and initial conditions. To compare the PDFs, we examine the difference of the PDFs and evolution of the width of the PDF. The trends with scale and redshift are as expected. A tertiary aim of this paper was to check if the fitting form for the non-linear density–velocity divergence relation, derived for constant equation of state (w) models by Nadkarni-Ghosh holds for the EDE model. We find that it does with an accuracy of 4 per cent, thus increasing its range of validity.
Non-linear density evolution from an improved spherical collapse model
