Halo nonlinear reconstruction

We apply the nonlinear reconstruction method to simulated halo fields. For halo number density $2.77\times 10^{-2}$ $(h^{-1} {\rm Mpc})^{-3}$ at $z=0$, corresponding to the SDSS main sample density, we find the scale where the noise saturates the linear signal is improved to $k\gtrsim0.36\ h {\rm Mpc}^{-1}$, a factor of $2.29$ improvement in scale, or $12$ in number of linear modes. The improvement is less for higher redshift or lower halo density. We expect this to substantially improve the BAO accuracy of dense, low redshift surveys, including the SDSS main sample, 6dFGS and 21cm intensity mapping initiatives.