20 years of Hubble Space Telescope optical modeling using Tiny Tim

TLDR: The Tiny Tim PSF simulation software package has been the standard HST modeling software since its release in early 1992 as mentioned in this paper, and has been used extensively for HST data analysis.

Abstract: Point spread function (PSF) models are critical to Hubble Space Te lescope (HST) data analysis. Astronomers unfamiliar with optical simulation techniques need access to PSF models that properly match the conditions of their observations, so any HST modeling software needs to be both easy-to-use and have detailed information on the telescope and instruments. The Tiny Tim PSF simulation software package has been the standard HST modeling software since its release in early 1992. We discuss the evolution of Tiny Tim over the years as new instruments and optical properties have been incorporated. We also dem onstrate how Tiny Tim PSF models have be en used for HST data analysis. Tiny Tim is freely available from tinytim.stsci.edu. Keywords: Hubble Space Telescope, point spread function 1. INTRODUCTION The point spread function (PSF) is the fundamental unit of imag e formation for an optical syst em such as a telescope. It encompasses the diffraction from obscurations, which is modified by aberrations, and the scattering from mid-to-high spatial frequency optical errors . Imaging performance is often described in terms of PSF properties, such as resolution and encircled energy. Optical engineering software, includi ng ray tracing and physical optic s propagation packages, are employed during the design phase of the system to predict the PSF to ensure that the imag ing requirements are met. But once the system is complete and operational, the software is usually packed away and the point spread function considered static, to be described in documentation for reference by the scientist. In this context, an optical engineer runs software to compute PSFs while the user of the optical system simply needs to know its basic characteristics. For the Hubble Space Telescope (HST), that is definitely not the case. To extract the maximum information out of an observation, even the smallest details of the PSF are important. Some examples include: deconvolvin g the PSF from an observed image to remove the blurring caused by diffraction and reveal fine structure; convolving a model image by the PSF to compare to an observed one; subtracting the PSF of an unresolved source (star or compact galactic nucleus) to reveal extended structure (a circumstellar disk or host galaxy) that would otherwise be unseen within the halo of diffracted and scattered light; and fitting a PSF to a star imag e to obtain accurate photometry and astrometry, especially if it is a binary star with blended PSFs Compared to ground-based telescopes HST is extremely stable, so the structure in its PSF is largely time-invariant. This allows the use of PSF models for data analysis. On the ground, the variable PSF structure due to the atmosphere and thermal-and-gravitationally-induced optical perturbations make it more difficult to produce a model that accurately matches the data. The effective HST PSF, though, is dependent on many parameters, including obscurations, aberrations, pointing errors, system wavelength response, object color, and detector pixel effects. An accurate PSF model must account for all of these, some of which may depend on time (focus, obscuration positions) or on field position within the camera (aberrations, CCD detector charge diffusion, obscuration patterns, geometric distortion). 1.1 Early HST PSF modeling: TIM Before launch of HST in 1990, a variety of commercial and proprietary software packages were used to compute PSFs. These provided predictions of HST’s imaging performance and guided the design, but they were not used by future HST observers. These programs were too complicated for general HS T users, and either were not publicly available or were too expensive. They also did not provide PSF models in forms that scientists would find useful, such as including the effects of detector pixelization and broadband system responses.