The use of weather types and air flow indices for GCM downscaling

Abstract

A variety of different methods have been proposed for downscaling large-scale General Circulation Model (GCM) output to the time and space scales required for climate impact studies. Using weather types to achieve this goal provides greater understanding of the problems that are involved compared to the many “black box” techniques that have been proposed. Analyses using Lamb weather types, counts of weather fronts and air flow indices over the British Isles show strong relationships with daily rainfall characteristics such as the probability and amount of rainfall. Three versions of a weather type method are described for generating daily rainfall series. Two methods use an objective scheme to classify daily circulation types over the British Isles along the lines of Lamb's subjective classification. The third method is based on user-defined categories of vorticity. Each method categorises rainfall events according to whether the previous day was wet or dry. All three methods successfully reproduce the monthly means, persistence and interannual variability of two daily rainfall time series during a validation period. A significant advantage of using vorticity is that it is a continuous variable and is strongly related to the probability of rainfall and the magnitude of rainfall events. The paper ends with a discussion of the issues relevant to the application of this method to the development of climate scenarios from GCMs.