基于GIS和改进灰色关联模型的岩溶区土壤肥力评价——以重庆市丰都县岩溶区为例

文章在 GIS 技术支持下，以重庆市丰都县岩溶区为研究对象，根据该县的地形特征、土地利用现状及土壤类型等共采集316个土壤样品，在此基础上，通过主成分分析法、隶属度函数等方法改进的灰色关联模型，对丰都县土壤肥力进行综合评价，并利用地统计学方法揭示其空间变异特征。结果表明：土壤肥力综合指数处于0.352-0.881之间，均值为0.579；其空间变异具有一阶趋势，去除趋势后所剩残差表现出较弱的空间相关性；研究区总体土壤肥力质量中等偏上，其中Ⅱ、Ⅲ、Ⅳ等级的土壤肥力分布面积分别占总区域面积的10.61%、31.09%、35.94%；土壤常规养分和微量元素养分的综合肥力贡献率分别为76.77%和23.23%。这表明将土壤微量元素养分同时纳入土壤肥力评价指标体系是合理可行的，利用改进的灰色关联模型能较好地定量化评价土壤综合肥力，地统计学方法能更好地揭示土壤肥力的空间变异。

Comprehensive evaluation of soil fertility in karst area by GIS and improved Grey Relation Model：A case study in Fengdu county karst area,Chongqing

Soil fertility evaluation can provide the basis for rational fertilization and agricultural production structure adjustment. Under the technical support of GIS, we take Fengdu county karst area as research object and collected 316 soil sample of this county, according to topographical features, land use status and soil types. In this paper, we report organic matter, total nitrogen, available nitrogen, available phosphorus, available potassium, available sulfur, available cuprum, available iron, available manganese, available zinc, available boron content and pH for all soil samples. Based on data, we evaluated the soil fertility of the county by using Grey Relation Model improved by principal component analysis and membership function, and discussed the special distribution characteristics by using statistical methods. The study shows that the soil integrated fertility index (IFI) in the study area ranged from 0.352 to 0.881 with the average value of 0.579 in the study area. Geostatistical trend analysis of IFI presented a one-order (linear) global trend and semi-variance analysis of the residuals showed a weak degree of spatial correlation. Overall fertility quality in the study area is middle or better, the second soil fertility grade region accounted for 10.61%, the third soil fertility grade region accounted for 31.09%, and the fourth soil fertility grade region accounted for 35.94%. The contribution rates for integrated fertility of soil conventional nutrient indexes and microelement indexes were 76.77% and 23.23%, respectively. Therefore, bringing microelement indexes into the soil fertility evaluation index system is reasonable and practicable, the improved Grey Relation Model can quantitatively evaluate integrated soil fertility, and the geostatistical method can satisfactorily reveal spatial variation patterns in soil fertility.