Title: Spatio-Temporal Stochastic Pattern Generator for ensemble prediction and ensemble data assimilation in geophysical applications

Abstract: A generator of spatio-temporal pseudo-random Gaussian fields that satisfy the "proportionality of scales" property (Tsyroulnikov 2001) is presented. The generator is a third-order in time stochastic differential equation with a pseudo-differential spatial operator defined on a limited area domain in Cartesian coordinate system. The spatial covariance functions of the generated fields belong to the Mat\'ern class. The spatio-temporal covariances are non-separable. A spectral-space numerical solver is implemented and accelerated exploiting properties of real-world geophysical fields, in particular, smoothness of their spatial spectra. The generator is designed to simulate additive or multiplicative, or other spatio-temporal perturbations that represent uncertainties in numerical prediction models in geophysics. The program code of the generator is publicly available.