[CII] luminosity models and large-scale image cubes based on COSMOS 2020 and ALPINE-ALMA [CII] data back to the epoch of reionisation

We have implemented a novel method to create simulated [CII] emission line intensity mapping (LIM) data cubes using COSMOS 2020 galaxy catalogue data. It allows us to provide solid lower limits for previous simulation-based model predictions and the expected signal strength of upcoming surveys. We applied [CII]158$\mu$m luminosity models to COSMOS 2020 to create LIM cubes covering a 1.2$\times$1.2 deg$^2$ sky area. These models were derived using galaxy bulk property data from the ALPINE-ALMA survey over the redshift range $4.4<z<5.9$, while additional models were taken from the literature. The cubes cover $3.42<z<3.87$, $4.14<z<4.76$, $5.34<z<6.31$, and $6.75<z<8.27$, matched to planned observations from the EoR-Spec module of the Prime-Cam instrument in the Fred Young Submillimeter Telescope (FYST). We also created predictions including additional galaxies below current detection limits by `extrapolating' from the faint end of the COSMOS 2020 luminosity function, comparing these to predictions from the literature. In addition, we computed signal-to-noise (S/N) ratios for the power spectra, using parameters from the planned FYST survey with predicted instrumental noise levels. We find lower limits for the expected power spectrum using the likely incomplete empirical data: when normalised by $2\pi^2$, the amplitudes at $k$ = 1 Mpc$^{-1}$ are 3.06$\times$10$^7$, 1.43$\times$10$^7$, 9.80$\times$10$^5$, 2.77$\times$10$^5$ (Jy sr$^{-1}$)$^2$ for the aforementioned redshift ranges. For the extrapolated sample, the power spectra are consistent with prior predictions, indicating that extrapolation is a viable method for creating mock LIM cubes. In this case, we expect a result of S/N>1 when using FYST parameters. However, our high-redshift results remain inconclusive because of the poor completeness of COSMOS 2020 at $z$>6.3. These predictions will be improved on the basis of future JWST data.