2010PASP..122.1510H

Image quality in mosaicked observations from interferometric radio telescopes is strongly dependent on the accuracy with which the antenna primary beam is calibrated. The next generation of radio telescope arrays such as the Allen Telescope Array (ATA) and the Square Kilometer Array (SKA) have key science goals that involve making large mosaicked observations filled with bright point sources. We present a new method for calibrating the shape of the telescope's mean primary beam that uses the multiple redundant observations of these bright sources in the mosaic. The method has an analytical solution for simple Gaussian beam shapes but can also be applied to more complex beam shapes through minimization. One major benefit of this simple, conceptually clean method is that it makes use of the science data for calibration purposes, thus saving telescope time and improving accuracy through simultaneous calibration and observation. We apply the method both to 1.43 GHz data taken during the ATA Twenty Centimeter Survey (ATATS) and to 3.14 GHz data taken during the ATA's Pi Gigahertz Sky Survey (PiGSS). We find that the beam's calculated full width at half-maximum (FWHM) values are consistent with the theoretical values, the values measured by several independent methods, and the values from the simulation we use to demonstrate the effectiveness of our method on data from future telescopes such as the expanded ATA and the SKA. These results are preliminary and can be expanded upon by fitting more complex beam shapes. We also investigate, by way of a simulation, the dependence of the accuracy of the telescope's FWHM on antenna number. We find that the uncertainty returned by our fitting method is inversely proportional to the number of antennas in the array.