Analysis and simulation of interception characteristics of broad-leaved forest canopy based on parameter optimization: Dianchi basin case

• Three forest types’ rainfall redistribution characteristics and key factors were analyzed. • Gauss-Newton algorithm optimizes parameter calibration to conform to natural laws. • A refined model introduces canopy density and LAI indicators to improve accuracy. The redistribution effect of the canopy precipitation interception is a critical component of the water cycle and water balance in basin ecosystem. However, traditional canopy interception models often fail to fully consider the reasonableness of model parameter values in practical applications. This study investigated the rainfall redistribution characteristics of three typical subtropical semi-humid evergreen broad-leaved forests in the Dianchi Basin of China. The model structure was improved by incorporating both canopy density and leaf area index (LAI) as indicators, and optimized the parameters using the Gauss-Newton algorithm, with appropriate threshold settings for absorption and additional interception. The results indicated that: (1) a power-function relationship exists between canopy interception and precipitation, which significantly influences precipitation redistribution. The interception rate for the three tree species ranged from 29.3% to 39.8%. (2) LAI and canopy closure are key factors influencing interception characteristics and should be included simultaneously in canopy interception model. (3) The Gaussian Newton’s algorithm effectively determines the optimal parameters of the forest canopy interception model under constraints. The refined model showed a good agreement with observed canopy interception for the three broad-leaved forests in the basin. These findings provide essential data for future studies on the role of forest canopy interception in watershed water cycles. The refined modelling approach minimizes deviations from natural processes when calibrating the model parameters, ensuring both scientific rigor and model accuracy. Furthermore, it offers valuable insights for parameter calibration in other hydrological and environmental models.