CONCERTO: High-fidelity simulation of millimeter line emissions of galaxies and [CII] intensity mapping

Abstract

The intensity mapping of the [CII] 158-µm line redshifted to the submillimeter window is a promising probe of the z>4 star formation and its spatial distribution into large-scale structures. To prepare the first-generation experiments (e.g., CONCERTO), we need realistic simulations of the submillimeter extragalactic sky in spectroscopy. We present a new version of the simulated infrared dusty extragalactic sky (SIDES) model including the main submillimeter lines around 1 mm (CO, [CII], [CI]). This approach successfully reproduces the observed line luminosity functions. We then use our simulation to generate CONCERTO-like cubes (125-305 GHz) and forecast the power spectra of the fluctuations caused by the various astrophysical components at those frequencies. Depending on our assumptions on the relation between the star formation rate and [CII] luminosity, and the star formation history, our predictions of the z∼6 [CII] power spectrum vary by two orders of magnitude. This highlights how uncertain the predictions are and how important future measurements will be to improve our understanding of this early epoch. SIDES can reproduce the CO shot noise recently measured at ∼100 GHz by the millimeter-wavelength intensity mapping experiment (mmIME). Finally, we compare the contribution of the different astrophysical components at various redshifts to the power spectra. The continuum is by far the brightest, by a factor of three to 100, depending on the frequency. At 300 GHz, the CO foreground power spectrum is higher than the [CII] one for our base scenario. At lower frequencies, the contrast between [CII] and extragalactic foregrounds is even worse. Masking the known galaxies from deep surveys should allow us to reduce the foregrounds to 20 % of the [CII] power spectrum up to z∼6.5. However, this masking method will not be sufficient at higher redshifts. The code and the products of our simulation are released publicly, and can be used for both intensity mapping experiments and submillimeter continuum and line surveys.

A NOEMA Molecular Line Scan of the Hubble Deep Field North: Improved Constraints on the CO Luminosity Functions and Cosmic Density of Molecular Gas

“…Given that there are no CO lines with spectroscopic confirmation entering this bin, it is also more sensitive to the assumed priors, and only upper limits can be derived if one limits to the top sources (see Appendix A for more details). We do find that the bright end of the CO(4-3) LF is consistent with the predictions from both SIDES and SPRITZ, while the models predict a lower luminosity density than is observed toward the fainter end (Bethermin et al 2022;Bisigello et al 2022). While it appears that there is a tantalizing break in the CO(4-3) LF, we find that it is not statistically significant (see Section 3.6), and the differences could again be caused by cosmic variance.…”

“…We compare the observed LFs to the recent theoretical predictions from the SIDES (Bethermin et al 2022) and SPRITZ (Bisigello et al 2022) simulations. In brief, both simulations use empirical prescriptions for the IR luminosity and CO-to-IR scaling relations.…”

A NOEMA molecular line scan of the Hubble Deep Field North: Improved constraints on the CO luminosity functions and cosmic density of molecular gas

“…Given there are no CO lines with spectroscopic confirmation entering this bin, it is also more sensitive to the assumed priors, and only upper limits can be derived if one limits to the top-sources (see Appendix A for more details). We do find that the bright end of the CO(4-3) LF is consistent with the predictions from both SIDES and S , while the models predict a lower luminosity density than is observed towards the fainter end (Béthermin et al 2022;Bisigello et al 2022). While it appears there is a tantalising break in the CO(4-3) LF we find it is not statistically significant (see § 3.6) and the differences could again be caused by cosmic variance.…”

“…We compare the observed LFs to the recent theoretical predictions from the SIDES (Béthermin et al 2022) and S (Bisigello et al 2022) simulations. In brief, both simulations use empirical prescriptions for the IR luminosity and CO-to-IR scaling relations.…”

“…This suggest that the variations seen between the fields can be attributed to cosmic variance. Interestingly both SIDES and S seem to predict a lower luminosity density for CO(2-1) than is observed in both fields, making it less clear whether this is due to cosmic variance (Béthermin et al 2022) or a missing ingredient in the models. Taken together, the combined measurements from the HDF-N and HUDF are (still) reasonably well described by a single Schechter function (Fig.…”