基于主成分分析的土壤肥力综合指数评价

Assessment of integrated soil fertility index based on principal components analysis

以杨凌为研究区域,选择田块尺度为评价单元,采集27个田块土壤,利用模糊数学中综合指数评价模型进行土壤肥力质量评价。经取样数合理性统计分析,27个样点估计研究区域土壤肥力评价能够满足90%置信水平条件下15%和20%的相对误差取样数的精度需求。运用主成分分析,通过计算变量的Norm值筛选速效钾、碳酸钙、全磷、全氮、土壤有机质、CEC、速效磷、全钾进入土壤肥力评价的最小数据库集（MDS）,利用非线性隶属度函数实现评价指标量纲归一化且评价单项肥力指标,利用雷达图直观地反映单因素肥力指标在土壤中的状态以及土壤肥力的整体状况,结果表明,全磷和碳酸钙是影响研究区土壤肥力的限制性因子。研究区域在田块评价尺度上的土壤肥力综合指数范围为0.7-0.8。

Taking the Yangling of Shaanxi as a case and selecting field plot scale as the assessment unit, the farmland soil fertility quality was evaluated by using integrated fertility index. According to the statistical analysis for sample number rationality, the regional soil fertility assessment based on 27 sampling sites were sufficient to meet the required accuracy of 15% and 20% relative error under 90% confidence level. The minimum data set components including soil organic matter, total nitrogen, total phosphorus, total potassium, available phosphorus, available potassium, CEC, and calcium carbonate were selected through calculating the Norm values based on principal components analysis, and the non-linear membership function was used to transform the indicators into dimensionless data and to evaluate single indicator. The radar plot was used to reflect the single factor fertility and the integral fertility. It was shown that total phosphorus and calcium carbonate were the limiting factors for the soil fertility quality in the study area, and the integrated fertility index at field plot scale was 0. 7-0. 8.