水土流失景观空间自相关与自相似的尺度特征分析——以长汀县根溪河小流域为例

Scaling analyses of spatial autocorrelation and self-similarity of soil and water loss landscape pattern——A case study of the small watershed of Genxi River, Changting County

基于遥感、地理信息系统和景观生态学研究方法,对长汀县根溪河小流域水土流失景观格局空间自相关与分形结构及其在不同粒度下的变化特征定量分析表明,水土流失景观中的强度以上流失类型空间自相关最低而自相似性最高;空间自相关性和自相似性依赖于粒度的选择,Moran's I与各种景观类型的FRAC_AM均存在尺度效应;空间自相关随粒度增加呈下降趋势;不同景观类型的自相似性随尺度的变化没有统一规律,但多数第一个较明显的转折点出现在粒度大小为10m处。

Soll erosion landscape pattern characters with different grains have been analyzed in the small watershed of Gcnxi River in Changting County, based on the means of the techniques of RS, GIS and landscape ecology. The results showed that Moran＇s I of the most serious loss landscape is minimum, whereas it＇s index of FRAC_AM is maximum among the 5 types of soll erosion landscape in the small watershed of Genxi River. The landscape indices of Moran＇s I and FRAC_AM were all related to the grains. The spatial autocorrelation and self-similarity of the soil erosion landscape pattern, depended on different scales within certain range of scale with the depending degree differed greatly, and have different sensitive points to the scaling. With the grain increased, the spatial autocorrelation of the total spatial landscape reduced gradually, while Moran＇s I had no obvious sensitive point. The spatial self-similarity did not follow the scaling regularly, however, most had the first prominent sensitive point to the scaling at the level of 10m among 5 different types of landscape patches.