基于主成分和聚类分析的村域稻田土壤肥力评价  

Evaluation on Rice Soil Fertility Based on Principal Component Analysis and Cluster Analysis at Village-scale

作　　者：王小艳[1] 冯跃华[1] 李云[1] 武彪[1] 陈山[1] 王旭[1] 李香玲[1] 莫银化 

机构地区：[1]贵州大学农学院,贵阳550025

出　　处：《中国农学通报》2014年第33期46-50,共5页Chinese Agricultural Science Bulletin

基　　金：国家自然科学基金项目"黔中喀斯特山区村级尺度土壤养分;水稻生长特性和产量的空间变异及其协同机理研究"(31160263);贵州省农业科技攻关项目"机插水稻高产栽培关键技术研究与示范"[黔科合NY(2011)3085];贵州省科学技术基金项目"水氮互作对杂交水稻碳氮代谢和产量及品质的影响机理"[黔科合(2011)2139]

摘　　要：为了解贵州省安顺市西秀区文星村稻田土壤肥力状况,采用主成分与聚类分析相结合的方法,对研究区土壤肥力进行了评价。结果表明：研究区土壤肥力因子除p H、总孔隙度、毛管孔隙度变异系数小于10%,为弱变异外,其他变异系数在10%~100%之间,均为中等变异,有效养分硫、铁、锰等元素的变异相对较大,变异系数均大于75%。主成分分析表明,可归为6个主要成分的22个评价因子,其累计贡献率达74.26%,可以从一定程度上反应田区土壤的肥力标准概况。土壤综合肥力指标（IFI）值的得分为-1.7422~1.8230,聚类分析将其分为Ⅰ、Ⅱ、Ⅲ、Ⅳ4个等级,所占样点比例分别为2.5%、52.5%、30%和15%。该方法能相对正确的对土壤养分供应能力及丰缺度作出评价,为稻田土壤的合理施肥提供参考依据。In order to clarity the soil fertility status of paddy fields in Wenxing Village, Anshun City, GuizhouProvince, principal component analysis combined with the cluster analysis was used to analyze the soil fertility.The results showed that the CV of soil p H, total porosity and capillary were lower than 10%, indicating theirweak variation. In contrast, other variation coefficients were from 10% to 100%, belonging to moderatevariation. The variability of available S, available Fe and available Mn were intensive with variation coefficientof more than 75%. Principal component analysis showed that 22 evaluation factors could be classified into sixmajor components, its cumulative contribution rate had reached 74.26%, could reflect the soil fertility standardat a certain extent. The IFI values of soil samples were from-1.7422 to 1.8230, which had been divided intofour levels as Ⅰ, Ⅱ, Ⅲ and Ⅳ by the cluster analysis with their proportion of 2.5%, 52.5%, 30% and 15%,respectively. This method could make a relatively correct evaluation on soil nutrient supply capacity and thedegree of abundance, and provide a reference for reasonable application of fertilizer in paddy field.

关 键 词：土壤肥力 主成分分析 聚类分析 评价 

分 类 号：S5[农业科学—作物学]