[Paper Review] The Catalina Real-Time Transient Survey (CRTS)

S. G. Djorgovski, A. J. Drake+10 more|arXiv (Cornell University)|Feb 24, 2011

Astronomical Observations and Instrumentation46 citations

TL;DR

The Catalina Real-Time Transient Survey (CRTS) leverages data from the Catalina Sky Survey to detect and classify astrophysical transients and variables in real time using image subtraction and catalog comparison. It enables immediate public dissemination of transients, leading to discoveries of supernovae, cataclysmic variables, flaring stars, and eclipsing white dwarf systems, with over 3,000 transients cataloged and a key role in preparing for future surveys like LSST.

ABSTRACT

Catalina Real-Time Transient Survey (CRTS) is a synoptic sky survey uses data streams from 3 wide-field telescopes in Arizona and Australia, covering the total area of ~30,000 deg2, down to the limiting magnitudes ~ 20 - 21 mag per exposure, with time baselines from 10 min to 6 years (and growing); there are now typically ~ 200 - 300 exposures per pointing, and coadded images reach deeper than 23 mag. The basic goal of CRTS is a systematic exploration and characterization of the faint, variable sky. The survey has detected ~ 3,000 high-amplitude transients to date, including ~ 1,000 supernovae, hundreds of CVs (the majority of them previously uncatalogued), and hundreds of blazars / OVV AGN, highly variable and flare stars, etc. CRTS has a complete open data philosophy: all transients are published immediately electronically, with no proprietary period at all, and all of the data (images, light curves) will be publicly available in the near future, thus benefiting the entire astronomical community. CRTS is a scientific and technological testbed and precursor for the grander synoptic sky surveys to come.

Motivation & Objective

• To systematically explore the time domain of the sky using real-time transient detection from synoptic surveys.
• To overcome the bottleneck in follow-up observations by enabling immediate public access to transient data.
• To develop automated classification techniques for transients using citizen science and machine learning.
• To serve as a technological and scientific precursor to large-scale surveys like LSST and SKA.
• To identify rare and extreme variable objects, including flaring dwarfs and eclipsing white dwarfs, through high-cadence imaging.

Proposed method

• Processes data from three wide-field telescopes (CSS, SSS, MLS) with 4k×4k CCDs, covering ~30,000 deg² with 19–21.5 mag depth.
• Compares real-time source catalogs to median-stacked baseline images (months to years old) to detect brightness changes.
• Applies a high contrast threshold (≥1 mag change, ≥5σ) to prioritize dramatic transients.
• Uses image subtraction techniques to detect new or variable sources, especially in dense fields.
• Publishes all detected transients immediately via VO-compliant protocols and alerts (ATel, CBET) with no proprietary period.
• Employs a citizen science platform, SkyDiscovery.org, to improve automated classification and artifact rejection.

Experimental results

Research questions

• RQ1 What types of astrophysical transients can be reliably detected in real time using existing synoptic survey data?
• RQ2 How effective is real-time open data dissemination in enabling rapid follow-up and discovery across the astronomical community?
• RQ3 Can high-cadence imaging detect rare events such as short-duration flares in late-type stars or eclipses by low-mass companions?
• RQ4 How do automated classification methods, including citizen science, improve transient identification and reduce false positives?
• RQ5 To what extent can CRTS serve as a testbed for future large-scale time-domain surveys like LSST?

Key findings

• CRTS has cataloged approximately 3,000 distinct transients, with many detected multiple times, using a contrast threshold of ≥1 mag and ≥5σ significance.
• Over 100 flares from UV Ceti-type stars were discovered, including events rising more than 5 mag in less than an hour, due to the 10-minute cadence.
• More than a dozen eclipsing white dwarf systems with low-mass companions were identified through archival analysis of over 10,000 white dwarfs.
• The survey identified a new FU Orionis-type star that brightened from V ≈ 14.8 to V ≈ 12.6 mag over 4.8 years, with images showing its brightening from 18 Jan 2005 to 10 Nov 2009.
• The open-data policy enabled rapid classification and publication of transients, with ~10% of interesting sources issued as ATel and CBET notices.
• The survey demonstrated that high-amplitude variability from flare stars is a major contaminant for future surveys like LSST, but can be effectively identified using archival data and light curve morphology.