Performance Assessment of the Mulder Model in Simulating Interception Loss in Tropical Forest

Abstract

This study evaluated the performance of the Mulder Model in estimating rainfall interception loss in the tropical rainforest of Bukit Lagong, Malaysia. Rainfall interception is a key component of the forest hydrological cycle, particularly in tropical regions characterised by high rainfall and dense canopy structures. Although the Mulder Model has been widely applied in temperate climates, its suitability for tropical environments remains insufficiently tested. A triangulation approach was adopted by comparing interception loss calculated from field measured gross rainfall, throughfall, and stemflow with model simulations. Secondary daily meteorological and canopy data from previous studies were used. The model was applied using site-specific parameters, including a canopy storage capacity of 0.8091 mm, a wet canopy evaporation rate of 0.25 mm/h, and an assumed rainfall duration of 4 hours. Model performance was evaluated using RMSE. Results indicated that the model tends to underestimate or overestimate interception during high-intensity rainfall and short dry intervals, typical of tropical climates. Variations in canopy density, leaf area index, and rainfall frequency contributed to discrepancies. The findings highlight the need for local parameter calibration to improve model reliability in tropical forests.