Validating the On-axis Point Spread Function Reconstruction for Keck Adaptive Optics

The largest telescopes (8–10 m) in the world regularly deploy adaptive optics (AO) systems utilizing natural and laser guide stars in tandem with deformable mirrors to correct the turbulent effects of the atmosphere and obtain Point Spread Functions (PSF) with Full Width at Half Maximum (FWHM) of less than 100 mas. While classical techniques seek to extract the PSF from the images, reconstructing the PSF from first principles will allow resolving high angular resolution phenomena in the absence of an unresolved point source in the science field. We use PRIME, a python package for PSF reconstruction and modelling to forward model the PSFs for NIRC2 data from the Keck II telescope’s AO system. We also use PRIME to fit the PSFs to a Fourier model and compare the performance between the two approaches by considering the Strehl, FWHM and the Fraction of Variance Unexplained. We then discuss the implications of our analysis for the upcoming Keck All-sky Precision Adaptive Optics upgrade on the Keck I telescope.