CosmoGridV1: a simulated

We present CosmoGridV1: a large set of lightcone simulations for map-level cosmological inference with probes scale structure. It is designed parameter measurement based on Stage-III photometric surveys non-Gaussian statistics and machine learning. CosmoGridV1 spans the $w$CDM model by varying $\Omega_m$, $\sigma_8$, $w_0$, $H_0$, $n_s$, $\Omega_b$, assumes three degenerate neutrinos $\sum m_\nu$ = 0.06 eV. This space covered 2500 grid points Sobol sequence. At each point, we run 7 PkdGrav3 store 69 particle maps at nside=2048 up to $z$=3.5, as well halo catalog snapshots. The fiducial cosmology has 200 independent simulations, along their stencil derivatives. An important part benchmark 28 which include larger boxes, higher counts, redshift resolution shells. They allow testing if new types analyses are sensitive choices made in CosmoGridV1. add baryon feedback effects map level, using shell-based correction model. shells used create weak gravitational lensing, intrinsic alignment, galaxy clustering, UFalcon code. main raw count that can be full-sky given $n(z)$. also release projected forecast, previously KiDS-1000 deep learning constraints data available www.cosmogrid.ai.