A DETAILED SPATIAL CLUSTER ANALYSIS TO FINDING DISTRIBUTION PATTERNS OF LONGEVITY POPULATION USING DASYMETRIC MAPPING METHOD

摘要

The majority cases of spatial analysis for the distribution of longevity population adopt census administrative boundary data. This approach is based on the assumption that Longevity Population is equally distributed throughout each administrative boundary unit. However, the census administrative boundaries which are artificially compartmentalized cannot reflect realistic characteristic of human residence and sensitive to Modifiable Areal Unit Problem (MAUP). Therefore, realistic and detailed set of spatial units is required when performing analysis for distribution of longevity population. The dasymetric mapping method enables to product more detailed and realistic than census boundary map for distribution of longevity population map utilizing diversity spatial information. Furthermore, cluster analysis enables to grasp the distribution of longevity population which considered spatial association. This study proposed a combined process of dasymetric mapping and cluster analysis to improve the spatial accuracy for distribution of longevity population. The dasymetric mapping method could be provided the boundary revision adopt Land-cover data and residential information with 2007 census data in Gangwon province, Korea. The cluster analysis was performed to find the Hot spot (or Cold spot) of Longevity Population distribution. The dasymetric map confirmed the new boundary which had characteristic of human residential distribution. And more clusters discovered with statistical significant than existing boundary data (such as census administrative sub-level data).