Satellite Streak Photometry for the NASA Landolt Mission Using PSF-Based MCMC Modeling

Abstract

The NASA Landolt Space Mission, led by George Mason University, will deploy an artificial star in space to aid in calibrating ground-based telescopes. The mission has a level one science requirement to measure flux brightnesses to 0.75% accuracy, beyond the current state of the art. This will help scientists gain a better understanding of the current challenges faced in astrophysics, including the rate of the Universe's expansion. Because the Landolt Mission spacecraft will move across the night sky, for some observing modes the satellite will appear as a streak in long-exposure images captured by ground-based telescopes associated with the mission. For other observing modes, the stars will streak as well. For this reason, traditional circular aperture photometry cannot accurately measure the total brightness of these streaked images, which are integral for calibrating ground-based telescopes. This project investigates the use of a Point Spread Function (PSF)-based Markov Chain Monte Carlo (MCMC) fitting approach to recover the physical properties and fluxes of satellite and stellar streaks. More specifically, this project used the emcee sampler and the streaktools Python packages. First, we validated the methodology using simulated streaks with known input parameters to evaluate flux recovery accuracy and characterize the performance of the fitting algorithm. The analysis was then extended to real astronomical images containing satellite streaks, where the true flux is unknown. Throughout the project, improvements were made to enhance MCMC walker initialization, parameter bounds, and convergence diagnostics to increase reliability. The developed PSF-based MCMC framework provides a reliable method for recovering satellite streak brightness, and it can support future analyses of multiple streaks and automated detection for the NASA Landolt Mission to achieve its level one science requirements.