The delivered image quality of the Gemini telescopes is an important metric for queue planning and scheduling, especially for adaptive optics instruments whose performance depends critically on seeing.  The commonly used statistics of seeing and other IQ factors, however, are based on site survey data that are more than a decade old.

We have surveyed Gemini South image quality and environmental data taken over the past several years in order to characterize the actual performance during operations.  These data are drawn from the Gemini Engineering Archive (GEA), which contains seeing monitor and telescope wavefront sensor image quality data.   We have developed an analysis package in Python using the technique of Principal Component Analysis (PCA), in order to identify the most important correlations in the data.