In assessing the risks associated with climate change, downscaling has proven useful in linking surface changes, at scales relevant to decision making, to large-scale atmospheric circulation derived from GCM output. Stochastic downscaling is related to synoptic climatology, weather-typing approaches (classifying circulation patterns) such as the Lamb Weather Types developed for the United Kingdom (UK), the European Grosswetterlagen (Bardossy and Plate, 1992) and the Perfect Prognosis (Perfect Frog) method from numerical weather prediction. The large-scale atmospheric circulation is linked with site-specific observations of atmospheric variables, such as precipitation, wind speed or temperature, within a specified region. Classifying each day by circulation patterns is achieved by clustering algorithms, fuzzy rule bases, neural nets or decision trees. The linkages are extended to GCM output to account for climate change. Stochastic models are developed from the probability distributions for extreme events. Objective analysis can be used to interpolate values of these models to other locations. The concepts and some applications are reviewed to provide a basis for extending the downscaling approach to assessing the integrated risk of the six air issues: climate change, UV-B radiation, acid rain, transport of hazardous air pollutants, smog and suspended particulates.
