【摘要】 以新疆绿洲滴灌棉田为研究对象,基于100个耕层土壤(0～30cm)有机质、碱解氮、速效磷和速效钾含量的分析数据,应用模糊K-均值算法进行了连续分类。研究结果表明:研究区最佳分区数为4个,混乱度指数平均值为0.23,地理空间上土壤的隶属关系相对明确;各管理分区土壤养分的变异系数都较分区前全研究区有所减小,分区间土壤养分差异显著。因此,基于模糊集理论的土壤养分预测方法可获得较好的分区结果,可用于实施变量投入和精确施肥推荐,为棉田土壤养分管理提供科学的理论依据。更多还原

【Abstract】 The objective of this research was to define management zones of oasis cotton field in Xinjiang Province.The variables of organic matter,available N,available P and available K data determined in 100 top soil(0～30 cm) samples were selected as data sources.The results indicated that the whole field was divided into four optimum management zones,the average confusion index was 0.23 in all management zones and the spatial distribution of membership grades was unambiguous.It showed that variation coefficient of soil nutrients decreased in every management zone than total field and differed significantly.It is recommended that the method of spatial prediction based on the fuzzy set theory could achieve a zoning result,it can be used for fertilizer recommendation and scientific direction to manage soil nutrient.更多还原