Neural network implementation of a mesoscale meteorological model

Abstract

Numerical weather prediction is a computationally expensive task that requires not only the numerical solution to a complex set of non-linear partial differential equations, but also the creation of a parameterization scheme to estimate sub-grid scale phenomenon. This paper outlines an alternative approach to developing a mesoscale meteorological model - a modified recurrent neural network that learns to simulate the solution to these equations. Along with an appropriate time integration scheme and learning algorithm, this method can be used to create multi-day forecasts for a large region. The learning method presented in this paper is an extended form of Backpropagation Through Time for a recurrent network with outputs that feed back through as inputs only after undergoing a fixed transformation. © 2014 Springer International Publishing.