研究目的
To develop and demonstrate a difference image analysis pipeline for extracting precision light curves from TESS full-frame images, addressing challenges like crowding and variable PSF, and to achieve the mission's noise floor specification.
研究成果
The DIA pipeline successfully achieves the TESS noise floor of 60 ppm hr^{-1/2} and handles the challenges of wide-field imaging with variable PSFs. It demonstrates recoverability of astrophysical signals like transits and variables, with performance independent of detector position. The pipeline is provided as open-source and will be used for real TESS data, with plans for public data release and further improvements.
研究不足
The pipeline may not fully capture all systematics of real TESS data, as it was tested on simulated images. Detrending routines may not remove all artifacts, with ～2% of stars showing residual systematics. The kernel size and stamp selection could be optimized, and additional sky subtraction might improve results. Computational resources limit the speed of processing large data sets.
1：Experimental Design and Method Selection:
The study uses a difference image analysis (DIA) pipeline with a δ-function kernel to handle non-Gaussian PSFs in TESS FFIs, based on methods from Alard & Lupton (1998), Bramich (2008), and Miller et al. (2008). The pipeline includes routines for background subtraction, image alignment, master frame creation, image subtraction, aperture photometry, and trend removal.
2：8). The pipeline includes routines for background subtraction, image alignment, master frame creation, image subtraction, aperture photometry, and trend removal.
Sample Selection and Data Sources:
2. Sample Selection and Data Sources: Simulated TESS FFIs from NASA's 'End-to-End 6' (ETE-6) data set were used, which model stars from the TESS Input Catalog (TIC) and include injected astrophysical signals. Data were selected from a single CCD (CCD #2 on Camera 2) for testing.
3：List of Experimental Equipment and Materials:
The pipeline is implemented in IDL and PYTHON, with a C component for differencing. It was run on Intel Quad-Core Xeon 2.33 GHz/2.8 GHz processors. No specific physical equipment is mentioned beyond computational resources.
4：33 GHz/8 GHz processors. No specific physical equipment is mentioned beyond computational resources.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Steps include background subtraction using thin plate splines, image alignment with WCS, master frame creation by median combining images, image subtraction with a δ-function kernel, aperture photometry with a 2.5-pixel radius, and detrending using ensemble light curves. The process was applied to 1348 images over ～4.5 days.
5：5-pixel radius, and detrending using ensemble light curves. The process was applied to 1348 images over ～5 days.
Data Analysis Methods:
5. Data Analysis Methods: Photometric precision was compared to TESS noise models. Variability metrics (e.g., Stetson J and L), Lomb-Scargle periodicity search, and box-least-square algorithm were used to identify variable stars and transits. Statistical methods included least-squares for kernel solving and noise normalization.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
