中文摘要：

针对土壤空间变异性问题,研究了分维数和土壤空间变异性的关系。应用Green-Ampt入渗模型模拟出十种土壤的下渗湿峰深度值,它们的大小依次是：壤质砂土、沙壤土、沙质黏壤土、壤土、粉砂壤土、砂质壤土、黏质壤土、粉砂黏壤土、黏土、粉砂黏土,然后计算出十种土壤随机组合构成的100种土壤剖面的湿峰形状分维数,用这些分维数来描述土壤空间变异性。同时用反距离加权法和相邻两参数平方均值法算出土壤区域属性值,并与分维数进行比较,再通过对土壤空间变异性（用土壤饱和水力传导度来表示）的统计分析和分维数关系的分析,得出湿峰的分形维数与KS具有较明显的分带关系,而且这种关系在实际流域中会更明显。总的来说,土壤空间变异性越大,分形维数df越大。因此可以看出,用分形来评价土壤空间变异性是可行的,并取得了不错的效果。

英文摘要：

In order to solve some problems of soil spatial variability,the relation between the fractal dimension and soil spatial variability is studied.Modeling wetting front depth of ten kinds of soil applying Green-ampt infiltration model,and the sizes of them are followed by loamy sand,sandy loam,sandy clay loam,loam,silt loam,sandy clay loam,clay loam,silty clay loam and silty clay.The wetting front fractal dimension of 100 kinds of soil composition are calculated.Soil spitial variability is described by using these fractal dimensions.At the same time,soil region attribute values can be obtained by the inverse distance power method and the adjacent two-parameter average method.By analyzing the statistical relation of soil spatial variation（soil saturated hydraulic conductivity to that） and the relation between it and fractal dimension,the relation between the fractal dimension of wetting front and Ks is a significant sub-zone,and this sub-zone relation will be more obvious in the real watershed.Overall,the fractal dimension is greater when soil spatial variability is high.Therefore,we can see that,application of multifractality to soil spatial variability is feasible and can achieve good results.