JWST COSMOS-3D: Spectroscopic Census and Luminosity Function of [O III] Emitters at 6.75<z<9.05 in COSMOS

TL;DR

This paper presents the largest spectroscopic census to date of [O III]+Hβ emitters at $6.75<z<9.05$ in COSMOS from the COSMOS-3D program, delivering 237 emitters and a well-characterized end-to-end completeness. Using two redshift bins, the authors derive the [O III] luminosity function with improved constraints on the knee and faint-end slope, finding $\log_{10}L^{*}_{[OIII]} \approx 42.96$–$43.10$ and $\alpha \approx -1.89$ to $-1.90$, while showing a rapid decline in luminosity density from $z\sim7$ to $z\sim8$ with $\rho_{[OIII]}(z\sim7) \approx 1.28\times10^{39}$ and $\rho_{[OIII]}(z\sim8) \approx 0.61\times10^{39}$ erg s$^{-1}$ cMpc$^{-3}$. The study compares the observed LF and EW distribution to several theoretical models, finding general agreement with JAGUAR, FLARES, and SC SAM but tension with THESAN and SPHINX for the LF, while THESAN/SPHINX better reproduce the high-EW tail. A robust cosmic variance analysis yields $\sim$15% variance at $z\sim7$–$8$, underscoring the importance of wide-area JWST surveys to map large-scale structure during reionization. These results demonstrate the essential role of COSMOS-3D in constraining early galaxy evolution and provide a framework for future wide-field spectroscopic studies with JWST.

Abstract

We present a spectroscopically-selected [OIII]+Hb emitters catalogue at 6.75<z<9.05 and the resulting [OIII] 5008 ÅLuminosity Function (LF) in the COSMOS field. We leverage the 0.3 deg$^{2}$ covered to date by COSMOS-3D using NIRCam/WFSS F444W (90% of the survey) to perform the largest spectroscopic search for [OIII] emitters at 6.75<z<9.05. We present our catalogue of 237 [OIII] emitters and their associated completeness function. The inferred constraints on the [OIII] LF enable us to characterise the knee of the [OIII] LF, resulting in improved [OIII] LF constraints at z~7,8. Notably, we find evidence for an accelerated decline of the [OIII] luminosity density between z~7 and z~8, which could be expected if the metallicity of [OIII] emitters, as well as the cosmic star-formation rate density, is declining at these redshifts. We find that theoretical models that reproduce the z~7,8 [OIII] LF do not reproduce well the [OIII] equivalent width distribution, pointing to potential challenges in the modelling of[OIII] and other nebular lines in the early Universe. Finally, we provide the first constraints on the cosmic variance of [OIII] emitters, estimating at 15% the relative uncertainty for the z~7,8 [OIII] LF in the 0.3 deg$^2$ field. This estimate is in good agreement with that inferred from clustering, and shows that the [OIII] LF derived from smaller extragalactic legacy fields is strongly affected by cosmic variance. Our results highlight the fundamental role that wide-area JWST slitless surveys play to map the galaxy large-scale structure down into the reionisation era, serving as a springboard for a variety of science cases.

Figures (12)

• Figure 1: 2D SNR spectra of all $237$ individual [O iii] [O iii] emitters reported in this work. The emitters are ordered by redshift (only a subsample of redshifts are shown on the y-axis for readability).
• Figure 2: Completeness functions for the [O iii] [O iii]+Hβ H$\beta$ emitter search as a function of the measured [O iii] [O iii] 5008 $\AA$ SNR. The black square datapoints and red curve denote the effective completeness including the initial Gaussian-matched filtering (GM) and the subsequent visual inspection (VI). The best-fit completeness function of each separate step is shown with yellow and blue lines, as well as the binned data values for the visual inspection (gray dots). Note that the visual inspection is specific to the quality threshold chosen (here $q\geq1.5$).
• Figure 3: Top Left: Redshift distribution of the [O iii] [O iii] emitter sample presented in this work. We also show the coverage of the [O iii] [O iii] line in the F444W filter and the variation of the effective volume, taking into account the specific mosaic pattern of COSMOS-3D and the sensitivity of the observations (see \ref{['app:rms']}).Top Right: Sky position of the distribution of the [O iii] [O iii] emitter sample presented in this work. The sources are colour-coded as a function of redshift, and the size of the dot represent the observed [O iii] [O iii] flux. The contour of the COSMOS-3D field is outlined in grey, with the remaining observations in the North-East corner in orange. Bottom: 3D visualisation of the [O iii] [O iii] emitter sample. The size and colour scaling are the same as in the top right plot.
• Figure 4: JWST [O iii] [O iii] Luminosity Functions at $5\lesssim z\lesssim 9$. The COSMOS-3D constraints determined in this work are shown with dark red and orange datapoints (empty markers indicate a completeness $<25\%$). We show the $z=5.33-6.96$ [O iii] [O iii] LF from EIGER Matthee2023_EIGER in yellow. The datapoints at $z=7.1,7.9$ from FRESCO Meyer2024 in gray squares and circles (with corrected errors including $45\%$ cosmic variance), and the NIRCam medium band excess-selected $z\sim 7$ [O iii] [O iii] LF from Wold2025 is shown in gray triangles. The best-fit relations at $z=7.1$ and $z=7.9$ are shown in orange and red lines (and envelopes denoting the $16-84$th percentiles) which are fit to all the NIRCam WFSS observations.
• Figure 5: Comparison between the spectroscopically-confirmed [O iii] [O iii] luminosity function (black and gray datapoints) and theoretical predictions at $z\simeq 7,8$. Empty datapoints indicate a low completeness ($<25\%$). The errors from Meyer2024 have been corrected for cosmic variance. The COSMOS-3D constraints are in significant tension with the THESAN and SPHIX predictions but in good agreement with JAGUAR, FLARES and SC SAM (see further text). Our new observations constrain the rapid decline beyond the knee of the LF ($L_{[O~{\sc iii}]}>10^{43}\rm{erg\ s}^{-1}$), in good agreements with models covering this regime.