Spatial Variability of Soil Fertility Using a Geostatistical Approach in Benin: A Case of Hlankpa Village, Adjohoun Commune

Abstract

In Benin, agricultural soils are degraded by poor farming practices and flooding, which reduces their fertility. This study aimed to analyze the variability of physico-chemical properties on a 6-hectare farm in Adjohoun municipality to guide soil management and crop planning. A total of 93 soil samples were collected at 25-meter intervals. The following parameters were determined: pH, nitrogen (N), phosphorus (P), potassium (K), organic matter, cation exchange capacity (CEC), soil depth, and water table level. Four kriging methods (simple, ordinary, universal, indicator) combined with different variogram models (exponential, Gaussian, circular, spherical) were tested to identify the most suitable approach for spatial variability assessment. The performance of the variogram models and the kriging methods was evaluated and compared using the Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE) as validation criteria. The exponential variogram proved to be the most appropriate for pH (MAE = 0.003; RMSE = 0.979), soil depth (MAE = 0.019; RMSE = 0.887), and water table level (MAE = 0.001; RMSE = 1.033), while the Gaussian model better explained the variability of nitrogen, phosphorus, and cation exchange capacity. Simple kriging showed good performance for soil depth (MAE = −0.003; RMSE = 0.979), pH (MAE = −0.003; RMSE = 0.979), nitrogen (MAE = −0.006; RMSE = 1.078), and organic matter (MAE = 0.00003; RMSE = 1.025), while indicator kriging excelled for water table level and phosphorus. The soils overall have satisfactory chemical fertility (N ≈ 0.29%; P ≈ 19.3 mg/kg; K ≈ 0.40%; organic matter ≈ 2.49%; CEC ≈ 15.9 cmol/kg) although the pH is slightly alkaline (≈7.51). The findings indicate that crops adapted to hydromorphic conditions are more suitable for the site. The geostatistical approach proved effective for precise mapping and soil fertility management, providing valuable insights for agricultural planning in Benin.