نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشجوی دکترای آب و هواشناسی، گروه جغرافیای طبیعی، دانشکده جغرافیا، دانشگاه تهران. تهران
2 استاد اقلیم شناسی، گروه جغرافیای طبیعی، دانشکده جغرافیا، دانشگاه تهران. تهران
3 دانشیار اقلیم شناسی، گروه جغرافیای طبیعی، دانشکده جغرافیا، دانشگاه تهران. تهران
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Jet streams are narrow and powerful air currents located in the upper troposphere and lower stratosphere. They form as a result of temperature gradients between polar and tropical air masses and play a key role in atmospheric dynamics, weather systems formation, and the transfer of energy and momentum within the Earth's atmosphere (Barnes and Screen, 2015). In the Northern Hemisphere, the polar and subtropical jet streams are globally significant due to their influence on weather phenomena such as storms, heatwaves, and cold spells (Francis and Vavrus, 2012; Screen and Simmons, 2014). This study identifies daily jet stream patterns from the Atlantic to the Middle East for the first time using hierarchical clustering. The objective is to classify and analyze the spatiotemporal patterns of the jet stream across the North Atlantic to the Middle East and to investigate their climatic trends using NCEP/NCAR reanalysis data and a hierarchical clustering method.
Daily horizontal (u) and meridional (v) wind components at the 250 hPa level were obtained from the NOAA database, covering the 31-year period from January 1, 1985, to December 31, 2015. The study domain extends from 10° to80°N and 80°W to80°E with a spatial resolution of 2.5°. The dataset was processed in MATLAB, converting spatial map data into numerical matrices: 11,322 rows representing daily records and 1,885 columns representing grid points, forming a matrix of over 21 million data points.
Jet streams were defined as wind speeds exceeding 30 m/s. Hierarchical clustering was performed using Euclidean distance for intra-cluster similarity and Ward’s method for inter-cluster linkage. A dendrogram was constructed, and a cut-off threshold was set to obtain nine optimal clusters. The spatial and temporal characteristics of each cluster were analyzed in Excel and GIS to compute variance, frequency, and the 95th percentile of wind speed values. Seasonal and monthly trends were assessed, and inter-cluster correlation matrices were generated using SPSS. For each cluster, a representative day was identified based on maximum internal correlation, and wind fields for those days were visualized using GrADS software.
Autumn patterns (Clusters 1 and 2) and spring patterns (Clusters 5 and 7) demonstrate high variability in jet stream intensity and position. Cluster 1 exhibits a decreasing trend and features a strong polar jet over the North Atlantic alongside a weakened subtropical jet near the Middle East, consistent with the projected weakening of subtropical jets under global warming (Archer & Caldeira, 2008; Overland & Wang, 2010). Cluster 2, characterized by a strong subtropical jet over southern Iran, reflects complex interactions between subtropical and polar jets and Rossby wave propagation, consistent with Hoskins and Ambrizzi (1993).
Significant negative correlations—such as those between Cluster 1 and Clusters 5 and 7, or between Cluster 3 and Cluster 9—suggest opposing atmospheric regimes (Michelangeli et al, 1995; Corti et al. 1999). Observed trends including declines in Clusters 1, 8, and 9 and an increase in Cluster 3—support previous findings regarding the weakening of the polar vortex (Kim et al., 2014). The results are also consistent with the findings of Thompson & Wallace (2000) and Ambaum et al. (2001).
This research addresses a gap in the direct analysis of jet stream patterns by, for the first time, applying spatiotemporal clustering across the North Atlantic to the Middle East. The main innovation of this study lies in the identification of nine distinct and recurrent jet stream patterns, each representing a specific atmospheric circulation regime in the region. These findings enhance our understanding of both seasonal and long-term atmospheric dynamics. The most significant quantitative outcome is the detection of contrasting and statistically meaningful trends in the frequency of these patterns over 31 years (1985-2015). Specifically, the increasing trend of cluster 3 (the winter pattern associated with a strong south-ward shifted polar jet) and the decreasing trends of clusters 1, 8, and 9 (autumn and winter patterns) provide strong evidence of evolving atmospheric circulation regimes in the region. The increased frequency of the pattern associated with extreme weather events (cluster 3) may represent one of the direct consequences of climate change on regional atmospheric dynamics. Furthermore, the discovery of significant negative correlations between certain clusters (e.g., clusters 3 and 9, and clusters 1 with 5 and 7) indicates competition and replacement among atmospheric regimes across different seasons.
کلیدواژه‌ها [English]
)