The Tianlai experiment is dedicated to the observation of large scale structures (LSS) by the 21 cm intensity mapping technique. In this paper we make forecasts on its capability at observing or constraining the dark energy parameters and the primordial non-Gaussianity. From the LSS data one can use the baryon acoustic oscillation (BAO) and the growth rate derived from the redshift space distortion (RSD) to measure the dark energy density and equation of state. The primordial non-Gaussianity can be constrained either by looking for scale-dependent bias in the power spectrum, or by using the bispectrum. Here we consider three cases: the Tianlai cylinder array pathfinder which is currently being built, an upgrade of the pathfinder array with more receiver units, and the full-scale Tianlai cylinder array. Using the full-scale Tianlai experiment, we expect σ w 0 ∼ 0.082 and σ wa ∼ 0.21 from the BAO and RSD measurements, σ local f NL ∼ 14 from the power spectrum measurements with scale-dependent bias, and σ local f NL ∼ 22 and σ equil f NL ∼ 157 from the bispectrum measurements.Subject headings: cosmological parameters -large-scale structure of universe

As an alternative way to describe the cosmological velocity field, we discuss the evolution of rotational invariants constructed from the velocity gradient tensor. Compared with the traditional divergence-vorticity decomposition, these invariants, defined as coefficients of the characteristic equation of the velocity gradient tensor, enable a complete classification of all possible flow patterns in the dark-matter comoving frame, including both potential and vortical flows. We show that this tool, first introduced in turbulence two decades ago, is very useful for understanding the evolution of the cosmic web structure, and in classifying its morphology. Before shell crossing, different categories of potential flow are highly associated with the cosmic web structure because of the coherent evolution of density and velocity. This correspondence is even preserved at some level when vorticity is generated after shell crossing. The evolution from the potential to vortical flow can be traced continuously by these invariants. With the help of this tool, we show that the vorticity is generated in a particular way that is highly correlated with the large-scale structure. This includes a distinct spatial distribution and different types of alignment between the cosmic web and vorticity direction for various vortical flows. Incorporating shell crossing into closed dynamical systems is highly non-trivial, but we propose a possible statistical explanation for some of the phenomena relating to the internal structure of the three-dimensional invariant space.

In this paper, we report a significant recovery of the linear baryonic acoustic oscillation (BAO) signature by applying the isobaric reconstruction algorithm to the non-linear matter density field. Assuming only the longitudinal component of the displacement being cosmologically relevant, this algorithm iteratively solves the coordinate transform between the Lagrangian and Eulerian frames without requiring any specific knowledge of the dynamics. For dark matter field, it produces the non-linear displacement potential with very high fidelity. The reconstruction error at the pixel level is within a few percent, and is caused only by the emergence of the transverse component after the shell-crossing. As it circumvents the strongest non-linearity of the density evolution, the reconstructed field is welldescribed by linear theory and immune from the bulk-flow smearing of the BAO signature. Therefore this algorithm could significantly improve the measurement accuracy of the sound horizon scale s. For a perfect large-scale structure survey at redshift zero without Poisson or instrumental noise, the fractional error ∆s/s is reduced by a factor of ∼ 2.7, very close to the ideal limit with linear power spectrum and Gaussian covariance matrix.

Cosmological large scale structure N -body simulations are computation-light, memory-heavy problems in supercomputing. The considerable amount of memory is usually dominated by an inefficient way of storing more than sufficient phase space information of particles. We present a new parallel, information-optimized, particle-mesh-based N -body code CUBE, in which information-efficiency and memory-efficiency are increased by nearly an order of magnitude. This is accomplished by storing particle's relative phase space coordinates instead of global values, and in the format of fixed point as light as 1 byte. The remaining information is given by complementary density and velocity fields (negligible in memory space) and proper ordering of particles (no extra memory). Our numerical experiments show that this information-optimized N -body algorithm provides accurate results within the error of the particle-mesh algorithm. This significant lowering of the memory-to-computation ratio breaks the bottleneck of scaling up and speeding up large cosmological N -body simulations on multi-core and heterogeneous computing systems.