عنوان مقاله [English]
Precipitation plays an important role in the global energy and water cycle. It is important for assessment and management of land use, agriculture and hydrology such as flood and drought risk reduction to know about the amount of precipitation reaching the ground water. Assessment of climate change and its effects require long-term rainfall analysis in all spatial scales. Growing concern in the scientific community (Nicholls and Alexander, 2007) is about whether there is a significant change in the amount of precipitation?. About 29% reduction in the daily maximum water flow caused higher temperatures and increased evaporation with any change in rainfall causing drought in southern Canada during the years 1847 to 1996 (Zhang, Harvey, Hogg and Yuzyk, 2001). Recently summer drought caused by unusually dry heating in the areas related to tropical West Pacific and Indian Ocean (Andreadis, Clark, Wood, Hamletand and Lettenmaier, 2005; Pagano and Garen, 2005) including studies of climate change on global precipitation regimes. Germer (2008) has examined monthly variations in rainfall, floods, droughts and runoff in the Yangtze River Basin in China. In another study, Dao (2004) studied the daily variation of rainfall in semi-arid regions of northern China. Raziyy and Azizi (2008) said that topography and latitude are main factors of controlling the precipitation in the west part of Iran. Also, Asakereh (1386) has investigated spatiotemporal variation in Iran precipitation. The results of their study indicated that about 51.4% of rainfall areas in Iran have been changed. In this context, the author aims to explore the development of spatiotemporal changes of the rainfall pattern with access to the available and reliable database and the new approach to extract patterns and possible trends in the data. This paper investigated the spatial patterns of precipitation amount trends in cold period of year, for Iran between 1950 and 2009.
We used reanalysis of monthly data of GPCC with 0.5*0.5 spatial resolution in Iran (From 44E until 63.5E and 25N until 40N) and global & local Moran’s spatial autocorrelation methods. Global Precipitation Climatology Center data accuracy is measured by the geographical weighted regression (GWR) method. Spatial autocorrelation of precipitation data were extracted by global Moran Index. This index shows only the overall clustering of precipitation data. Therefore, to detect the different local patterns spatial autocorrelation, local Moran was used. The index measured spatial differences in rainfall amounts between each grid point and its neighboring points and evaluated its significant level. Trend analysis of the spatial patterns was configured based on Man Kendall’s τ nonparametric test.
Results and Discussion
Geographically weighted regression between the global climatology center data and station data showed that the gridded data have verified to be acceptable to be replaced by the station data. Rainfall Gridded data and station data indicated an average correlation of 76%. Global spatial autocorrelation results and precipitation data in all of the months indicate significant positive spatial autocorrelation (or clustered pattern). Local spatial autocorrelation results of each month show proprietary precipitation pattern and monthly precipitation. They entailed any significant trends. October shows the lowest average index values and dispersion in rainfall patterns. Unlike the months of December rainfall patterns have shown concerted. A spatial clustering map shows, usually in October, a strong spatial cluster is formed on the southern coast of the Caspian Sea. However, during the months of November and December, a strong spatial cluster formed in Zagros on rain again. In February and March the Caspian cluster is reduced in proportion severity.
Temporal variations of the No, high - high points and No, low-low points in all months indicated no significant change. This suggests a lack of space to expand or reduce the size of the cluster rainfall during the study period and comparison between pervasive drought events. Low value of Global Moran’s Index indicates a strong relationship between them. Thus, it is required to be used for other variables for water defect researches.
Comparing of general Moran index values and widespread drought in Iran showed that the low index values are based on the years of drought. The results of this research suggest that the evidence of dehydration in other climatic variables such as temperature and evaporation.