| description abstract | A synoptic climatological weather classification scheme incorporating both surface and upper-air data is developed for the central United States based on an automated two-step cluster analysis. It employs daily NCEP?NCAR reanalysis data over all seasons of 57 yr (1948?2004) in creating synoptic types from surface and upper-air (925, 850, 700, and 500 hPa) temperature and humidity data as well as sea level pressure, geopotential heights, and winds aloft. The cluster analysis creates 10 synoptic types exhibiting distinct seasonal preferences, with three each that occur primarily in summer and winter, and four that occur primarily in winter and the transition seasons, particularly spring. The typing scheme generates synoptic patterns largely characterized by distinctive surface circulations, baroclinic vertical structure, and thermal advection. Interannual variations occur in the frequencies of the synoptic types, some of which are out of phase with each other. The annual frequencies of two winter synoptic types, associated respectively with strong zonal and meridional flow, are highly correlated (r ? 0.63?0.73) to the phase of the Pacific?North American teleconnection pattern, while Niño-3.4 equatorial Pacific sea surface temperatures are linked to a synoptic type producing low pressure around the Gulf Coast. | |
