Simulating point spread functions for the James Webb Space Telescope with WebbPSF

Abstract

Experience with the Hubble Space Telescope has shown that accurate models of optical performance are extremely desirable to astronomers, both for assessing feasibility and planning scientific observations, and for data analyses such as point-spread-function (PSF)-fitting photometry and astrometry, deconvolution, and PSF subtraction. Compared to previous space observatories, the temporal variability and active control of the James Webb Space Telescope (JWST) pose a significantly greater challenge for accurate modeling. We describe here some initial steps toward meeting the community's need for such PSF simulations. A software package called WebbPSF now provides the capability for simulating PSFs for JWST's instruments in all imaging modes, including direct imaging, coronagraphy, and non-redundant aperture masking. WebbPSF is intended to provide model PSFs suitable for planning observations and creating mock science data, via a straightforward interface accessible to any astronomer; as such it is complementary to the sophisticated but complex-to-use modeling tools used primarily by optical designers. WebbPSF is implemented using a new exible and extensible optical propagation library in the Python programming language. While the initial version uses static precomputed wavefront simulations, over time this system is evolving to include both spatial and temporal variation in PSFs, building on existing modeling efforts within the JWST program. Our long-term goal is to provide a general-purpose PSF modeling capability akin to Hubble's Tiny Tim software, and of sufficient accuracy to be useful to the community.