부산대학교 도서관

Given the basic features of drought, it is difficult to monitor its spatiotemporal development. Though in recent decades drought monitoring has grown exponentially, its spatiotemporal dynamics are still a serious challenge. In this study, using the S-TRACK method for spatial drought tracking, which is based on machine learning to build drought paths, the dynamics of drought in Iran has been investigated. To analyze the drought, the Global Precipitation Climatology Centre (GPCC) networked data were used in the statistical period of 1952-2019. Using binary method, the spatial characteristics of drought were extracted and four of the largest droughts in Iran were analyzed. The results show that in all four droughts, the maximum area of the largest clusters is in the second half of the year and the first half of the following year, and it peaks in the warm period of the year. From a spatial perspective, the droughts of 2000 and 2008 were the most extensive droughts in Iran. Also, in terms of time sequence, the 1984 and 2000 droughts were the most continuous droughts in Iran. It was also found that with an increase in surface area, the frequency of long distances between these areas decreases, while the frequency of short distances increases. The highest density of pervasive droughts was observed in western and eastern Iran. In all four statistical periods, there is a dominant path, which includes southeast to north and northwest, as well as northwest to south and southeast.
Author(s): Alireza Hosseini [sup.1], Yousef Ghavidel [sup.1], Manuchehr Farajzadeh [sup.1] Author Affiliations: (1) grid.412266.5, 0000 0001 1781 3962, Department of Physical Geography, Tarbiat Modares University, , Tehran, Iran Introduction Although [...]