THE UTILITY OF SURFACE AND UPPER AIR DATA IN SYNOPTIC CLIMATOLOGICAL SPECIFICATION OF SURFACE CLIMATIC VARIABLES

摘要: The relative merits of applying 1000 and 500 hPa map classifications to specifying departures in daily monthly climatic elements have been examined for a network 82 stations over New Zealand. classification technique was based on eigenanalysis twice geopotential height fields at 2.5° resolution, followed by cluster analysis. Five significant EOFs were obtained each level subsequent analysis resulted stable patterns seven ten synoptic classes the levels respectively. application EOF combined data from both led case equivalent-barotropic patterns, so that it not possible distinguish between varying vertical structures associated with same low-level circulation. On basis, better able maximum minimum temperatures precipitation, but comparable results types duration bright sunshine wind run. difference skill least partly related number level. Monthly mean all variables 1980-1993 estimated regression frequencies class, weighted departure values month. In general first two methods showed similar skill, estimates slightly superior those analyses, combinations predictors gave best results. Direct proved overall than classifications, limiting value stochastic weather generator approach down-scaling Zealand's regional climate.