A Mechanistic Approach Accounting for Progressive Soil Surface Sealing in Interrill Erosion Modeling

Abstract

Raindrop‐induced soil surface sealing alters infiltration, runoff, and sediment detachment, yet most erosion models fail to capture its dynamic impacts on interrill processes at event scales. Conceptually informed by the process‐based approach of Water Erosion Prediction Project, we develop a modeling framework that couples seal formation, two‐layer infiltration, kinematic wave overland flow, and interrill erosion. The framework integrates slope gradient and rainfall intensity into the temporal evolution of interrill erodibility, enabling explicit representation of sealing dynamics. Laboratory data of Assouline and Ben‐Hur (2006, https://doi.org/10.1016/j.catena.2006.02.005 ) covering rainfall intensities of 24–60 mm hr −1 and slopes of 5%–25%, were used for evaluation. The model reproduced observed sediment loads with normalized RMSE of 14%–20% and Nash‐Sutcliffe efficiencies of 0.70–0.89, outperforming the version in which the progressive sealing effect is omitted. Results highlight two competing sealing effects: reduced interrill erodibility suppresses detachment, while increased rainfall excess enhances transport capacity. Their relative importance depends on slope: increasing rainfall excess dominates erosion on gentle slopes, whereas interrill erodibility reduction controls erosion on steep slopes. The model also reveals dynamic regime shifts from transport‐limited to transitional or detachment‐limited during sealing. Comparisons show that neglecting seal formation can substantially underestimate sediment loads, whereas assuming instantaneous sealing overestimates erosion by 37%–71% on gentle slopes and underestimates it by 2%–7% on steep slopes. By resolving slope‐ and rainfall‐dependent sealing dynamics, the proposed model provides a physically based framework for event‐scale interrill erosion prediction, improving the reliability of rainfall‐runoff‐erosion assessments in erosion‐prone landscapes.