Abstract: 【Objective】 To carry out omprehensive evaluation of soil nutrients in orchard of Yecheng County, Xinjiang and provide reference for regional soil nutrient management and fertilization decision. 【Method】Based on the data of 782 soil samples in Yecheng County, Xinjiang, five nutrient indexes were obtained: organic matter, total nitrogen, hydrolytic nitrogen, available phosphorus and available potassium, three principal components were obtained by SPSS running principal component analysis. Scores of main components and comprehensive scores of soil nutrients were calculated respectively. SPSS average clustering process was used to classify soil nutrient scores. According to the classification results of soil nutrient comprehensive evaluation score, the township level corresponding to the sample points was located. 【Result】The comprehensive scores of soil nutrients were divided into four categories: very high, high, middle and low, respectively. The ratio of sample number was 2.69%, 20.46%, 41.94% and 34.91%, respectively; Ratio of sample number to middle and low class partition reached 76.85%. Significantly higher nutrients are mainly distributed in the township of Sayibago. The high nutrients were mainly distributed in the townships of Sayibago, Tuguqi, Yitikong, and Usakkashi. The middle nutrient contents were mainly distributed in the townships of Sayibague, Gaytimuk, Yitimu, Jiangles Grain Farm and Chaerbagg. Lower nutrients were mainly distributed in Xiahe Township, Yitimukong Township.【Conclusion】The distribution of soil nutrients in Yecheng County was uneven, and the soil nutrient synthesis was of middle and low level. The first principal component was comprehensive reflection of organic matter and total nitrogen; The second principal component reflects soil nutrient supply status of available potassium content; Third principal component is describing effective phosphorus content.

Key words: principal component analysis; cluster analysis; comprehensive evaluation of soil nutrients; orchard

摘要： 【目的】对新疆叶城县果园土壤养分进行综合评价,为区域土壤养分管理和施肥决策提供参考。【方法】基于新疆叶城县果园 782 份土壤样本数据:有机质、全氮、水解性氮、有效磷、速效钾5个养分指标,通过SPSS运行主成分分析得出3个主成分,计算各个样点各主成分的得分以及土壤养分综合得分,SPSS运行K-均值聚类过程将土壤养分综合得分分类,依据土壤养分综合评价得分分类结果查找样点对应的乡级分布。【结果】土壤养分综合得分划分极高、高、中、低四类得分区,样本数所占比率分别为2.69%、20.46%、41.94%、34.91%,中及低类得分区样本数比率达76.85%;极高养分主要分布在萨依巴格乡;高养分主要分布在萨依巴格乡、吐古其乡、依提木孔乡和乌夏克巴什乡;中等养分含量主要分布在萨依巴格乡、加依提勒克乡、依提木孔乡、江格勒斯粮种场和恰尔巴格乡;低养分主要分布在夏合甫乡、依提木孔乡、乌吉热克乡、加依提勒克乡。【结论】叶城县土壤养分分布不均,土壤养分综合属于中低水平。第一主成分是全氮和有机质的综合反映;第二主成分反映速效钾含量对土壤养分供给状况;第三主成分是有效磷含量的描述。

关键词: 主成分分析, 聚类分析, 土壤养分综合评价, 果园