Light Streak Photometry and Streaktools

Abstract

The accuracy of photometric calibration has gradually become a limiting factor in various fields of astronomy, limiting the scientific output of a host of research. Calibration using artificial light sources in low Earth orbit remains largely unexplored. Here, we demonstrate that photometric calibration using light sources in low Earth orbit is a viable and competitive alternative/complement to current calibration techniques, and explore the associated ideas and basic theory. We present the publicly available Python code Streaktools as a means to simulate and perform photometric calibration using real and simulated light streaks. Using Streaktools, we perform "pill" aperture photometry on 131 simulated streaks, and Markov chain Monte Carlo based point-spread-function (PSF) model-fitting photometry on 425 simulated streaks in an attempt to recover the magnitude zeropoint of a real exposure of the Dark Energy Camera instrument on the Blanco 4 m telescope. Our results show that calibration using pill photometry is too inaccurate to be useful, but that PSF photometry is able to produce unbiased and accurate (1{\sigma} error = 3.4 mmag) estimates of the zeropoint of a real image in a realistic scenario, with a reasonable light source. This demonstrates that light-streak photometry is a promising alternative and complement to established techniques, which should be explored and tested further.