Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Detection of Faint Sources by the UltraViolet Imaging Telescope Onboard AstroSat Using Poisson Distribution of Background

B. Ananthamoorthy, Debbijoy Bhattacharya, P. Sreekumar, Swathi B

TL;DR

This work tackles faint UV source detection in UVIT images by treating the astronomical background as a Poisson process. It develops a Poisson-based background estimation and thresholding pipeline, applying it to M31 Field 13 and the SMC to reveal a substantial number of new faint sources with robust multiwavelength counterparts. The catalog is validated through cross-matches with PHAT, Gaia, and prior UVIT studies, and exhibits high completeness around 22–22.5 mag while noting reduced performance in crowded regions. Overall, the Poisson-background approach yields a 1.5–2× increase in detections, enabling a more complete UVIT catalog and the study of new UV source populations.

Abstract

We present an improved approach for constructing the UV source catalogs using observations from the UltraViolet Imaging Telescope (UVIT) onboard AstroSat, by considering the Poisson distribution of the UV background. The method is tested extensively using fields that are not crowded, the Small Magellanic Cloud (SMC) and M31 (Field 13). The results are compared with previous studies that used UVIT observations. This approach is successful in detecting fainter sources and produces a large number of new sources ($\sim 15$ to $92\%$ more). Most of the newly discovered UV sources fall in the faint end of the source distribution (m $\gtrsim 22$). The counterparts at other wavelengths are identified for most sources. This approach is more efficient for source detection and provides an opportunity to explore new classes of UV sources.

Detection of Faint Sources by the UltraViolet Imaging Telescope Onboard AstroSat Using Poisson Distribution of Background

TL;DR

This work tackles faint UV source detection in UVIT images by treating the astronomical background as a Poisson process. It develops a Poisson-based background estimation and thresholding pipeline, applying it to M31 Field 13 and the SMC to reveal a substantial number of new faint sources with robust multiwavelength counterparts. The catalog is validated through cross-matches with PHAT, Gaia, and prior UVIT studies, and exhibits high completeness around 22–22.5 mag while noting reduced performance in crowded regions. Overall, the Poisson-background approach yields a 1.5–2× increase in detections, enabling a more complete UVIT catalog and the study of new UV source populations.

Abstract

We present an improved approach for constructing the UV source catalogs using observations from the UltraViolet Imaging Telescope (UVIT) onboard AstroSat, by considering the Poisson distribution of the UV background. The method is tested extensively using fields that are not crowded, the Small Magellanic Cloud (SMC) and M31 (Field 13). The results are compared with previous studies that used UVIT observations. This approach is successful in detecting fainter sources and produces a large number of new sources ( to more). Most of the newly discovered UV sources fall in the faint end of the source distribution (m ). The counterparts at other wavelengths are identified for most sources. This approach is more efficient for source detection and provides an opportunity to explore new classes of UV sources.

Paper Structure

This paper contains 19 sections, 3 equations, 9 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Data
  3. Analysis and Results
  4. Nature of the UVIT background
  5. Astrometry Correction
  6. Background and Threshold Calculation
  7. Background Calculation:
  8. Threshold Calculation:
  9. Source detection and photometry
  10. Catalog Parameters
  11. Discussions and Conclusions
  12. Comparison with other catalogs
  13. Consistancy with earlier work
  14. Confidence in the new sources detected
  15. Completeness
  16. ...and 4 more sections

Figures (9)

  • Figure 1: Regions used for comparison in this work. Left panel: UVIT observation of M31 Field 13 in NUVN2 filter. White dashed circles represent the central $12^{\prime}$ region, and green boxes represent the PHAT survey overlapping regions. Right panel: UVIT observations of three partially overlapping MC regions in the NUVB13 filter. Red, green, and blue correspond to SMC1, SMC2, and SMC3 fields, respectively. The inner dashed circle corresponds to the central $12^{\prime}$ regions of each field used in the current analysis.
  • Figure 2: Histogram of UVIT background distribution in $128\times128$ pixel source-free region compared with the expected Poisson and Gaussian distribution in the field of M31 Field 13 NUVN2 filter.
  • Figure 3: Ratio of $3\sigma$ equivalent threshold from Gaussian and Poisson distribution as a function of background counts.
  • Figure 4: Magnitude distribution of all detected and new sources from this work compared to L20 and D23. Also shown are sources detected only in previous studies in M31 Field 13 and SMC. Top left panel: M31 Field 13 PHAT survey overlapping region in FUVCaF2 filter compared with L20 Leahy2020 Top right panel: M31 Field 13 PHAT survey overlapping region in NUVN2 filter compared with L20. Bottom left panel: SMC in FUVBaF2 filter compared with D23 Ashish2023 Bottom right panel: SMC in NUVB13 filter compared with D23.
  • Figure 5: Gaia G magnitude in NUVB13 filter of SMC field for sources NUVB13 magnitude less than $22$ (dotted) and greater than $22$ (dash-dotted).
  • ...and 4 more figures