The PRIMVS Catalogue - Periodic

The VISTA Variables in the Vía Láctea (VVV) survey provides an unparalleled near-infrared view of the Galactic bulge and inner disk, where optical variability studies are hampered by dust extinction. Infrared periodic variables are powerful probes of stellar evolution, Galactic structure, and distance scales in obscured regions. We introduce the PeRiodic Infrared Milky-way VVV Star-catalog (PRIMVS), designed to deliver a reliable and largely complete census of periodic near-IR variables across the bulge and southern disk, with highest completeness for periods above ~1 day, enabling population studies and structural mapping where extinction is severe. We analyzed multi-epoch Ks light curves from VVV/VVV Infrared Astrometric Catalogue (VIRAC) with rigorous quality cuts, cleaning, and detrending. Periods are measured with Lomb–Scargle, Phase Dispersion Minimization, Conditional Entropy, and Gaussian-process models. A unified false-alarm probability (FAP) provides cross-method significance. Machine-learning classification uses gradient-boosted trees trained on the third Gaia Data Release (DR3) cross-matches; feature embeddings support validation and exploration. PRIMVS processes 86.5 million cleaned light curves; 5,161,222 sources show robust periodicity (FAP < 0.3). The catalogue recovers ≈97% of previously reported VVV periodic variables and substantially extends completeness for periods P ≳ 1 d, notably for eclipsing binaries, long-period variables (LPVs), and embedded young stellar objects; completeness is reduced at shorter periods (P ≲ 1 d) owing to aliasing effects on the false-alarm estimator. Classical sequences (RR Lyrae, Cepheids, and LPVs) appear at the expected loci in period–amplitude space, and their spatial distributions trace old, metal-poor bulge populations and young thin-disk components. The near-IR baseline yields improved sensitivity to obscured populations relative to optical surveys. PRIMVS constitutes the largest homogeneous catalogue of periodic near-infrared variables in the Milky Way to date, with highest completeness at periods above ~1 day. It provides a foundation for mapping structure in high-extinction regions, refining IR distance indicators, and testing evolutionary pathways of variable stars. Future releases will incorporate quasi-periodic sources more fully, improve short-period recovery, extend multi-wavelength cross-matches, and deploy enhanced classification schemes.