Key points are not available for this paper at this time.
Accurate numerical modelling of wind-crown interactions is essential for regulating urban temperature, ecological forecasting, and assessing tree failure risks. Conventional modelling approaches typically simplify tree crowns as locally homogenous porous media with macroscopically uniform or vertically varying leaf area density (LAD) distributions. However, real-world foliage distributions can be spatially heterogeneous in both horizontal and vertical directions, dependent on variations in tree architecture, species, and environmental conditions. This study first utilizes LiDAR (Light Detection and Ranging) scanning to obtain the detailed three-dimensional (3D) foliage distributions of selected trees in Singapore, and the LiDAR-derived results demonstrate the substantial crown heterogeneity for these selected trees, suggesting the importance of incorporating heterogeneity considerations in the numerical modelling of wind-crown interactions in real-world scenarios. The LiDAR-derived data are further processed to obtain a three-dimensional LAD distribution, which is incorporated into the crown source term. Subsequently, a computational fluid dynamics (CFD) model for wind-crown interactions is established to examine the sensitivity of the predictions to the simplifications of crown morphology and LAD heterogeneity. The results illustrate the significant differences in airflow patterns, aerodynamic drags, and bending moments depending on the simulated crown geometries and LAD distributions, and underscore the limitations of the simplified LAD assumption that are still commonly made in the literature. Overall, the present study illustrates that incorporating spatially explicit LAD distributions is critical to improving the accuracy of predictions for wind-crown interactions and should be prioritized in future research.
Zhang et al. (Fri,) studied this question.