A Deep ALMA Survey of the Redshift Distribution of Dusty Star-forming Galaxies

TL;DR

This study uses ALMA spectroscopic follow-up to accurately determine the redshift distribution of dusty star-forming galaxies, revealing a steep decline in their abundance at very high redshifts.

Contribution

It provides the highest completeness to date for an unbiased DSFG sample at this flux limit and evaluates the accuracy of photometric redshift methods.

Findings

69% of DSFGs have secure or tentative spectroscopic redshifts.

Most JWST/NIRSpec targeted DSFGs were spectroscopically identified.

Less than 10% of DSFGs with flux >2 mJy are at z>4.

Abstract

We present an Atacama Large Millimeter/submillimeter Array (ALMA) spectroscopic follow-up survey of an 870 $\mu$m-selected sample of dusty star-forming galaxies (DSFGs) in the GOODS-S field. We use these linescans to identify or confirm spectroscopic redshifts (spec-zs) for 20 sources. Including spec-zs from the literature, there are now secure or tentative spec-zs for 52 out of 75 DSFGs (69%). At $f_{870}>2.5$ mJy, the sample is 97% spectroscopically complete, allowing us to model the full DSFG redshift distribution down to nearly the confusion limit for a 15-m telescope at 850 $\mu$m. This is the highest completeness for an unbiased sample at this flux limit to date. We find that nearly all of the DSFGs in our sample that were targeted with JWST/NIRSpec were spectroscopically identified, without much dependence on near-infrared or submillimeter flux or redshift. However, only 29% of our sample have JWST spectroscopic coverage. We use the spec-zs to evaluate various photometric redshift (photo-z) estimates, finding that all methods exhibit an outlier fraction of at least $>20$%. Nearly all of the photo-z methods tend to overshoot the redshifts, leading to overestimates of the number of DSFGs at high redshift ($z>4$). Our results suggest that $\lesssim10$% of $f_{870} \gtrsim 2$ mJy DSFGs lie at $z>4$ and $\lesssim2$% lie at $z>5$, reflecting a steep decline in the abundance of massive dusty galaxies in the first 1.5 Gyr.