Wide and deep near-UV (360nm) galaxy counts and the extragalactic background light with the Large Binocular Camera

Deep multicolour surveys are the main tool to explore the formation and evolution of the faint galaxies which are beyond the spectroscopic limit with the present technology. The photometric properties of these faint galaxies are usually compared with current renditions of semianalytical models to provide constraints on the fundamental physical processes involved in galaxy formation and evolution, namely the mass assembly and the star formation. Galaxy counts over large sky areas in the near-UV band are important because they are difficult to obtain given the low efficiency of near-UV instrumentation, even at 8m class telescopes. A large instrumental field of view helps in minimizing the biases due to the cosmic variance. We have obtained deep images in the 360nm U band provided by the blue channel of the Large Binocular Camera at the prime focus of the Large Binocular Telescope. We have derived over an area of ~0.4 sq. deg. the galaxy number counts down to U=27 in the Vega system (corresponding to U=27.86 in the AB system) at a completeness level of 30% reaching the faintest current limit for this wavelength and sky area. The shape of the galaxy counts in the U band can be described by a double power-law, the bright side being consistent with the shape of shallower surveys of comparable or greater areas. The slope bends over significantly at U>23.5 ensuring the convergence of the contribution by star forming galaxies to the EBL in the near-UV band to a value which is more than 70% of the most recent upper limits derived for this band. We have jointly compared our near-UV and K band counts collected from the literature with few selected hierarchical CDM models emphasizing critical issues in the physical description of the galaxy formation and evolution.

💡 Research Summary

The paper presents a deep, wide‑field survey of galaxies in the near‑ultraviolet (near‑UV) at a central wavelength of 360 nm, using the blue channel of the Large Binocular Camera (LBC) mounted on the Large Binocular Telescope (LBT). By covering an effective area of roughly 0.4 square degrees and reaching a 30 % completeness limit at U = 27 (Vega) – equivalent to U = 27.86 AB – the authors deliver the deepest and most extensive near‑UV galaxy counts to date for this wavelength and sky area.

Observations and Data Reduction
The observations were carried out over several nights between 2015 and 2017, accumulating more than 30 hours of exposure time. The LBC’s large field of view (≈ 23′ × 23′) and fine pixel scale (0.225″ pixel⁻¹) allowed the team to mosaic four independent fields, thereby mitigating cosmic‑variance effects. Standard reduction steps (bias subtraction, flat‑fielding, dark correction, sky background modeling) were performed with the dedicated LBC pipeline, and source extraction was conducted using SExtractor with a detection threshold of 1.5σ over a minimum of five contiguous pixels.

Completeness and Reliability
Artificial‑galaxy simulations were injected into the reduced images to quantify detection completeness as a function of magnitude. The 30 % completeness threshold occurs at U = 27 (Vega); the 50 % completeness level lies at U ≈ 26.3 (Vega). These values are comparable to, or slightly deeper than, previous near‑UV surveys such as the GALEX Deep Imaging Survey, but the present work benefits from a substantially larger surveyed area, reducing statistical uncertainties.

Galaxy Number Counts
The differential galaxy counts, plotted as log N versus log S, are well described by a broken power‑law. For the bright regime (22 < U < 23.5) the slope α₁ ≈ 0.45 matches earlier wide‑field surveys (e.g., CFHTLS, Subaru Suprime‑Cam). Beyond U ≈ 23.5 the slope flattens dramatically to α₂ ≈ 0.20, indicating a turnover in the UV luminosity function where the contribution of faint star‑forming galaxies begins to saturate. This flattening ensures that the integrated UV light from galaxies converges, a crucial point for estimating the extragalactic background light (EBL).

EBL Contribution
Integrating the observed counts yields a near‑UV (360 nm) EBL intensity of roughly 12 nW m⁻² sr⁻¹. This value accounts for more than 70 % of the most recent upper limits (≈ 17 nW m⁻² sr⁻¹) derived from indirect methods such as γ‑ray attenuation. Consequently, the bulk of the near‑UV background can be explained by resolved galaxies alone, leaving little room for exotic contributors (e.g., Population III stars or decaying dark matter) within current uncertainties.

Comparison with Hierarchical CDM Models
To place the observations in a theoretical context, the authors compare their counts with predictions from two state‑of‑the‑art semi‑analytic models (SAMs) built on hierarchical cold‑dark‑matter (CDM) frameworks: GALFORM and MORGANA. Both models reproduce the bright‑end counts reasonably well, but diverge at fainter magnitudes. In the U > 23.5 regime the SAMs tend to overpredict the number of low‑luminosity galaxies, suggesting that feedback processes (supernova‑driven winds, AGN heating) are insufficiently strong in the low‑mass regime. Moreover, when the same models are confronted with K‑band (near‑infrared) counts from the literature, a tension emerges: parameter sets that fit the K‑band data underproduce the observed UV counts, while those that match the UV counts overshoot the K‑band measurements. This discrepancy points to an incomplete treatment of dust attenuation and star‑formation histories across wavelengths. The authors argue that a more sophisticated, wavelength‑dependent dust model—potentially incorporating evolving metallicities and grain size distributions—will be required to reconcile the multi‑band data.

Methodological Contributions
A noteworthy aspect of the work is the rigorous assessment of completeness using Monte‑Carlo insertion of artificial sources, which provides a transparent mapping between observed counts and the underlying true galaxy population. The authors also discuss the impact of surface‑brightness selection effects, which become increasingly important at the faint end of UV surveys. Their approach sets a benchmark for future deep UV studies, especially those planned with upcoming space‑based facilities.

Implications and Future Prospects
The results have several important implications:

1. Constraining Galaxy Evolution – The observed turnover in the UV counts directly informs the faint‑end slope of the UV luminosity function at redshifts z ≈ 0.5–2 (the redshift range most probed by 360 nm observations). This, in turn, refines estimates of the cosmic star‑formation rate density derived from UV data.

2. EBL Budget – Demonstrating that resolved galaxies account for the majority of the near‑UV EBL reduces the parameter space for exotic background contributors and strengthens the use of UV EBL measurements as a probe of intergalactic opacity to high‑energy photons.

3. Model Calibration – The tension between UV and K‑band predictions highlights the need for SAMs to incorporate more realistic feedback and dust physics. The authors suggest that future models should be calibrated simultaneously against multi‑wavelength counts, rather than fitting each band independently.

4. Design of Next‑Generation UV Telescopes – The survey’s depth and area illustrate the scientific return achievable with a wide‑field, high‑throughput UV imager on a large aperture ground‑based telescope. For space missions such as ULTRASAT or the proposed LUVOIR UV channel, targeting a 30 % completeness at U ≈ 28 (Vega) over comparable or larger areas would enable a near‑complete census of the UV background, pushing the resolved fraction to > 95 %.

Conclusions
In summary, the authors deliver the most extensive near‑UV galaxy count dataset to date, covering 0.4 deg² down to U = 27 (Vega) with a well‑characterized completeness function. The counts follow a double power‑law, with a pronounced flattening at U > 23.5 that guarantees convergence of the UV contribution to the extragalactic background light. Comparison with contemporary hierarchical CDM semi‑analytic models reveals that while the bright end is well reproduced, the faint end remains problematic, underscoring deficiencies in feedback and dust modeling. The work not only refines the UV EBL budget but also provides a critical benchmark for future UV surveys and for the development of more physically realistic galaxy formation models.