Presentation #353.10 in the session “Computational Augmentation to Observations”.
The surface of the Green Bank Telescope (GBT) is subject to deformations due to heating, wear and tear, and other external factors. These physical deformations do not harm the telescope, and they can be accounted for by using the actuators that are built in between the panels of the main reflector, the active surface. However, efficient observations require the shape of the dish to remain as close to a paraboloid as possible. In order to improve the efficiency of the observations, the surface of the telescope is scanned using the laser antenna surface scanning instrument (LASSI). Many small holes exist in between panels, which makes it difficult to accurately measure the deformations. In order to make accurate measurements of these deformations, a mask is applied to any small holes or unwanted artifacts. Once these holes and/or artifacts are masked, the deformations of the primary reflector may be measured with a higher accuracy. In order to determine how to best implement the mask, the author compared masking at two different steps during data processing using two different strategies; computing a mask based on the noise properties of the scans, and based on the positions of the holes on the surface. The scans were compared against the movement of the active surface to determine which masking strategy was more effective. These residuals were compared with different methods of statistical analysis. Masking using the noise properties of the scans showed the lowest residuals with respect to the active surface, the approach currently used by LASSI to mask holes and spaces between panels.