KMTNet Synoptic Survey of Southern Sky

Figure: All-sky map in galactic coordinates displaying KS4's footprint on the sky.

Overview

Description

This is the first public data release (DR1) from the KMTNet Synoptic Survey of the Southern Sky (KS4). KS4 is a deep, four-band (B, V, R, I) imaging survey covering a large area of the southern hemisphere (-85° < Dec < -28.8°). The survey was conducted using the Korea Microlensing Telescope Network (KMTNet), which consists of three identical 1.6-m telescopes located at Cerro Tololo Inter-American Observatory (CTIO) in Chile, South African Astronomical Observatory (SAAO) in South Africa, and Siding Spring Observatory (SSO) in Australia.

The primary science goal of KS4 is to provide deep, wide-field reference images to aid in the identification of optical counterparts to gravitational wave (GW) events. This DR1 provides science-ready data products for approximately 4,000 square degrees.

The data release includes deep co-added images reaching a 5σ limiting magnitude of 22-23.5 AB mag. It also provides two complementary source catalogs containing over 279 million unique objects:

• idual_master (228.1 million objects): An I-band-selected, forced-photometry catalog optimized for consistent color measurements.

• single_master (279.9 million objects): A band-merged catalog based on independent detections in each filter, optimized for completeness.

For a full description of the data release, the processing pipeline, and data validation, please see Chang et al. (2024, in prep) and Jeong et al. (2024, in prep).

Figure: Multi-Order Coverage map of the KS4 DR1 sky coverage, presented in a polar projection centered on the South Celestial Pole (Decl. = -90°).

Survey Summary Table

KS4 DR1 Summary

Parameter	Value
Area Covered	~4,000 deg²
Bands	B, V, R, I
Median 5σ Depth	B = 22.72 AB mag, V = 22.56 AB mag, R = 22.75 AB mag, I = 22.07 AB mag
Median Seeing (FWHM)	B=2.10", V=2.03", R=1.94", I=1.93"
Number of Fields	979
Total Exposures	17,626
Unique Objects	Over 200 million (SNR > 5) (High-confidence sources: 116,393,930 in B, 140,896,979 in V, 195,153,754 in R, and 202,256,982 in I)
Astrometric Accuracy	0.125" (modal offset vs. Gaia DR3, no proper motion correction)
Photometric Homogeneity	Median offsets for point sources within ±0.03 mag (vs. Gaia XP < 19 ABmag)

Figure: Essential statistics on depth, seeing, and astrometry for each of the bands covered in this survey (B, V, R, I).

Data Reduction & Calibration Table

Data Reduction and Calibration

The KS4 DR1 data products were generated using a dedicated pipeline (described in Jeong et al., in preparation), which builds upon standard KASI pre-processing. The key steps are summarized below.

Step	Description
1. Pre-processing	Standard KASI pipeline (Overscan correction, Dark subtraction, Flat-fielding, Crosstalk removal).
2. Quality Assurance (QA)	Individual frames were filtered based on image quality metrics (e.g., FWHM > 6", ELONGATION > 1.9) to reject poor-quality data.
3. Astrometry	Astrometric solutions were computed using SCAMP (initially with UCAC-4) and validated against the Gaia EDR3 catalog to achieve sub-pixel accuracy.
4. Photometric Calibration	A two-stage process: (1) Initial ZP homogenization of single frames using APASS DR9 (BVR) & SkyMapper DR3 (I). (2) A final 2D spatial correction was applied to stacked images using Gaia XP synthetic photometry.
5. Bad Pixel Masking	Masks were generated for each frame to flag cosmic rays, cross-talk, bleeding from saturated stars, and static CCD defects.
6. Image Stacking	Calibrated frames were combined using SWarp with a median-combine algorithm and tangential projection to create deep, co-added images.
7. Catalog Generation	SExtractor was run in two modes: (1) Dual-mode with I-band detection for forced photometry (idual_master). (2) Single-band independent detection for a merged catalog (single_master).
8. Post-processing	Spurious sources (e.g., from satellite trails, stray light) were identified using the DBSCAN algorithm and removed. A magnitude-dependent bias correction was applied to MAG_AUTO.

Application

One of the primary scientific applications of KS4 is to support time-domain astronomy and the identification of optical transients. The KS4 DR1 reference images can be subtracted from newly acquired KMTNet images to search for transient events such as gravitational-wave counterparts or gamma-ray burst afterglows. To support such applications, we also provide a real-time image-reduction and transient-identification pipeline, which is publicly available through the GitHub repository.

Figure: Preview of data access. Visit here to explore the data.