别拉洪河流域湿地变化的多尺度空间自相关分析

Spatial autocorrelation analysis of multi-scale changes in Bielahong River basin wetlands

利用空间自相关模型,结合G IS技术,对别拉洪河流域不同尺度上湿地空间自相关的动态变化进行了研究。结果表明,自20世纪50年代至2000年,别拉洪河流域湿地面积逐渐减少,湿地全局空间自相关呈先增加后减小再增加的"N"字型趋势;别拉洪河流域湿地高-高自相关类型有逐渐减少并向流域下游扩张的趋势,而低-低自相关类型有逐渐增加并向流域上游扩张的趋势;湿地空间自相关表现出明显的尺度效应,随着研究尺度的增大湿地空间自相关性逐渐增强;别拉洪河流域湿地空间自相关的动态变化主要是由人类对湿地的开发造成的。

The multi-scale dynamic change in the Bielahong River basin wetland of the Sanjiang Plain is studied using a spatial autocorrelation model and the geographic information system（GIS） technology.Results show that the wetland area was gradually decreasing and the spatial autocorrelation of the entire wetland area exhibited an up-down-up pattern during the period 1950s-2000.A downward trend is detected in the high-high autocorrelation type,which tends to spread toward the lower reach of the Bielahong River basin.The opposite result is obtained for the low-low autocorrelation type.The wetland spatial autocorrelation exhibits an appreciable scale effect,as the spatial autocorrelation increases with the increase of simulation scales.The result indicates that the dynamic change of wetland spatial autocorrelation is mainly caused by human activities.The spatial autocorrelation analysis of wetland changes can reveal the regularity of the wetland spatial cluster as well as the dynamic changes of the wetland spatial correlation pattern,which could be one of the effective approaches to studying the spatial heterogeneity of wetlands.This study provides a scientific basis for wetland classification and decision-making management.