Pii: S0304-3800(00)00353-7

We investigated the effects of the shape of leaf area profiles and the number of canopy layers on simulated sensible and latent heat fluxes using a gradient diffusion-based biometeorological model. Three research questions were addressed through simulation experiments: (1) Given the same amount of cumulative leaf area in the vertical direction, how does the shape of the leaf area profile affect simulation results? (2) For a given leaf area profile, how does the number of layers influence the simulation results? (3) How do these two factors interact with each other in affecting the simulated energy fluxes? Our results demonstrated that the scheme of canopy stratification could substantially affect the simulated energy fluxes, and that the effect exhibited a consistent pattern — an S-shaped response curve. There existed a minimal number of layers for achieving a required degree of accuracy. The turning point of the S-shaped curve represented the optimal number of layers, indicating a desirable balance between the accuracy of the model and demands for computation and data collection. Our results also showed that differences in the shape of leaf area profiles alone could significantly alter the simulated energy fluxes even if the total amount of leaves in the canopy and the values of all other model parameters remained the same. Furthermore, the shape of leaf area profiles interacted with the number of layers in affecting the simulated energy fluxes. While considerable impacts were observed for all leaf area profiles, complex (nonlinear) shapes exacerbated the effects of changing the number of layers. © 2000 Elsevier Science B.V. All rights reserved.