Prediction and Control of the MUSLE Method Results: Geo-Statistical Model

Abstract

Soil erosion and sediment transport by runoff are among the largest environmental problems threatening agricultural lands in the world. Their often irreversible degradation is a major source of socio-economic imbalance in vulnerable areas. The management and control of soil loss in these areas are essentially based on erosion rate modeling generally using two methods: Revised Universal Soil Loss Equation and Modified Universal Soil Loss Equation. The Modified Universal Soil Loss Equation method results often present outliers' values compared to the Revised Universal Soil Loss Equation method. This paper aims to predict and control the MUSLE method values in watersheds based on a database of Revised Universal Soil Loss Equation method values in more than 70 watersheds around the world. For this purpose, Geo-statistical analysis (Simple Linear Regression method) was used to allow the development of a model with a coefficient of determination R2= 0.96 (validity of 96%) and thus the estimation of the tolerable uncertainties of the Modified Universal Soil Loss Equation method by the equation: Y (V Modified Universal Soil Loss Equation LE) = 0.76 X (V_RUSLE) + 1.7.