Research into possible impacts of a climate change requires descriptions of local and regional descriptions of climate. For instance, the local and regional aspect of a climate change is stressed in the U.S. Strategic Plan for the Climate Change Science Program (CCSP) (http://www.climatescience.gov/Library/stratplan2003/default.htm). Global climate models (GCMs) are important tools for studying climate change and making projections for the future. Although GCMs provide realistic representations of large-scale aspects of climate, they generally do not give good descriptions of the local and regional scales. It is nevertheless possible to relate large-scale climatic features to smaller spatial scales. There are two main approaches for deriving information on local or regional scales from the global climate scenarios generated by GCMsc (1) numerical downscaling (also known as dynamical downscaling) involving a nested regional climate model (RCM) or (2) empirical-statistical downscaling employing statistical relationships between the large-scale climatic state and local variations derived from historical data records.