The Large Array Survey Telescope-Pipeline. II. Image Subtraction and Transient Detection

TL;DR

The paper describes the LAST pipeline's image subtraction and transient detection methods, validating their effectiveness with commissioning data and achieving high detection efficiency and purity.

Contribution

It presents a detailed validation of the LAST pipeline's image subtraction and transient detection component using the ZOGY algorithm and deterministic filtering.

Findings

Achieves a 5σ limiting magnitude of 20.3-20.7 mag.

Detects transients with about 80% efficiency.

Maintains over 90% purity at high S/N ratios.

Abstract

Context. The Large Array Survey Telescope (LAST) is a wide-field visual-band survey designed to explore the variable and transient sky with high cadence. Its raw data stream is automatically processed in near real time at the observatory site, producing science-quality images, catalogs, and transient alerts. Transient alerts are then reported to the Transient Name Server (TNS). Aims. The LAST pipeline comprises two major components: (i) processing and calibration of single images followed by coaddition of $20\times20$s exposures, producing single-image and coadded-image catalogs; and (ii) subtraction of coadded images from calibrated reference images followed by transient detection. In this work we present a detailed description and validation of the second component of the pipeline. Methods. Transient detection is based on the algorithm for proper image subtraction (ZOGY). We combine ZOGY subtraction with the Translient statistic for sub-pixel motion discrimination, together with a sequence of deterministic filtering steps, to produce a clean stream of transient candidates without the use of machine learning. Results. Using commissioning data, the pipeline achieves a preliminary $5\sigma$ limiting magnitude of $20.3$-$20.7$mag, a single-epoch transient detection efficiency of $\sim80$%, and a purity of $\gtrsim90$% at signal-to-noise ratio of $\geq7.5\sigma$.