Fine-scale physiological and ecological processes are of great importance in governing large-scale ecosystem functions as well as the global carbon cycle. In the individually-based terrestrial biosphere model ED developed by Moorcroft et al. (2001), such fine-scale spatial heterogeneity is accounted for using a size- and age-structured approximation, where vegetation in the same grid cell of the model is divided horizontally into different patches according to their age (i.e., number of years since last disturbance), and vegetation in the same patch is divided vertically into different cohorts according to their size (i.e., height). As a result, vegetation canopy in ED features a large number of leaf layers in the vertical direction. To couple this biosphere model with global climate models, here we develop a new multiple-layer canopy model which efficiently solves the interactions between size-structured vegetation and the atmosphere using an implicit approach but without simultaneously solving the large matrix of equations. The formulation of the multiple-layer canopy model and some preliminary results will be presented, and the implications of assumptions made will be discussed.