基于地统计学和GIS技术的北京市大兴区礼贤镇土壤养分空间变异性研究

运用地统计学和GIS方法，对北京市大兴区礼贤镇的表层土壤全氮、全磷、全钾、全硼、碱解氮、有效磷、速效钾、有效硼、有机质和pH值等参数的空间变异及分布特征进行了研究。结果表明，研究区内有效磷、速效钾、碱解氮变异系数分别为10599%、10043%、8029%，分异性较强。经过不同趋势阶数指标插值误差的综合比较，初步确定全磷、有效磷、全钾、速效钾、全硼、有效硼应该选择一阶，碱解氮、有机质和pH值应该选择无趋势，而全氮则应该选择二阶。全磷、有效磷、全钾、全硼、pH值的理论模型为指数模型，速效钾、全氮、碱解氮、有效硼、有机质的理论模型为线性模型。全磷、有效磷、全钾、全硼、pH值主要受到结构性因素的影响，全氮、有效硼、有机质受到结构性因素和随机性因素的共同影响，速效钾、碱解氮主要受到随机性因素的影响。通过普通克里格插值图清晰可见，按照含量分布特点土壤中全氮、碱解氮、全磷、有效硼、有机质、有效磷的空间展布相似，全钾和速效钾的空间展布相似，全硼的空间展布特点与其他指标都不相同，这与统计学分析结果具有相似性。通过了解研究区土壤中10项养分指标分布情况，对及时调整肥料用量比，实现有针对性施肥，调整农业结构以及优势农产品规划和布局提供数据支持。

Spatial Variation Analysis of Soil Nutrients in Lixian Town of Daxing District in Beijing Based on Geostatistics and GIS

Spatial variation and distribution of parameters related to soil nutrients in Lixian, Daxing district in Beijing were investigated by geostatistics and GIS. These parameters include total nitrogen, total phosphorus, total potassium, total boron, alkaline hydrolyzable nitrogen, available phosphorus, available potassium, available boron, organic matter and pH. The results reveal strong variations in available phosphorus, available potassium, and alkaline hydrolyzable nitrogen with variation coefficients of 10599%, 10043%, and 8029%, respectively. The integrative comparison of interpolation error in order index under different trends has been made. It indicates that the spatial variations of total phosphorus, available phosphorus，total potassium, available potassium, total boron, and available boron are characterized by first order trend. While that of total nitrogen shows second order trend and no trend is observed in alkaline hydrolyzable nitrogen, organic matter and pH. The spatial variations of total phosphorus, available phosphorus, total potassium, total boron and pH fit exponential model. While those of available potassium, total nitrogen, alkaline hydrolyzable nitrogen, available boron and organic matter fit linear model. The total phosphorus, available phosphorus, total potassium, total boron and pH are mainly influenced by structural factors. The total nitrogen, available boron and organic matter are controlled by both structural and stochastic factors. While available potassium, alkaline hydrolyzable nitrogen are only influenced by stochastic factors. In the ordinary Kriging interpolation diagram, distribution patterns of element contents in the soil reveal the same spatial distribution for total nitrogen, alkaline hydrolyzable nitrogen, total phosphorus, available boron, organic matter and available phosphorus, as well as for total potassium and available potassium. The total boron has distinct distribution that is different from other elements. This is in agreement with the result of the statistical analysis. Understanding the spatial distribution of the ten nutrient indicators in soil is essential to adjusting the ratio of fertilizer consumption and achieving targeted fertilization. It also provides data support for the structural adjustment and standardized layout of agriculture industry.