吉林省伊通县农田土壤养分空间变异特征

以伊通县耕层土壤有机质、碱解氮、有效磷和速效钾为研究对象。利用地统计与GIS技术相结合的方法。研究4种土壤养分的空间分布特征。结果表明：各变量均服从正态分布，有机质和碱解氮含量的最佳半方差函数理论模型为指数模型，有效磷和速效钾的最佳模型为孔穴效应模型。有机质具有较强的空间相关性，碱解氮具有中度的空间相关性，有效磷和速效钾具有较弱的空间相关性，各变量的空间自相关距离分别为822．4km、977．6km、194．2km和116．4km。有机质和碱解氮含量的空间变异主要受结构性因素（土壤类型、地形地貌和水文条件等）的影响，并且它们的空间分布格局具有相似性；有效磷和速效钾主要受随机性因素（如施肥、土地利用）的影响，其分布表现出较强的空闻异质性。研究结果可为土壤取样间距设计、定位施肥和农田管理提供一定的参考依据。图4，表3，参16。

Spatial Variability of Soil Nutrients in Farmland of Yitong County, Jilin Province

The research aimed at soil nutrients organic matter （OM）, available nitrogen （AN）, available phosphorus （AP） and available potassium （AK） in Yitong County, Jilin Province. The geostatistical analysis technology and geography information system were used to analyze the four soil nutrients. The statistical results showed that data of each variable fitted normal distribution, and an exponential model for semivariogram fitted OM and AN, and a hole effect model fitted AP and AK. Semivariograms indicated that OM con- tent had a strong spatial relevance with the ratio of nugget to sill, and AN content had a moderate spatial dependence, and AP and AK contents had a poor spatial dependence. The ranges of spatially dependent for OM, AN, AP, and AK were 822. 4, 977.6, 194. 2, 116. 4km, respectively. The spatial variability for OM and AN was fundamentally affected by structural factors such as soil types, topography and hydrological conditions and so on, and that for AP and AK was by random factors such as fertilization and land use. The findings could provide reference for designing soil sampling distance, locating fertilization and managing farmland.