Characterizing crosstalk within the Pan-STARRS GPC1 camera
Using data from a year-long dedicated campaign to observe bright stars, we study the crosstalk channels present in the GPC1 camera. By analyzing these data, we construct a dataset that checks source stars on almost every CCD of every chip within the camera against all possible crosstalk destinations. We use a clustering algorithm to find potential crosstalk occurrences, and then also check all possible combinations (driven by the hardware layout) by eye. This results in a total of 640 rules, with a flux attenuation factor ranging from 2.5x10$^{2}$ for the bright end to 2.5$\times$10$^{4}$ at the faint end. The average value of m$_{cross}$-m$_{src}\approx$-10.25 corresponds to an attenuating factor of 1.25x10$^{4}$, which produces crosstalk ghosts with an average signal-to-noise ratio of 0.64$\pm$0.1 on the bright images. We find no evidence of crosstalk signals between CCDs not connected in the hardware setup. The distribution of attenuation factors is also found to be dependent on crosstalk movement. A clear dependence on cell column offsets is found, consistent with the idea that the source star charge is progressively attenuated during the traversal of cell readout lines. While we can see the trends, the uncertainties on aperture magnitude measurements are large at this stage.