شناسایی الگوهای همدیدی بارش های شدید شمال شرق ایران

North East of Iran sometime is influenced by heavy precipitation (HP) that causes plentiful damages. This paper aim to classifying and identifying the atmospheric circulation patterns related to heavy precipitation. This work increases my understanding from dynamic mechanisms that is generated by heavy precipitation. In this research, a heavy precipitation day has been considered if at least in three stations HP is more than 15 mm. Therefore, we have selected 186 days from October to May for a 37-year period (1970 – 2006). As well as, NCEP-NCAR Reanalysis data in the spatial resolution of 2.5 × 2.5 degrees including daily average of 500 hpa geopotential height, Sea level pressure and wind components (U, V) at the 500 hpa and 850 hpa level were used over the region 20 – 60 N and 20˚W-70˚E. For classifying the data of geopotential height and obtaining their patterns, we have used subjective and quantitative methods. At the first, we used Principle Component Analysis (PCA) method on S-mode analysis. Then by using scree plot, 9 principal components (PC) - that is covered about 91% of total of the variance- were selected. Then the non-hierarchical K-means method was used to cluster of the data that were obtained from PCA. Also we have analyzed 6-hour maps of each day by using subjective method from 72-hours ago. As a result, this paper showed that there are three main patterns for heavy precipitations including: Sudanese Low-pressures, Sudanese -Mediterranean composite low-pressures and Mediterranean low-pressures. Among these patterns, the heaviest of precipitations is related to Soudan low pressure.