Representation of Temperature Variability in Fars Province Using Spatial Statistics

In order to identify the changes of spatial autocorrelation between heat clusters in Fars province, the minimum and maximum temperatures have been recorded to form a network database in Fars province. Then a 33 year period was selected for this research from the afore-mentioned database, the selected daily period was from 1/01/1980 to 12/31/2012, and an area with the dimensions of 18×18 km was added to the region under study. In order to achieve the temperature changes of these heat clusters, the newest spatial statistical methods such as spatial autocorrelation (Global Moran’s I), Anselin Local Moran’s and hot spots by environment GIS were used. The results of this study showed that the pattern of spatial and temporal variations of heat clusters of Fars Province is high-cluster. However, based on local Moran and hot spots, heat clusters in the South, South West and South East have positive spatial autocorrelation pattern (heat clusters) and parts of North and North East have a negative spatial autocorrelation (cold clusters). In the study period, a large part of the province, in most cases, almost half of the total area had no pattern of spatial autocorrelation, significant or non-significant.