A research team led by Prof. Xiaohu Yang proposes the ELUCID project: Using the PM method can effectively enhance the accuracy of reconstruction numerical simulation. To carry out the reconstructed numerical simulation (ELUCID) with galaxies in the local universe can make best use of the observational data to study the non-linear evolution theory of structure formation. By comparing the reconstructed simulation and original simulation, they found that this method could reproduce amplitude and phase of perturbation in the initial density field exactly. In this approach, when adopting a grid cell size of 0.75 Mpc/h and 40 time steps PM iteration, Hamilton Markov Chain Monte Carlo (HMC) algorithm can recover more than half of the phase information down to a scale k ~ 0.85 h/Mpc at high z and to k ~ 3.4 h/Mpc at z=0. These results attain an extremely high accuracy, which far beyond all accuracies achieved by other reconstruction models. At the same time, the team also verified that by using a higher spatial resolution could further improve the accuracy of reconstruction model, providing a standard line for the limitation of accuracy can be reached by applying this method to observations. On the other hand, via comparison of related power spectra、mass function of dark matter halos and so on, they found that this method brought the smallest systematic errors.
Figure: ELUCID numerical simulation test. Left: distribution of particles in L100 numerical simulation; right: distribution of particles predicted by the corresponding reconstructed numerical simulation.