Most spatial optimal network designs have been developed for Gaussian processes. However, environmental data rarely conform to this assumption and usually reveal non-Gaussian features suchas asymmetry, so there is a need for novel methods that can account for skewness. To overcome this limitation, this article develops an optimal network design based on the closed skew Gaussian process and introduces new optimality criteria for different aims using information measures. In the Bayesian framework, the design that maximizes the average overall observationalinformation is optimal. The effect of skewness on the configuration of points in the optimaldesign is demonstrated through simulation examples. Besides, the proposed approach is implemented to expand the precipitation monitoring network in Khuzestan province, south of Iran.