PRINCIPAL COMPONENT AND CLUSTERING APPROACH TO DELINEATION OF SOIL MANAGEMENT UNIT IN PELLA DISTRICT OF HONG LOCAL GOVERNMENT AREA, ADAMAWA STATE
Publication Type: Journal Article.
Publication Year: 2026
Author(s): Maunde, M. M., Alhassan, I., Saddiq, A. M., Tahir, A. M., Ibrahim, A., and Onokebhagbe, V.O.
Journal Name: (IJAEMD)
ABSTRACT
Principal Component Analysis (PCA) was applied to 29 soil physicochemical parameters to identify the most influential indicators of variability in soils of the Pella district. The first seven principal components (PC1–PC7) explained 96% of the total variance, with strong contributions from Cu, K, Porosity, Clay, Effective Cation Exchange Capacity (ECEC), Fe, and Exchangeable Sodium Percentage (ESP). These variables emerged as key drivers of soil heterogeneity, thereby reducing the set of parameters required for effective soil characterization. The component structure highlighted Cu, ECEC, Fe, AvP, and ESP as dominant contributors across multiple components, underscoring their relevance in soil fertility assessment. Cluster analysis further classified soils from eleven sampling units (Pella, Dzuma, Midila, Zhedinye, Mbulinyi, Dagza, Uding, WuroBokki, Fachi, PellaGwaja, and Holma) into three distinct management zones. Dzuma, Pella, Uding, PellaGwaja, and Holma formed one similarity group; Zhedinye and Dagza clustered together, while WuroBokki, Fachi, Mbulinyi, and Midila constituted a third group. This classification demonstrates that PCA combined with clustering provides a robust framework for delineating site-specific soil management zones, enabling precision agriculture strategies that target variability in soil fertility. By identifying critical soil properties and grouping soils into management units, this study offers a practical basis for improving crop productivity and promoting sustainable land management in the Pella district.
Keywords: Principal Component Analysis, clustering, nutrients, Hong, Adamawa State