Effective Use of Regional Models for Climate Prediction

Regional models are our best tool for providing the local-scale information about climate change needed for impacts analysis and planning as the climate warms in response to greenhouse gas increases. Lessons learned from many years of using regional atmospheric modeling to simulate present, past and future climate are reviewed. The advantages and disadvantages of various options for simulation design are discussed, including an analysis of techniques for initializing regional model simulations and constraining the lateral boundaries. We show when and how regional models can be used to provide reliable predictions, and review ensemble simulation design and other approaches to evaluating confidence in predictions. Several examples are provided using predictions of 21st c. climate change over Africa and the U.S.