Fuzzy logic for evaluation of the fertility of soil and productivity of conilon coffee

Abstract The objective of this study was to analyze, using the geoestatistic and a system of classification fuzzy, the fertility of an experimental area with base in chemical attributes of the soil and its relationship with the productivity of the conilon coffee. The study was accomplished in the experimental farm of the INCAPER - ES. The soil samples were collected in the depth of 0 - 0.2 m, being analyzed the attributes: matches, potassium, calcium and magnesium, aluminum, sum of bases, cation exchange capacity (pH 7), and saturation percentage. The data were submitted to a descriptive, exploratory, and geostatistical analysis. A system of fuzzy classification was applied using the attributes described to infer about the fertility of the soil and its relationship with the productivity of the culture. The fertility possibility presented positive spatial relationship with the productivity of the culture, with higher values of this where the possibility of fertile soil is superior.摘要：本研究旨在借助地统计学（geostatistics）与模糊分类（fuzzy classification）系统，基于土壤化学属性分析某试验田的肥力状况，并探究其与科尼隆咖啡（conilon coffee）产量的关联。本研究于圣埃斯皮里图州农业研究与推广机构（INCAPER-ES）的试验农场开展。研究采集了0~0.2 m深度的土壤样品，分析的土壤化学属性包括：钠、钾、钙、镁、铝、碱总量、pH7条件下的阳离子交换量及饱和度百分比。对获取的数据开展了描述性、探索性及地统计学分析。基于前述土壤化学属性构建模糊分类系统，以此推断土壤肥力状况及其与作物产量的关联。研究结果表明，土壤肥力潜力与作物产量呈显著正空间相关性，在土壤肥力潜力更高的区域，作物产量也相应更高。