ASSESSMENT OF SOIL ERODIBILITY ESTIMATORS ON SELECTED SOIL TYPES IN SOUTHERN NIGERIA

Abstract

Soil erosion is a major threat to food security and agricultural development at local, national, and international levels. Various soil erosion models have been developed globally to address erosion issues and safeguard soil resources. Studies in southern Nigeria have demonstrated the use of these models and GIS for estimating soil erodibility and mapping erosion hazards. Empirical models have also enhanced soil erodibility assessments, aiding sustainable land management in the region. Soil erodibility, represented by the K-factor, is a crucial parameter for assessing soil vulnerability to water erosion and predicting erosion rates. Therefore, this study used a literature survey to compute soil erodibility data from 24 natural runoff plots in southern Nigeria from 1975 to 2023 to evaluate the accuracy of K-factor estimates by the Soil Erodibility Nomograph (Nomo), Erosion Productivity Impact Calculator (EPIC), and Geometric Mean Particle Diameter (Dg) models on Alfisols and Ultisols. The models' performance was assessed using metrics such as Nash-Sutcliffe Efficiency (NSE), percentage bias, coefficient of determination (R²), and the p-value from the Mann-Whitney U test. Results showed that field-measured values ranged from 0.007 to 0.108 Mg h MJ^-1 mm^-1. The nomograph overestimated the K-factor by 14.1%, while EPIC and Dg underestimated it by 80.1% and 82.8% respectively. The nomograph model performed best for Ultisols (R² = 0.75, NSE = 0.56) and to a lesser extent for Alfisols (R² = 0.45, NSE = 0.40). To develop new empirical equations for predicting soil erodibility from soil data in southern Nigeria, where natural runoff plots are unavailable, a significant non-linear regression analysis is required between field-measured and predicted K-factors.