Catalogue of Periodic, Quasi-Periodic, and Aperiodic AGN Light-Curve Signatures from LSST DP1

Overview

This release provides the first catalogue of AGN light-curve periodicity signatures from the Vera C. Rubin Observatory Legacy Survey of Space and Time Data Preview 1, derived with the Quasar harmonic eXplorer (QhX, v0.2.0) pipeline. All quasars in this catalogue were identified via cross-matching LSST DP1 objects with major external quasar catalogues (Milliquas, Quaia, SDSS, DESI , ZTF), ensuring reliable AGN classification and ancillary redshift estimates.
The catalogue contains both rare reliable periodic candidates and the more common non-periodic cases, systematically classified into subcategories that capture pseudo-periodic, transiently coherent, and red-noise–dominated behaviours.

Two data products are provided:

• Object-level table — one row per AGN, summarizing the most significant detected period and cross-band diagnostics.
• Per-band table — one row per AGN×band, containing band-wise fit quality and phase coherence metrics.

Schema: ephemerides_objects_z.csv

- Taken from QhX formal errors when available.

- If absent, conservative fallback errors are imputed using light-curve baseline and S/N.

- json — formal QhX estimate,

- fallback_baseline — inferred from light-curve baseline and S/N,

- fallback_fractional — conservative fraction of period,

- none/no_period — no estimate available.

pseudo — pseudo-periodic features with intermittent coherence,

transient — coherence confined to specific epochs,

red_noise — red-noise–dominated false positives.

z_adopted — redhsift of object.

Column	Description
object_id	unique source identifier (from DP1)
best_period (days)	most significant period from QhX
per_err_lo / per_err_up (days)	asymmetric lower/upper errors on the detected period.
per_err_sym (days)	symmetric error proxy (average of lower/upper if available, else fallback).
per_err_source	flag describing where the error came from:
per_err_note	additional context (e.g. baseline length, S/N used in fallback).
n_good	number of bands satisfying the “good” criterion (see below).
n_coherent	number of bands satisfying the “coherent” criterion.
med_snr	median amplitude signal-to-noise ratio across bands.
med_R2	median fraction of variance explained across bands.
max_lag (cycles)	largest phase lag relative to the reference band.
baseline_days (days)	temporal baseline of the object’s light curve in DP1.
verdict	classification tier: reliable, medium, or discarded.
bucket	subcategory for non-reliable sources:
support_count	number of pairwise cross-band comparisons supporting the period.
supporting_labels	list of band-pair labels contributing to support.
sig_pairs	number of pairwise comparisons with significance ≥ 0.5.

Schema: ephemerides_bands.csv

Column	Description
object_id	matches the object table
band	LSST filter (u, g, r, i, z, y).
period_days (days)	per-band period detection (may be absent if only cross-band).
amp_mag (mag), amp_snr	fitted sinusoidal amplitude.
amp_snr	signal-to-noise ratio of the amplitude.
R²	fraction of variance explained by the sinusoidal fit.
phase_cycle	fitted phase (in cycles) within the band.
ref_band	band chosen as the phase reference.
ref_phase	phase of the reference band.
phase_lag_vs_ref (cycles)	phase lag relative to the reference band.
n_points_used	number of photometric points used in this band’s fit.
median_err (mag)	median photometric uncertainty of the light curve in this band.
good_band	flag: True if amp_snr ≥ 1.5 and R² ≥ 0.10.
coherent_band	flag: True if good band and phase lag ≤ 0.20 cycles.
verdict	carries the object-level verdict (reliable, medium, discarded).
bucket	carries the object-level subcategory (pseudo, transient, red_noise).

Verification criteria

Download diagnostic figures

Download the complete sets of figures showing folded LSST DP1 of known quasars light curves (with sinusoidal fits) and the corresponding spectral window functions for all AGN having DP1 observations in 4 bands and each light curve number of points larger than 50:

• Light_curve.zip — Folded and time–binned light curves for all objects.
• Spectrum_window.zip — Spectral window functions illustrating cadence alias structures.

Each ZIP archive contains one image per object and band. The light–curve plots show the best–fit sinusoidal model at the detected period, while the spectral windows quantify aliasing due to LSST cadence.

How to use the catalogue

1. Start from ephemerides_objects.csv to filter by classification, period range, or number of good/coherent bands.
2. Join on object_id to ephemerides_bands.csv to inspect per-band diagnostics (amplitude S/N, R², phase lags).
3. For population studies, group the object-level table by verdict or bucket, and the per-band table by band.
4. Period errors (per_err_lo, per_err_up, per_err_sym) are provided where possible; treat fallback errors (per_err_source=fallback_*) as conservative bounds rather than formal uncertainties.