The scale and redshift variation of density and velocity distributions in dark matter flow and two-thirds law for pairwise velocity 后印本

摘要: A halo-based non-projection approach is proposed to study the scale and redshift dependence of density and velocity distributions (PDF) in dark matter flow. All particles are divided into halo and out-of-halo particles such that PDF can be studied separately. Without projecting particle fields onto grid, scale dependence is analyzed by counting all pairs on different scales $r$. Redshift dependence is studied via generalized kurtosis. From this analysis, we can demonstrate: i) Delaunay tessellation can be used to reconstruct density field. Density correlations/spectrum are obtained, modeled and compared with theory; ii) $m$th moment of pairwise velocity can be analytically modelled. On small scale, even order moments can be modelled by a two-thirds law $\langle(\Delta u_L)^{2n}\rangle\propto{(-\epsilon_ur)}^{2/3}$, while odd order moments $\langle(\Delta u_L)^{2n+1}\rangle=(2n+1)\langle(\Delta u_L)^{2n}\rangle\langle\Delta u_L\rangle\propto{r}$ and satisfy a generalized stable clustering hypothesis (GSCH); iii) Scale dependence is studied for longitudinal velocity $u_L$ or $u_L^{'}$, pairwise velocity (velocity difference) $\Delta u_L$=$u_L^{'}$-$u_L$ and velocity sum $\Sigma u_L$=$u^{'}_L$+$u_L$. Fully developed velocity fields are never Gaussian on any scale; iv) On small scale, both $u_L$ and $\Sigma u_L$ can be modelled by a $X$ distribution to maximize system entropy. Distributions of $\Delta u_L$ is different with its moments analytically derived; v) On large scale, both $\Delta u_L$ and $\Sigma u_L$ can be modelled by a logistic function; vi) Redshift evolution of velocity distributions follows prediction of $X$ distribution with a decreasing shape parameter $\alpha(z)$ to continuously maximize system entropy.