COSMOS Spectroscopic Redshift Compilation (First Data Release): 488k Redshifts Encompassing Two Decades of Spectroscopy

TL;DR

The COSMOS Spec-$z$ Compilation DR1 unifies nearly two decades of spectroscopic redshifts across $138$ programs over a $10$ deg$^2$ field, enabling robust calibration and galaxy-population studies up to $z \sim 8$. It details a uniform data-processing pipeline that harmonizes quality flags, corrects astrometry to COSMOS2020, and flags duplicates, while re-deriving physical properties with redshifts fixed to spectroscopic values using both CIGALE and LePHARE. The work analyzes completeness, stellar-mass distributions, and the demographics of quiescent, star-forming, and bursty galaxies, and demonstrates practical uses in photo-$z$ validation and SOM-based gap analysis of spectroscopic coverage. As a living resource, it sets the stage for future data releases that will incorporate JWST, Euclid, Roman, and ground-based surveys, expanding coverage to fainter populations and enabling detailed environmental studies of large-scale structure.

Abstract

We present the COSMOS Spectroscopic Redshift Compilation encompassing ~ 20 years of spectroscopic redshifts within a 10 deg$^2$ area centered on the 2 deg$^2$ COSMOS legacy field. This compilation contains 487,666 redshifts of 266,284 unique objects from 138 individual observing programs up to $z \sim 8$ with median stellar mass $\sim 10^{8.4}$ to $10^{10}$ M$_\odot$ (redshift dependent). Rest-frame $NUVrJ$ colors and SFR -- stellar mass correlations show the compilation primarily contains low- to intermediate-mass star-forming and massive, quiescent galaxies at $z < 1.25$ and mostly low-mass bursty star-forming galaxies at $z > 2$. Sources in the compilation cover a diverse range of environments, including protoclusters such as ``Hyperion''. The full compilation is 50\% spectroscopically complete by $i \sim 23.4$ and $K_s \sim 21.6$ mag; however, this is redshift dependent. Spatially, the compilation is $>50$\% ($>30$\%) complete within the central (outer) region limited to $i < 24$ mag and $K_s < 22.5$ mag, separately. We demonstrate how the compilation can be used to validate photometric redshifts and investigate calibration metrics. By training self-organizing maps on COSMOS2020/Classic and projecting the compilation onto it, we find key galaxy subpopulations that currently lack spectroscopic coverage including $z < 1$ intermediate-mass quiescent galaxies and low-/intermediate-mass bursty star-forming galaxies, $z \sim 2$ massive quiescent galaxies, and $z > 3$ massive star-forming galaxies. This highlights how combining self-organizing maps with our compilation can provide guidance for future spectroscopic observations to get a complete spectroscopic view of galaxy populations. Lastly, the compilation will undergo periodic data releases that incorporate new spectroscopic redshift measurements, providing a lasting legacy resource for the community.

Figures (15)

• Figure 1: Offsets in the astrometry of all sources in the compilation after astrometric corrections were applied using COSMOS2020 Classic as the reference catalog. The dashed, dash-dotted, and dotted lines are the $1\sigma$, $2\sigma$, and $3\sigma$ confidence levels. The median offset is shown as the blue square and is $\sim69.6$ mas with a median absolute deviation of $\sim 35$ mas and standard deviation of $105$ mas.
• Figure 2: Top Panel shows the published spectroscopic redshift distribution of all sources within the final compilation (grey) along with those that have high-quality spec-$z$ ($Q_f$$= 3 - 4$) and broad-line features ($Q_f$$= 13 - 14$). The range of HSC $i$ magnitudes and spec-$z$ for each individual program is shown in the lower panel with each symbol corresponding to the median $i$ magnitude and redshift for each program. Error bars represent the 16$^\textrm{th}$ and 84$^\textrm{th}$ percentile ranges. Programs with no COSMOS2020 Classic matches are placed at 31 mag with a slight perturbation for visualization purposes. In total, 266,284 sources comprise the total compilation with 487,666 redshift measurements from $138$ programs.
• Figure 3: The spectroscopic completeness in terms of HSC $i$ (blue) and UltraVISTA $K_s$ (red) magnitudes with the full population and high $Q_f$ subset shown as solid circles and empty squares, respectively. We find 50% spectroscopic completeness for the full (high $Q_f$) sample at $i \sim 23.4$ ($22.8$) mag and $K_s \sim 21.6$ ($\sim 20.7$) mag. Note this does not include any redshift dependency.
• Figure 4: The spectroscopic completeness (in percentage) based on HSC $i$ (top panel) and UltraVISTA $K_s$ (bottom panel) magnitudes factoring in redshift dependency. The left panels are for the full compilation and the right panels are for the high $Q_f$ subset. The 50% completeness limit is roughly constant at $i \sim 23 - 23.5$ mag but is variable in the high $Q_f$ subset shifting to fainter magnitudes from $z \sim 0$ to $z \sim 2$. The 50% completeness limit in $K_s$ is variable in both the full compilation and high $Q_f$ subset. Note that $>100\%$ is expected and signifies disagreements between COSMOS2020 photometric redshifts and the spectroscopic redshifts in our compilation. Overall, this highlights the need to understand the completeness not just of the whole sample, but also at various redshift and magnitude slices.
• Figure 5: The spatial spectroscopic completeness of the compilation limited to $i < 24$ mag (top) and $K_s < 22.5$ mag (bottom) with the full (high $Q_f$) samples shown on the left (right) panels. Each numerical value corresponds to the spectroscopic completeness of that specific 0.01 deg$^2$ region. The central region has the highest spectroscopic completeness due to the many follow-up programs focused on the CANDELS HST/ACS (light red) and WFC3 (red) region (e.g., 3D-HST -- Brammer2012; MOSDEF -- Kriek2015; HETDEX -- Mentuch2023). The outer regions also have $\gtrsim 30$% completeness even for the high $Q_f$ subset reaching the edges of the HST/ACS F814W imaging (green). Future JWST spectroscopic follow-up within COSMOS-Web (orange) and PRIMER (pink) and other COSMOS-based programs including PASSAGE (PI: Matthew Malkan) and COSMOS-3D (PI: Koki Kakiichi) will both increase the spatial spectroscopic completeness as well as extend it to fainter magnitudes.