Galaxy surveys are crucial for studying large-scale structure (LSS) and cosmology, yet they face limitations—imaging surveys provide extensive sky coverage but suffer from photo-$z$ uncertainties, while spectroscopic surveys yield precise redshifts but are sample limited. To take advantage of both photo-$z$ and spec-$z$ data while eliminating photo-$z$ errors, we propose a deep learning framework based on a dual UNet architecture that integrates these two datasets at the field level to reconstruct the three-dimensional photo-$z$ density field. We train the network on mock samples representative of stage-IV spectroscopic surveys, utilizing CosmicGrowth simulations with a $z=0.59$ snapshot containing ${2048}^3$ particles in a $(1200h^{-1}\mathrm{Mpc}{)}^3$ volume. Several metrics, including correlation coefficient, mean absolute error, mean squared error, peak SNR, and structural similarity index measure, validate the model’s accuracy. Moreover, the reconstructed power spectrum closely matches the ground truth at small scales ($k\gtrsim 0.06h/\mathrm{Mpc}$) within the $1\sigma$ confidence level, while the UNet model significantly improves the estimation of photo-$z$ power spectrum multipoles. This study demonstrates the potential of deep learning to enhance LSS reconstruction by using both spectroscopic and photometric data.

Close

PrevNext

View figure in article

• Large scale structure of the Universe