ABSTRACT
This chapter explains the weather-type downscaling. One of the major advantages of the weather-type downscaling is to relate physical weather conditions in statistical downscaling by defining weather types. The chapter presents the empirical weather typing by H. H. Lamb and its result is employed to generate daily rainfall sequences following the study of R. L. Wilby. Among other weather types, empirical weather typing requires extensive research and complete review of weather and atmospheric circulations. Future revolution of precipitation is simulated by conditioning the weather type obtained from global climate model (GCM) outputs. P. D. Jones et al. suggested an objective classification with mean sea-level pressure in 16 grid points. The chapter introduces a rather simple empirical approach. Further complex classification methods can be applied such as k-means, machine learning, and Gaussian mixture distributions. Employing the Lamb weather type catalog, transition matrices and precipitation statistics can be extracted.
