4. Applications of NNs to Developing Hybrid Earth System Numerical Models for Climate and Weather

In this chapter, numerical modeling background is introduced and a number of neural network (NN) applications developed for numerical weather prediction (NWP) models and climate simulation systems are presented. The hierarchy of numerical models describing weather and climate processes of different scales is introduced and discussed. The notion of hybrid models that combine deterministic physically based parts with statistical blocks is introduced. Several atmospheric and oceanic applications of the NN technique to produce statistical blocks for hybrid numerical models are introduced and discussed in detail. These applications include fast and accurate NN emulations of atmospheric radiation parameterizations and new NN-based convection parameterization for atmospheric models, and fast and accurate NN emulations of nonlinear wave-wave interaction parameterization for ocean wind wave models. The chapter contains an extensive list of references giving extended background and further detail to the interested reader on each examined topic. It can serve as a textbook and an introductory reading for students and beginner and advanced investigators interested in learning how to apply the NN emulation technique to different numerical modeling problems.