Assessing simulations of daily temperature and precipitation variability with global climate models for present and enhanced greenhouse climates

The enhanced greenhouse climates of five different global climate models are examined with reference to the ability of the models to characterize the frequency of extreme events on both a regional and global scale. Ten years of model output for both control and enhanced greenhouse conditions are utilized to derive return periods for extreme temperature and precipitation events and to characterize the variability of the model climate at both regional and global scales. Under enhanced greenhouse conditions, return periods for extreme precipitation events are shorter and there is a general increase in the intensity of precipitation and number of wet spells in most areas. There is a decrease in frequency of cold temperature extremes and an increase in hot extremes in many areas. The results show a reasonable level of agreement between the models in terms of global scale variability, but the difference between model simulations of precipitation on a regional scale suggests that model derived estimates of variability changes must be carefully justified. Copyright (C) 1999 Royal Meteorological Society.