Analysis of dynamical statistical characteristics of temperature correlation networks of 1-30 d scales

作者：Gong Zhi Qiang^*; Zhou Lei; Zhi Rong; Feng Guo Lin

来源：Acta Physica Sinica, 2008, 57(8): 5351-5360.

DOI：10.7498/aps.57.5351

• multi-scale
• small world network
• global coupled network
• correlation condition value

Based on reanalysis data of National Center of Environment Prediction of America, temperature correlation networks of different time scales are constructed and their dynamical statistical properties are also analysed. Results show that temperature correlation networks of 1 and 30 d belong to small world and semi-globally coupled network, respectively. The time scale condition is 11-12 d for networks to shift from one type to the other. For 1 d scale, the total number of links at each geographic location is semi-symmetrical between north and south hemisphere, the equatorial nodes have a larger number of connections in three regions: the equatorial Indian ocean, equatorial east Pacific ocean and equatorial east Atlantic ocean. Sub-system networks of both mid-latitude and low-latitude belong to the small world network, while the accumulative degree distribution of low-latitude follows the exponentical and Gaussian distributions and that of mid-latitude follows the semi-Gaussian distribution. The range of correlation condition value for temperature network belonging to small world network is 0.2-0.9. Meanwhile for the semi-globally coupled and the small world networks, respectively. The range of correlation condition value for temperature network belonging to semi-globally coupled network is 0-0.6.

• 出版日期2008-8
• 单位兰州大学; 国家气候中心; 扬州大学; 中国科学院大气物理研究所