Numerous works have been completed in the atmospheric science modeling community based on changing or replacing of parameterization schemes with other schemes. However, few studies have considered how to make these models less time- consuming. Most of the changes that are made to models involve increasing their degree of complexity (and time burden) as opposed to their simplicity. However, a recent study showed that the processing time of the CAM2 climate model was sped up 50-80 times by replacing its longwave parameterization scheme with a neural network. In this book, we show that the entire shortwave parameterization scheme can be collapsed into a function of three independent variables: latitude, longitude, and day of year. Three separate empirical methods are then used to approximate the scheme currently in use in the model. The analysis provides further evidence that suggests that subroutines of model formulas can be replaced with empirically-based methods that drastically speed up processing time without a significant loss of output accuracy. As a result, this book is a must- have for climate academicians and statisticians alike.