耕地土壤有机质空间变异性的随机模拟

Stochastic simulation of cultivated soil organic matter spatial variability

有机质是土壤重要的肥力特征,研究盐渍土改良区耕地土壤有机质空间变异特征可为土壤质量提升提供科学依据。以山东省禹城市盐渍土改良区典型地块耕地土壤有机质为研究对象,在全面野外调查和室内化验分析以获得大量的土壤有机质相关信息的基础上,运用地统计学方法对有机质进行了序贯高斯模拟各次实现(SGSV)、序贯高斯模拟平均实现(SGSA)和ordinary Kriging插值(OK),并将SGSV、SGSA、OK与实测数据进行了统计参数、变异函数、空间分布趋势等方面进行了对比分析。结果表明OK、SGSA改变了有机质数据的空间结构,具有"平滑"效应,SGSA在消除平滑影响方面优于Kriging插值;SGSV具有与实测数据相同的空间自相关结构,对预测点的模拟值具有不确定性,为揭示研究区域土壤有机质的空间结构特征提供了有力的工具,对盐渍土改良区土壤有机质空间不确定性的风险研究具有更实际的意义。

Soil organic matter is one of important features of soil fertility. Studying the spatial variability of the soil organic matter can provide consults to improve the saline soil quality. Based on plentiful information that obtained by field-survey,soil sampling and lab analysis,the study was conducted on an area of typical saline soil improvement districts in Yucheng City. The unobserved values of soil organic matter were estimated by sequential Gaussian Simulation to achieve variance （SGSV）,sequential Gaussian simulation of the average achieved （SGSA） and the ordinary Kriging interpolation （OK） were applied on cultivated soil organic matter. Their statistic characteristics,semi-variances and spatial distribution trend were compared. The results indicated that the estimated values by OK and SGSA changed the original data configuration with obviously smoothing-average effect,and SGSA was better than OK method in eliminating smoothing-average effects. While SGSV had the same data configuration as the measured values,and the simulated value was uncertain for the unobserved points. So SGSV was efficient in analyzing the uncertain and risk variants of soil organic matter,and could bring some negative effect on uncertain and risk variants.