Astronomers Pull a Faint Carbon Monoxide Signal out of Two Public Archives

Three astronomers have pulled a faint carbon monoxide signal out of two public archives, with no instrument built for the job. In a Letter published Nov. 11, 2026, in The Astrophysical Journal Letters, Anirban Roy of New York University and the Flatiron Institute, with Nicholas Battaglia of Cornell University and Anthony R. Pullen of New York University, report emission tied to galaxies at a redshift of about 0.5, found by stacking a Planck map on the positions of galaxies recorded by the eBOSS survey.

The method under test is line intensity mapping. Rather than resolving galaxies one at a time, it adds up the light of a single spectral line from many of them at once, which the authors describe as a way to study how galaxies form and how matter is arranged on the largest scales. New instruments are being designed for that job. The paper asks how far the data already on hand can get, and both the Planck maps and the eBOSS galaxy catalog are public.

Carbon monoxide is the usual stand-in for the cold clouds that stars form out of, because the hydrogen that makes up most of that gas is hard to see directly.

How firm the signal is depends on the analysis. The authors report that reading the stacked emission as the CO(3-2) line depends on the continuum model, the spectral response and the background treatment, and that those choices put the detection confidence somewhere between 2.2 and 3.4 sigma. Sigma is a measure of how far a signal stands above the noise, and at the low end of that range noise alone could account for what they see. That is why the confidence is given as a range rather than a single number.

Roy and colleagues write that stacking is already workable with existing observations from cosmic microwave background experiments and galaxy surveys, and that it complements the purpose-built experiments.