微生物复垦技术近年来已成为矿区土地复垦新的研究方向，微生物复垦多年后，新的人工生态系统已建立，其灌木林下土壤养分的空间异质性研究成为关注的重点。选择神东大柳塔采煤沉陷地微生物复垦地为研究区，利用经典统计学、地统计学和协同克里金插值方法，研究采煤沉陷地微生物复垦区灌木林下土壤全氮、有机质和速效磷的空间异质性及其影响因素。结果表明:研究区土壤全氮、有机质和速效磷的含量均值分别为0.50 g/kg,11.93 g/kg,2.24 mg/kg，均呈现出中等强度变异，存在空间自相关性。选择能表征丛枝菌根真菌特性的关键因子土壤菌丝密度，发现土壤菌丝密度与土壤全氮、有机质和速效磷含量呈极显著正相关性。以土壤菌丝密度为协变量的土壤全氮、有机质和速效磷的最优模型分别为高斯模型、球状模型和高斯模型，决定系数分别为0.709,0.757和0.467，均高于单一变量的土壤全氮、有机质和速效磷的决定系数。以菌丝密度为辅助变量的土壤全氮、有机质和速效磷均呈现强空间自相关性，有机质的变程较长，空间连续性较好;土壤全氮空间分布呈块状分布，对空间依赖性较强;土壤速效磷呈不规则斑块状，破碎化程度高。以菌丝密度为辅助变量进行协同克里格插值能够显著提高土壤全氮、有机质和速效磷含量的预测精度。

Micro-reclamation technology had become a new research direction of land reclamation in coal mining areas recently. The man-made ecosystem had been established after micro-reclamation for several years,the spatial heterogeneity of soil nutrients under shrubbery in the micro-reclamation demonstration base had been paid close attention to the further research. The microbial reclamation area in the coal mining subsidence area of Daliuta,Shendong,was selected as the research area. The total nitrogen,organic matter,and available phosphorus in shrub forests at microbial reclamation area in the coal mining subsidence area were studied using classical statistics,geostatistics,and coordinated kriging interpolation methods to obtain the spatial heterogeneity and its influencing factors. The results showed that the average values of soil total nitrogen,organic matter,and available phosphorus content in the study area were 0. 50 g / kg, 11. 93 g / kg and 2. 24 mg / kg respectively,all of which exhibited a moderate-intensity variation and spatial autocorrelation. The soil mycorrhizal hyphal density had been chosen as the key factors of the arbuscular mycorrhizal fungi characteristics. The results showed that the hyphal density has very significant positive correlations with soil total nitrogen,organic matter and available phosphorus. The best fitting models for the cooperative variation functions of soil total nitrogen,organic matter,and available phosphorus with hyphal density as an auxiliary variable are gaussian model,spherical model and gaussian model,and their coefficients of determination are 0. 709,0. 757 and 0. 467,which are higher than those of single-variable soil total nitrogen,organic matter,and available phosphorus. The soil total nitrogen,organic matter and available phosphorus all showed strong spatial autocorrelation with hyphal density as auxiliary variable,which can significantly improve the prediction accuracy of soil total nitrogen,organic matter and available phosphorus content. The spatial distribution of soil total nitrogen was block-shaped,and had a strong dependence on space. Soil available phosphorus was irregularly patchy and highly fragmented. Cooperative Kriging interpolation with mycelium density as an auxiliary variable can significantly improve the prediction accuracy of soil total nitrogen organic matter and available phosphorus. There are different spatial heterogeneity and spatial correlations of soil total nitrogen and organic matter in shrub forests at the microbial reclamation zone of coal mining subsidence area.