Cosmic CO and [CII] backgrounds and the fueling of star formation over 12 Gyr

TL;DR

This study provides the first empirical detections of the mean cosmic CO and [CII] backgrounds through tomographic intensity mapping, revealing a molecular gas reservoir (Ω_H2) larger than that traced by current galaxy surveys and a global depletion time t_dep ≈ 3(1+z)^{−1} Gyr. By linking CO excitation to a universal, super-linear Kennicutt–Schmidt relation, the work shows star formation is sustained by a large, short-lived molecular supply governed by cosmic inflows and feedback. The CO and [CII] histories, together with line-to-continuum and equivalent width analyses, establish a calibrated, 3D LIM framework with direct empirical line strengths, guiding future instrument design and enabling robust cosmological inferences from line backgrounds across 0<z<4.2. These results anchor LIM as a mature probe of galaxy formation, providing global constraints on gas fueling, cooling, and star-formation efficiency over a 12 Gyr span.

Abstract

Molecular gas, modest in mass yet pivotal within the cosmic inventory, regulates baryon cycling as the immediate fuel for star formation. Across most of cosmic history, its reservoir has remained elusive, with only the tip of the iceberg revealed by luminous carbon monoxide (CO) emitting galaxies. Here we report the first detections of the mean cosmic CO background across its rotational ladder at 7$σ$, together with ionized carbon ([CII]) at 3$σ$, over $0<z<4.2$. This uses tomographic clustering of diffuse broadband intensities with reference galaxies, directly probing aggregate emission in the cosmic web. From CO(1-0) we infer the total molecular gas density, $Ω_{\rm H_2}$, finding it about twice that resolved in galaxy surveys. The global depletion time is $\sim$1 Gyr, shorter than the Hubble time, requiring sustained inflow. CO excitation links to star-formation surface density and, with depletion time, yields a super-linear Kennicutt-Schmidt law that appears universal. Together these results establish a global picture of galaxy growth fueled by a larger, short-lived molecular reservoir. The CO and [CII] detections mark a turning point for line-intensity mapping, replacing forecasts with empirical line strengths and defining sensitivity requirements for upcoming 3D experiments poised to open new windows on galaxy formation and cosmology.

Figures (11)

• Figure 1: Tomographic CIB spectrum and cosmic CO and [CII] histories.a, Evolving CIB spectrum shown as bias-weighted emissivity $\epsilon_{\nu}b$ versus rest-frame frequency, color-coded by redshift. Data points are from 11-band intensity-galaxy cross-correlations in the two-halo clustering regime2025ApJ...992...65C. Vertical bands mark nine CO and the [CII] 158 $\mu$m lines, with widths set by the mean spectral resolution $R\sim3.5$; data points within these bands receive line excess. Colored curves show the best-fit continuum-plus-line model, with unresolved CO and [CII] drawn as tophats at 1% of their frequencies. b, We detect the cosmic CO and [CII] backgrounds by exploiting their characteristic frequency-redshift line-excess patterns in panel a. The comoving luminosity densities, with CO summed over nine lines (black and red curves with shaded intervals), peak broadly near cosmic noon. Dashed lines give posterior medians for individual CO transitions, revealing evolving excitation and the dominance of mid-$J$ transitions. Black and red points are per-redshift refits, supporting our redshift parameterization; the single transparent [CII] point is negative due to noise. [CII] is unconstrained at $z<0.4$ owing to a gap in spectral coverage. For broader context, yellow data points and upper limits (95%) indicate Ly$\alpha$ intensity-mapping constraints2019ApJ...877..150C2018MNRAS.481.1320C, and the green dashed line shows predicted hydrogen 21 cm amplitudes2024ApJ...963...23A. [CII] likely rivals Ly$\alpha$ as the brightest line background, while CO remains much brighter than 21 cm. All error bars and shaded bands indicate 1$\sigma$ (68%) uncertainties.
• Figure 1: Landscape of millimeter line monopole intensities. Same as Fig. \ref{['fig:T_b']}, but shown in specific intensity $I_\nu$ (Jy sr$^{-1}$) rather than brightness temperature. Curves show posterior median CO and [CII] line monopoles, color-coded by redshift. The jointly fitted CIB dust-continuum monopole is shown in black, and the sum with the line contribution matches the CIB-only fit in the companion work2025ApJ...992...65C, based on the same data vector. The thermal Sunyaev–Zeldovich spectral distortion is shown in red, with the dashed segment indicating the decrement2020ApJ...902...56C. All shaded bands correspond to 68% uncertainties. Filled gray regions show atmospheric transmission windows at a dry site2001ITAP...49.1683P. This figure is analogous to Supplementary Fig. \ref{['fig:line_monopole']} but includes only updated empirical constraints from cross-correlation-based tomographic intensity mapping.
• Figure 1: Sampling of CO and [CII] lines over redshift. The matrix shows whether a CO or [CII] line contributes, at a given redshift, to a data point in the bias-weighted CIB emissivity (Fig. \ref{['fig:spectrum_and_lines']}a), measured from tomographic cross-correlations with spectroscopic reference objects2025ApJ...992...65C. Blue-filled and black-hatched cells mark coverage by Planck and Herschel bands, respectively. The number of data points sampling each line is listed at right. Together, these provide sufficient coverage to track CO and [CII] emission, along with the CIB continuum, over 12 Gyr.
• Figure 2: Landscape of millimeter line monopoles in brightness temperature. The mean sky monopoles $T_{\rm b}$ in $\mu$K versus observed frequency for the CO and [CII] backgrounds, derived from our posterior constraints, are shown as curves color-coded by redshift, with shading indicating the 68% uncertainties. The black curve and band show the updated CIB dust-continuum monopole in this work fitted jointly with the lines, with their sum closely matching the previous continuum-only fit2025ApJ...992...65C. The spectral distortion of the thermal Sunyaev-Zeldovich background is shown in red, with the dashed segment indicating the decrement ($y \approx 1.22 \times 10^{-6}$)2020ApJ...902...56C. All component amplitudes are now empirically constrained, with the lines tracing LSS in 3D and the continuum tracing it in 2D, informing both LIM analyses and foreground modeling or mitigation for CMB experiments. Filled gray regions mark atmospheric transmission windows at a relatively dry site2001ITAP...49.1683P, highlighting more accessible low-frequency bands from the ground. This figure establishes key benchmarks for the far-infrared to millimeter line backgrounds, replacing previously divergent model forecasts with direct measurements and defining amplitude baselines and sensitivity targets for future LIM experiments.
• Figure 2: Contribution of CO and [CII] to the CIB monopole. Sky-averaged intensities of the total CIB (dust continuum plus lines), CO (nine transitions combined), and [CII] are shown as functions of observed frequency, complementing the brightness-temperature view in Fig. \ref{['fig:T_b']}. Black lines and shaded bands show our posterior medians with 68% credible intervals. Cyan markers denote total CIB monopoles (1$\sigma$) from FIRAS and Planck2019ApJ...877...40O, which we use as external integral constraints. Earlier claims of exceptionally high [CII] at $z\sim2.6$ (yellow, $2\sigma$)2018MNRAS.478.1911P2019MNRAS.489L..53Y lie closer to the total CIB than to plausible [CII] amplitudes. Our empirical line measurements yield CO and [CII] constraints that broadly agree with two recent models2022AA...667A.156B2022ApJ...929..140Y while disfavouring another CO prediction2024PhRvD.110b3513C. Although [CII] has a higher absolute intensity, CO becomes increasingly important relative to the CIB continuum toward lower frequencies, underscoring its relevance for precision CIB and CMB analyses.