Abstract: In order to discuss the impact of Picea crassifolia forest on the soil pH and nutriments,the experimental places were selected in a dynamical monitoring sample plot(340 m×300 m)of Picea crassifolia forest in Dayekou Basin of the Qilian Mountains. The soil sampling was collected based on grid(20 m×20 m)method in August 2012. By using the method of classical statistics coupled with geostatistical techniques such as semivariogram, kriging interpolation,and spatial distribution maps,the spatial heterogeneity of soil pH and nutrients were studied in this paper. The experimental results showed as follows:Soil pH,hydrolyze nitrogen and total phosphorus showed small variation; soil organic carbon,total nitrogen,available phosphorus,total potassium and available potassium showed moderate variation;and the variation intensity of soil pH and nutrients showed:available potassium(29.68%) >organic carbon(19.81%) > available phosphorus(19.37%) > total nitrogen(18.08%) > total potassium(15.58%) > hydrolyze nitrogen(8.18%) > total phosphorus(6.77%) > pH(2.52%). Parameters of theoretical variogram models showed that soil pH,total nitrogen,hydrolyze nitrogen,total phosphorus,available phosphorus and available potassium all fitted spherical model,while organic carbon and total potassium can be best described by exponential model;variation of soil pH,organic carbon,total nitrogen,hydrolyze nitrogen,total phosphorus,available phosphorus,total potassium and available potassium are 108.8 m,88.5 m,112.8 m,131.9 m,143.3 m,73.3 m,73.3 m and 134.71 m,respectively. In terms of spatial structure,soil pH showed moderate spatial autocorrelation,this was resulted from both soil structural factors and random factors,and soil nutrients showed high spatial autocorrelation,which was greatly affected by structural factors. Soil pH and nutrients presented patch distribution. Distribution of pH showed that soil is mainly alkaline; organic carbon and nitrogen has similar spatial distribution pattern; soil total phosphorus and available potassium distribution changes are more obvious; changes of available phosphorus and total potassium contents are relatively moderate. The study results can provide not only references for sampling design and spatial distribution of soil pH and nutrients of Picea crassifolia forest in Qilian Mountains,but also a scientific basis for soil environment recovery and reconstruction of Picea crassifolia forest.

Key words: Picea crassifolia forest, soil pH, soil nutrient, spatial heterogeneity, geostatistical, Qilian Mountains