SAMPLING DENSITY FOR CHARACTERIZING THE PHYSICAL QUALITY OF A SOIL UNDER COFFEE CULTIVATION IN SOUTHWESTERN MINAS GERAIS

ABSTRACT The elaboration of maps to characterize the spatial variability of soil attributes assists in the strategic planning and decision making of agricultural managers. Precision and accuracy of maps are related to the ideal sampling density to characterize the variability pattern. This study was conducted with the aim of identifying the sampling density to represent the variability of soil physical quality using attributes with different magnitudes of variation in an area cultivated with coffee. Three approaches were used to find the most adequate sampling density (geostatistical analysis, percentage of error associated with the sampling density, and coefficient of variation). A total of 145 soil samples were collected at a depth of 0-0.20 m at the crossing points of a regular grid with a spacing of 50 m. The percentage of clay, silt, and sand, macroporosity, microporosity, total pore volume, and soil density were determined. The data were submitted to descriptive statistical analysis. For elaborating the variability maps with up to 15% error and soil attributes with a coefficient of variation close to 50%, a sampling density of 3 points ha−1 is suggested.

摘要 为表征土壤属性的空间变异性而绘制的土壤图，可为农业管理者的战略规划与决策提供辅助支撑。土壤图的精度与准确性，与用于表征其变异性模式的理想采样密度密切相关。本研究选取某咖啡种植区作为研究对象，选取变幅各异的土壤属性，旨在确定可表征该区域土壤物理质量变异性的采样密度。本研究采用三种方法筛选最优采样密度：地统计分析、基于采样密度的误差百分比法以及变异系数（coefficient of variation）法。研究共在间距为50m的规则网格交点处，采集了145份0~0.20m深度的土壤样品，并测定了黏粒、粉粒、砂粒的百分含量，大孔隙度、微孔隙度、总孔隙体积及土壤容重。对所得数据开展描述性统计分析。结果表明，若需绘制误差不超过15%的变异性分布图，且目标土壤属性的变异系数接近50%，则建议采用每公顷3个采样点的采样密度。