Full adaptive optics images of asteroids Ceres and Vesta; Rotational poles and triaxial ellipsoid dimensions

Adaptive optics (AO) images of asteroids Ceres and Vesta were obtained on September 18-20, 1993, with the 1.5-m telescope at the Starfire Optical Range of the USAF Phillips Laboratory located near Albuquerque, NM. The light source for higher-order wavefront correction was a Rayleigh laser beacon focused at a range of 10.5 km generated by a copper vapor laser. On April 27 and May 11, 1996, Vesta was again imaged, this time at its perihelic opposition using Vesta itself as the beacon for the AO. Images obtained at an effective wavelength of 0.85 mu m were analyzed with a new reconstruction technique, called parametric blind deconvolution. The technique allows fits for relevant parameters in the frequency domain, where the convolution of the asteroid ellipse with the variable Lorentz-shaped point spread function produced by the AO system during these observations can be separated into a multiplication of analytic functions. The triaxial ellipsoid dimensions and rotational pole (with a two-fold ambiguity) of Ceres were obtained from 17 images and found to be in excellent agreement with its 1983 stellar occultation outline (Millis et al., 1987, Icarus 72, 507-518) and the 1991 AO image produced by the COME-ON system (Saint-Pe et al., 1993, Icarus 105, 271-281). A similar analysis of eight images of Vesta in 1993, 19 images in April 1996, and 28 images in May 1996, and combined with results from four previous speckle interferometry (SI) sets, yield dimensions and a rotational pole in reasonable agreement with the Hubble Space Telescope's results (Thomas ef al. 1997). Differences may arise from a non-alignment between principal axes of inertia and the spin axis. The AO data from May 11, 1996, is of sufficient quality that Vesta's lightcurve can be decomposed into its cross-sectional area and surface brightness components, proving analytically that Vesta's lightcurve minimum is caused by a dark hemisphere. Combining the AO and SI observations with previous lightcurves yields a sidereal period of 0.22258874 days with an uncertainty of 4 in the last decimal place (3.5 ms) and shows that lightcurve minimum occurs 6 degrees before a maximum in cross-section area. (C) 1998 Academic Press.