Simulating the distribution of rainfall interception ratio in a Masson's pine plot using terrestrial laser scanning data

Abstract

AbstractThe distribution and structure of leaves and branches influence the rainfall interception ratio (RIR) and water cycling processes in forests. However, measuring the three‐dimensional structure features of canopy organs to evaluate their effects on the RIR is labour‐intensive and time‐consuming. Modelling the spatial distribution of the RIR within a forest canopy after rainfall thus remains challenging. To address this difficulty, we used point clouds scanned by a terrestrial laser scanner (TLS) from Masson's Pines to extract fine‐scale canopy structure features. Next, we applied an exploratory factor analysis (EFA) to evaluate quantitatively how the various canopy structure features affect the RIR of canopy components. Finally, we simulated the RIR distribution by combining the interception coefficient of canopy organs estimated from the EFA results and throughfall volume measured by rain gauges set in the study site. The results show that the inclination angle of canopy organs and their upper gap fraction more strongly affect their RIR. The average absolute error of simulated rainfall interception volume (VIN) is less than 0.57 mm, and the related root mean squared error (RMSE) is less than 0.41 mm after rainfall. For measured rainfall events, the mean absolute error and RMSE of simulated VIN intercepted by nearby canopy organs above each rain gauge are less than 0.52 and 0.39 mm, respectively. The accuracy of the RIR simulation is degraded by the uneven distribution of canopy receivable rainfall volume, noise and missing point clouds in the TLS datasets. Modelling RIR distribution within the canopy based on TLS point clouds provides a chance for simulating water cycling processes in forests during rainfall.