Abstract
The AGN-sCAN project consists of deep VLBA observations in the COSMOS field, targeting about 500 galaxies at high resolution. The primary goals of AGN-sCAN are: (A) assessing the incidence of radio-faint AGN in massive star-forming galaxies; (B) measuring the level of contamination from radio-faint AGN to the total radio emission. These steps will be crucial to provide unbiased constraints on the composition of the faint (sub-mJy) radio sky between AGN and star-forming galaxies, paving the way to the future SKA. At least two papers are foreseen by the AGN-sCAN team: one focusing on the survey paper and the other on the impact of radio-faint AGN on radio-SFR relations. The preliminary results obtained by the team seem to confirm the incidence of radio-faint AGN extrapolated from previous works, and a negligible contamination from radio-faint AGN to the total radio emission in star-forming galaxies.
Results achieved
: The project "AGN-sCAN: zooming-in on the AGN-galaxy connection since the cosmic noon" aims to deliver an unbiased radio view of distant Active Galactic Nuclei (AGN) and galaxy star formation activity. The project relies on a 120-hour observing program (PI: I. Delvecchio) obtained with the Very Long Baseline Array (VLBA), and targeting a representative sample ~500 massive star-forming galaxies (SFGs) in the COSMOS field at 0.5<5.0. By reaching an unprecedented combination of high angular resolution (~27x7 milli-arcsec) and sensitivity (5σ limit of 25 μJy/beam), these pilot observations enable us to perform a morphological decomposition based on brightness temperature (TB), separating pure AGN-driven emission (TB ≥ 105 K) from galaxy star formation on a source-by-source basis. The primary goals are twofold: (i) accurately calibrating AGN-corrected radio-vs-star formation rate (SFR) relations by isolating radio-faint AGN previously missed by clear-cut classification criteria (e.g. radio-excess); (ii) assessing the role of environment and internal effects in shaping the triggering and incidence of radio AGN over time. As a bonus goal, given the plethora of JWST and broad-band data available for all targets, the excellent VLBA resolution allows us to search for peculiar AGN and follow them up with multi-band observations. This success was fundamentally driven by the strategic use of the project budget to recruit two full-time postdoctoral researchers who served as the primary analytical workforce: Dr. G. Peluso (INAF) and Dr. F. Ubertosi (UniBo), who led the WP1 and WP2, respectively. The high-level achievements, divided by the core project objectives, are as follows: Objective 1 (WP1): Calibrating radio-continuum emission as a SFR tracer. We successfully quantified the contribution of elusive radio-faint AGNs to the infrared-radio correlation (IRRC). By mastering wide-field VLBA data reduction for 500 targets in the COSMOS field, we achieved a highly sensitive (rms~5 μJy/beam) detection limit. Despite the low VLBA detection rate (~9%), in line with previous expectations, we accounted for the large fraction of non-detections by modeling them in the framework of the radio AGN number counts. We further demonstrated that the cumulative contribution of VLBA detections and non-detections does have a negligible impact on the radio-SFR distribution (by less than 0.1 dex). This result confirms that standard radio-continuum observations can be safely used as robust star formation rate (SFR) indicators, once radio-excess AGN are removed. This work was published in Peluso et al. (2026), in collaboration with the full AGN-sCAN team. In parallel, the full AGN-sCAN design and survey strategy will be presented in a forthcoming publication (Delvecchio et al. 2026, to be submitted), delivering the full catalog and data products. Objective 2 (WP2): Assessing the Drivers of Radio AGN Activity Over Time and Environment Led by the UniBo unit (Dr. F. Ubertosi and Prof. M. Gitti), this primary objective investigated what drives the triggering, demographics, and kinetic feedback of radio AGN populations across different galaxy stellar masses, environments, and cosmic epochs. To achieve this, the team followed a twofold approach: bridging the physical understanding of immediate environmental triggers at low redshifts (z < 0.5, WP 2.1) with a framework for interpreting high-redshift feedback in distant galaxies (0.5 < z < 3.5, WP 2.2). WP 2.1: Dissecting Immediate Environmental Triggers at Low Redshift (z < 0.5) To understand the mechanics of kinetic feedback, the UniBo unit focused on high-resolution JVLA, VLBA, Chandra, and VLT/MUSE mapping of multiphase gas in local cool-core groups and intermediate-redshift clusters. In foundational work led by the unit (Ubertosi et al. 2025a), the team analyzed 25 local (z < 0.2) cool-core systems. They discovered that while significant spatial offsets between the supermassive black hole (SMBH) radio core and the hot X-ray gas peak exist in 80 percent of these systems, the offset between the SMBH and H-alpha gas shrinks to just 15 percent. This proved that the spatial alignment of cooling warm gas directly dictates the AGN lifecycle; when the cooling gas is spatially offset, the AGN is starved of fuel and becomes quiescent. To maximize the scientific output of the broader AGN-sCAN VLBI dataset, Dr. Ubertosi is currently developing new procedures for VLBA data reduction and analysis by utilizing multi-frequency observations of the central radio AGN in the local galaxy group NGC 5044. Expanding to intermediate redshifts, Ubertosi et al. (2025b) studied the merging cluster CHIPS 1911+4455 as a test case for larger-scale environmental disturbances. The UniBo team successfully identified an "infant" AGN alongside a rapidly star-forming brightest cluster galaxy. The data confirmed that this radio AGN was recently awakened, triggered by a rapid, merger-induced fuel supply increase. WP 2.2: Demographic Framework of High-Redshift Feedback (0.5 < z < 3.5) Because high-redshift galaxies cannot be resolved with the same physical detail as local clusters, the UniBo unit has leveraged the vast multi-wavelength data of the COSMOS collaboration to conduct a multi-parametric investigation into radio AGN activity. Since October 2025, Dr. F. Ubertosi has been leading a pilot project analyzing the full sample of over 700 radio-selected AGN in the COSMOS-Web area. Building on WP1 results, the team selected all radio-excess AGN and cross-matched them with state-of-the-art group catalogs (Toni et al. 2024, 2025) across the 0.5 < z < 3.5 range. After correcting for selection purity and catalog completeness, the UniBo researchers explored the AGN fraction based on environment (group vs. field), stellar mass, and redshift. The team has successfully derived the radio AGN luminosity function in groups versus the field. This revealed a systematic prevalence of radio AGN in groups, which is also visible when separately computing radio AGN fractions for galaxies in the field and for galaxies in galaxy groups. The ultimate goal is dissecting the covariance between environment, stellar mass, and redshift to shape the incidence of radio AGN. This work (Ubertosi, Delvecchio et al., in prep) will be submitted by summer 2026, setting a crucial benchmark that bridges our knowledge of internal and external radio-AGN triggers from the local universe up to cosmic noon. Bonus Objective: Searching for peculiar objects The team expanded the initial goals of the AGN-sCAN project by leveraging the extreme resolution and astrometry of VLBA imaging for identifying unique systems and following them up with multi-wavelength facilities. The AGN-sCAN team has cross-matched COSMOS VLBA with JWST/NIRCam source catalogues for identifying offset (aka “wandering”) AGN. This search has led to a promising wandering AGN candidate at z=0.93, which has been followed-up with JWST/NIRSpec and HSA observations, at which the UniBo team has been contributing. Moreover, Dr. F. Ubertosi led as PI a deep low-frequency VLBA observation of a binary AGN candidate, aimed at distinguishing core from jet origins. To overcome unexpected data quality issues in that task, Dr. F. Ubertosi successfully proposed new VLBA observations to search for binaries in 80 local galaxy clusters. In the investigation of Little Red Dots, the UniBo team is involved in analysing newly acquired ultra-deep HSA data to test if the high-z PRIMER-COS 3866 is powered by an AGN core. Finally, the UniBo unit has contributed to the early analysis of a Direct Collapse Black Hole candidate, analysing team-led e-MERLIN observations of the Infinity Galaxy at z=1.14, in order to map this extended emission and distinguish between a fossil AGN bubble and merger-induced star formation.Project details
Unibo Team Leader: Myriam Gitti
Unibo involved Department/s:
Dipartimento di Fisica e Astronomia "Augusto Righi"
Coordinator:
Inaf - Istituto Nazionale Astrofisica(Italy)
Total Unibo Contribution: Euro (EUR) 87.789,00
Project Duration in months: 24
Start Date:
28/09/2023
End Date:
28/02/2026
