ASAS-SN Light Curves Overview

Updated 30 July 2025

ASAS-SN light curves are uniformly-calibrated, all-sky photometric records with 2–5 year baselines and ~2–3 day cadence, enabling diverse temporal studies.
They employ advanced methods like image subtraction, forced aperture photometry, and rigorous error modeling ensuring robust and precise measurements.
Their extensive, accessible databases support automated classifications, citizen science projects, and detailed period analyses in time-domain research.

The All-Sky Automated Survey for Supernovae (ASAS-SN) light curves constitute a foundational resource for modern time-domain astronomy, providing calibrated, uniformly-sampled photometric histories for millions of celestial objects across the entire sky. Designed originally to discover nearby supernovae to completeness, ASAS-SN’s data products now underpin a broad range of studies on variable stars, transients, binaries, and extragalactic sources. ASAS-SN light curves are characterized by their long time baselines (~2–5 years or more), broad coverage (down to V~17 or g~18.5 mag), and systematic reduction pipelines that use image subtraction, forced aperture photometry, and rigorous calibration against all-sky standard catalogs.

1. Observational Strategy and Data Acquisition

ASAS-SN achieves all-sky, high-cadence monitoring via an international network of small robotic telescopes, each equipped with wide-field optics. The principal observational parameters are:

Cadence: Each field is observed with a nominal 2–3 day cadence (sometimes higher after upgrades), yielding ~100–500 epochs per source over the survey duration (Jayasinghe et al., 2018).
Depth: Photometric sensitivity reaches V~17 mag (or g~18.5 mag in later, deeper data; see (Christy et al., 2021)).
Field of View and Pixel Scale: Each telescope array covers ~4.5 deg² per camera, with an 8″ pixel scale and FWHM of ~2 pixels (Kochanek et al., 2017).
Exposure: Nominal exposures of 90s with multiple cameras provide redundancy and improve S/N per field.
Photometric Calibration: Zero-points are set using cross-matches with standard catalogs (e.g., APASS for V-band; Refcat2 or Gaia for g-band) with careful treatment of atmospheric extinction, and iterative rejection of outliers (Kochanek et al., 2017, Jayasinghe et al., 2019).

Image subtraction, based on the ISIS pipeline, is applied prior to photometry to isolate genuine variability from static sky sources and background. Photometry is performed on difference images using a 2-pixel aperture, with background determined from an annulus and sigma-clipping for robust estimation.

2. Data Reduction, Light Curve Construction, and Error Handling

Multiple quality control and statistical methods are implemented to produce scientific-grade light curves:

Quality Control: Images affected by poor focus, weather, or astrometric mismatch are excluded. Zero point offsets between cameras are harmonized, and the “primary” camera for each object is used as the baseline (Jayasinghe et al., 2018).
Error Modeling: Errors are initially derived from photon statistics, then empirically re-scaled using the reduced χ² statistic for calibration stars, fitting the χ²/N_DOF vs. magnitude distribution to ensure error bars realistically reflect observed scatter (Jayasinghe et al., 2018, Jayasinghe et al., 2018).
Systematics: A systematic error floor of ~0.02 mag is adopted for bright sources, with a polynomial model fitted for fainter sources to capture additional scatter (Jayasinghe et al., 2018).
Database Products: ASAS-SN Sky Patrol V2.0 and light curve servers provide access to precomputed, continuously updated light curves for >110 million sources—with each epoch recording magnitude, error, flux, FWHM, filter, and limiting magnitude. Forced photometry reports a 99.99 error code for non-detections (Hart et al., 2023).

3. Scientific Utility and Analysis Methodologies

ASAS-SN light curves are central to the classification and paper of variable sources owing to their time coverage and cadence. Key methodologies include:

Period Search: A suite of algorithms—Generalized Lomb-Scargle (GLS), Multi-Harmonic Analysis of Variance (MHAOV), Phase Dispersion Minimization (PDM), Box Least Squares (BLS)—is used to extract periodicities (Jayasinghe et al., 2018, Jayasinghe et al., 2018).
Feature Extraction and Machine Learning: Quantitative features (amplitude, period, color indices, light curve moments, Fourier decomposition coefficients) are used as input to random forest classifiers, both for initial classification (e.g., Upsilon [Kim & Bailer-Jones 2016], scikit-learn) and for refinement. Weighted precision and recall ~89–99% are achieved for large-scale catalogs (Jayasinghe et al., 2018, Jayasinghe et al., 2018).
Citizen Science Integration: The Zooniverse-based Citizen ASAS-SN project leverages volunteer classifications of phased light curves, with workflows designed to resolve degeneracies (e.g., half-period ambiguity in eclipsing binaries), and to flag ambiguous or non-variable (“Junk”) light curves that supplement machine learning modules (Christy et al., 2021, Christy et al., 2021).
Advanced Light Curve Models: For specialized analyses (e.g., flare energetics), empirical flare templates parameterized by the half-light time (t₁/₂) are fitted to sparse data. For binary and pulsating stars, Fourier decomposition and string-length statistics (e.g., Lafler–Kinman T(P)) are employed to discriminate classes and measure periodicity (Schmidt et al., 2018, Jayasinghe et al., 2018).

4. Types and Scope of Variability Characterized

ASAS-SN provides the largest homogeneous sample of time-resolved light curves for variable objects in the V and g bands, encompassing:

Variable Star Taxonomy: Pulsating stars (RR Lyrae, Cepheids, Miras, δ Scuti), eclipsing binaries (EA, EB, EW), red semi-regular variables (SR), rotational variables, cataclysmic variables, Be/GGCAS stars, and rare types (e.g., AM CVn, RV Tau) are included (Jayasinghe et al., 2018, Jayasinghe et al., 2018, Nere et al., 23 Aug 2024, Kato et al., 2021).
Discovery Yield: Recent catalogs include 66,179 new bright variables (Jayasinghe et al., 2018), 220,000 new variables in the southern hemisphere alone (Jayasinghe et al., 2019), and more than 1 million expected variable sources in total (Kochanek et al., 2017).
Supernovae and Transients: Early-time and long-baseline detections enable the paper of luminous SNe, pre-explosion variability, and physical inferences about the progenitor and environment of rare events (e.g., ASASSN-13dn, SN 2017hcc) (Prieto et al., 2017, Hueichapán et al., 11 Mar 2025).
Rare Systems: Multi-year, high-cadence coverage allows for detection and characterization of rare classes—AM CVn candidates via rapid, short outbursts and double superoutburst signatures (Kato et al., 2021), double periodic variables (DPVs) with dual-scale periodicities (Rosales et al., 2019), and long-period eclipsing binaries exceeding 7 years (Jayasinghe et al., 2018).

5. Infrastructure, Data Access, and Community Tools

ASAS-SN light curves are accessible via robust, high-throughput database systems and client interfaces:

Access Portals: Light curves can be retrieved on-demand using the Sky Patrol web interface or a scalable Python client, with support for cone searches, catalog cross-IDs, and custom queries up to 1 million objects per session (Hart et al., 2023).
Database Design: The architecture employs a hybrid in-memory source table (for rapid coordinate-based filtering) with disk-based document storage of light curves, providing order-of-magnitude speed improvements and near real-time updates (~1 hour latency) (Hart et al., 2023).
Visualization and API: The Python client delivers bundled LightCurveCollection objects for batch analysis, periodograms, and compatibility with time series analysis libraries, supporting the needs of both survey-scale statistical research and targeted studies.

6. Systematic Challenges and Validation

Robust classification and physical inference from ASAS-SN light curves must contend with several challenges:

Crowding and Saturation: Blending with nearby sources and saturation in V<11–12 mag sources require custom correction procedures, including Gaussian redistribution of bleed trails and blended-source amplitude suppression assessment (Kochanek et al., 2017).
Period Ambiguities: Alternating minima (notably in RV Tau stars) can bias automated period-finding algorithms toward half the true period, necessitating manual review and the use of both P and 2P phasing (Nere et al., 23 Aug 2024).
Transient Detection and False Positives: Citizen science workflows and ML “Junk” filters are increasingly used to cull artifacts, blend-induced signals, and non-astrophysical variability, with targeted training sets built on both positive and negative examples (Christy et al., 2021).

7. Scientific Impact and Future Prospects

ASAS-SN light curves are transformative for the construction of large, bias-minimized, time-domain samples that enable:

Population Studies: Homogeneously classified all-sky samples ( $>400,000$ variables) support calibration of period–luminosity relations, stellar population synthesis, and Galactic structure mapping (Jayasinghe et al., 2018, Jayasinghe et al., 2019).
Rare Transient Characterization: High-cadence coverage allows for precise timing and physical diagnosis of early SN light curve features (e.g., double power-law rises in SN Ia (Shappee et al., 2018)), interaction signatures in SNe II (Hueichapán et al., 11 Mar 2025), and the behavior of interacting binaries and cataclysmic phenomena.
Community Science: Rapid, global access to calibrated, updated light curves supports citizen science discovery work, real-time alerts, multi-messenger synergy (e.g., with gravitational wave and neutrino detection), and the development of cross-survey training sets for next-generation time-domain missions (Christy et al., 2021, Hart et al., 2023).

Continuing expansion in both database scope and analysis methodology, including real-time custom patrols, improved calibration for bright sources, and deeper integration with machine learning and citizen engagement, is expected to further increase the scientific utility of ASAS-SN light curves in the coming decade.