arXiv subjects

V. Petrecca

Publications and source records attributed to V. Petrecca.

6 recordsLinked to original sources

arXiv2026

VAR-PZnn: A machine-learning framework for AGN photometric redshifts using color and variability-based features

Photometric redshift estimation for active galactic nuclei (AGNs) remains a fundamental challenge for current and upcoming large-scale photometric surveys. Traditional spectral energy distribution (SED) fitting suffers from color-redshift degeneracies, particularly for AGNs whose power-law continua hide the strong spectral features required to anchor redshift estimates. While AGN variability provides additional constraining power, existing frameworks require multi-band light curves that are not always available. This work presents VAR-PZnn, a fully connected mixture density network that integrates 26 variability features extracted from ZTF g-band light curves with optical photometry from Pan-STARRS1, mid-infrared (MIR) photometry from CatWISE, and, for a subsample, NIR photometry from UKIDSS. The model is trained and tested on 72,728 spectroscopically confirmed AGNs/QSOs spanning 0.01 < z < 4.5 and g-band magnitudes from 17 to 21.5. For the main sample, we achieve σ_{NMAD} = 0.058 and an outlier fraction of η= 8.2%, which reduces to 5.4% when the 10% of sources with the highest predicted uncertainty are excluded. An ablation study demonstrates that MIR photometry provides the dominant constraint for photo-z accuracy, while variability features serve as a secondary refiner. Using UKIDSS NIR data as a proxy for future synergies between LSST and space-based missions like Euclid and Roman, we obtain η= 13.3% without MIR data and η= 4.6% when MIR is available. We benchmark against Low-Resolution Templates (LRT) SED fitting (η= 28.7%) and the VAR-PZ framework; applying single-band VAR-PZ priors worsens LRT performance to η= 39.4% due to single-band light-curve degeneracies, confirmed via simulations (η= 27.6% to 28.1%). This framework provides a scalable approach for the Legacy Survey of Space and Time (LSST).

astro-ph.GA↗

arXiv2026

Toward decision-aware AI for LSST-scale time-domain astronomy

The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) will generate approximately (10^7) alerts per night, pushing time-domain astronomy beyond pipelines that treat discovery as a static labeling problem. We argue that LSST is better understood as a partially observed dynamical environment, in which scientific return depends on the quality of follow-up decisions made under uncertainty and finite observational resources. The central challenge is therefore to maintain evolving, uncertainty-aware representations of astrophysical sources and to select actions that maximize long-term scientific value. We propose that foundation models trained on heterogeneous time-domain data can learn survey-scale representations of source state, while decision-theoretic policies support principled, auditable allocation of follow-up resources. Embedded within human-supervised agentic systems, these components position AI as part of the operational inference loop rather than as a downstream predictive tool. The way such systems represent belief, optimize utility, and expose their reasoning will shape observational efficiency, the distribution of scientific agency, including who participates in discovery and the scientific questions that receive priority.

astro-ph.IM↗

arXiv2026

AGILE: an end-to-end Rubin-LSST simulation of AGNs, galaxies, and stars I. Software description and first data release

Contemporary large-scale surveys such as the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) and Euclid present an unprecedented discovery potential for studying AGNs at the population level in the big data era. However, one major challenge is the accurate identification and classification of AGNs from optical/NIR photometry, or variability data alone. In order to optimize AGN selection, classification, and systematics, as well as to test different data analysis tools, we present AGILE (AGNs In the LSST Era), an LSST end-to-end simulation software. AGILE -- developed as part of the INAF LSST in-kind contribution -- is capable of simulating the anticipated AGN population in LSST and Euclid. We based AGILE on existing simulations of galaxies and stars, while we developed an AGN recipe based on empirical relations. AGILE populates complete galaxy samples with AGNs according to the observed AGN accretion rate distribution, and each AGN is assigned an optical/UV spectral energy distribution. Optical AGN variability is added using a damped random walk model connected to the AGN physical parameters. Finally, AGILE creates both LSST-like images and related data products. Using AGILE, we build a $24$ deg$^2$ complete mock truth catalog of AGNs, galaxies, and stars with $0.2 < z < 5.5$, $\log M/M_\odot > 8.5$ (AGNs and galaxies), and $r < 27.5$ mag (stars). We perform a pilot simulation (AGILE DR1) consisting of $1$ deg$^2$ of LSST operations in the COSMOS field observed up to three years according to the survey strategy. We use AGILE DR1 to quantify the accuracy of the LSST Science Pipelines in recovering true fluxes of AGNs, galaxies, and stars. We quantify the LSST completeness and purity in recovering Type 1 AGNs using typical color-color and variability selections. We share the AGILE DR1 dataset, an ideal test-bench for further scientific exploitation.

astro-ph.GA↗

arXiv2026

Euclid: An automated system to match Rubin transient alerts to Euclid observations

The Vera C. Rubin observatory is expected to produce 10 million transient alerts per night in ugrizy filters, whilst Euclid is a visible to near-infrared space telescope engaged in a wide field survey. We present a prototype system to automatically match the transient alerts from Rubin to Euclid observations. The system produces joint light-curves containing both visible and near-infrared photometry, and joint image cutouts. Using Zwicky Transient Facility alerts as a proxy for Rubin, we demonstrate the system in use in cases where Euclid did and did not detect the transient and highlight the value that can be added in each case. For transients detected by Euclid these benefits include identifying the supernovae (SNe) in observations taken prior to ground-based detection, thereby better constraining the explosion time, such as SN 2024pvw detected ~3 d prior to ground based detections. In cases where Euclid did not detect the transient, we demonstrate the benefit of adding Euclid observations to improve host morphology measurements and associations.

astro-ph.IM↗

arXiv2025

Euclid: Quick Data Release (Q1) -- Photometric studies of known transients

We report on serendipitous Euclid observations of previously known transients, using the Euclid Q1 data release. By cross-matching with the Transient Name Server (TNS) we identify 164 transients that coincide with the data release. Although the Euclid Q1 release only includes single-epoch data, we are able to make Euclid photometric measurements at the location of 161 of these transients. Euclid obtained deep photometric measurements or upper limits of these transients in the $I_E$, $Y_E$, $J_E$, and $H_E$ bands at various phases of the transient light-curves, including before, during, and after the observations of ground-based transient surveys. Approximately 70\% of known transients reported in the six months before the Euclid observation date and with discovery magnitude brighter than 24 were detected in Euclid $\IE$ images. Our observations include one of the earliest near-infrared detections of a Type~Ia supernova (SN 2024pvw) 15 days prior to its peak brightness, and the late-phase (435.9 days post peak) observations of the enigmatic core-collapse SN 2023aew. Euclid deep photometry provides valuable information on the nature of these transients such as their progenitor systems and power sources, with late time observations being a uniquely powerful contribution. In addition, Euclid is able to detect the host galaxies of some transients that were previously classed as hostless. The Q1 data demonstrate the power of the Euclid data even with only single-epoch observations available, as will be the case for much larger areas of sky in the Euclid Wide Survey.

astro-ph.SR↗

arXiv2024

The extremes of AGN variability: outbursts, deep fades, changing looks, exceptional spectral states, and semi-periodicities

The extremes of Active Galactic Nuclei (AGN) variability offer valuable new insights into the drivers and physics of AGN. We discuss some of the most extreme cases of AGN variability; the highest amplitudes, deep minima states, extreme spectral states, Seyfert-type changes, and semi-periodic signals, including new X-ray observations. The properties of changing-look (CL) AGN are briefly reviewed and a classification scheme is proposed which encompasses the variety of CL phenomena; distinguishing slow and fast events, repeat events, and frozen-look AGN which do not show any emission-line response. Long-term light curves that are densely covered over multiple years, along with follow-up spectroscopy, are utilized to gain insight into the underlying variability mechanisms including accretion disk and broad-line region physics. Remarkable differences are seen, for instance, in the optical spectral response to extreme outbursts, implying distinct intrinsic variability mechanisms. Furthermore, we discuss methods for distinguishing between CL AGN and CL look-alike events (tidal disruption events or supernovae in dense media). Finally, semi-periodic light curve variability is addressed and the latest multiwavelength (MWL) light curve of the binary supermassive black hole (SMBH) candidate OJ 287 from the MOMO project is presented. Recent results from that project have clearly established the need for new binary SMBH modelling matching the tight new constraints from observations, including the measurement of a low (primary) SMBH mass of ~10^8 Msun which also implies that OJ 287 is no longer in the regime of near-future pulsar timing arrays.

astro-ph.HE↗