Spatial cluster analysis of hydrological drought indicator in an Oasis ecosystem

The objective of this study was to identify the spatial extent of high drought risk areas using remotely sensed drought indices in arid environments. Using remote sensing and geographic information system techniques, this study spatially quantified drought risk using hydrological drought indices derived from satellite data in Ejin Oasis, western China. All indices were computed from the MODIS data in the summer for year 2003. Another group is based on soil moisture, including the Normalized Difference Moisture Index (NDMI), and Standardized Moisture Index (SMI). The use of local Getis statistic (Gi*) provides insights on the spatial relationships of land cover pattern to drought risk assessment. Specially, locations of significant Gi* values identified areas where the differences in hydrological drought indices occur and are spatially clustered. This information may then be used to map high drought risk areas and help governments to improve the allocation of local water resources in arid environments.