Seeing Effects and Their Calibration for Astronomical Speckle Interferometry Observations

The effects of seeing upon the techniques for recovery of high spatial resolution information are discussed. The need for calibration of the seeing to avoid incorrect interpretation of the reduced data is presented. A method which allows resolved and unresolved objects observed in variable seeing conditions to be reduced with corresponding seeing is presented and analyzed. This technique also allows study of the seeing statistics of the atmosphere. These measurements lead to observation constraints for the use of speckle interferometry as well as a refined observational protocol. The effects of non-isoplanatism, especially when applied to the imaging technique of speckle holography, are also discussed along with measurements of the spatial frequency dependent isoplanatic parameter. It is shown that the degree of isoplanicity is dependent upon spatial frequency requirements and the telescope being used. An imaging technique which appears to be seeing self-calibrating is presented. An unresolved source need not be observed to sample identical seeing statistics thus avoiding the problems discussed above. The behavior and dynamic range of this imaging technique is investigated for different seeing conditions. Results from a point source yield apparently seeing corrected images which match the theoretical Airy patterns at a number of wavelengths with an rms difference of less than 1%. The resolved red supergiant Alpha Orionis is studied using this self-calibrating imaging technique. Diameter measurements at six different wavelengths yield a consistent stellar radius when different limb darkening is taken into account. Images of Gamma Orionis (unresolved) are used as the point spread functions for the images. The radio astronomer's CLEAN algorithm is used to remove the effects of the point spread function from the Alpha Orionis images. The H-alpha image shows non-symmetric envelope structure and its azimuthally averaged radial profile compares favorably with an Alfven wave driven stellar wind model, and evidence for excess emission at (TURN)2 R(,*) is noted.