A stand-level light interception model for horizontally and vertically heterogeneous canopies

Process-based forest growth models often rely on estimates of absorbed photosynthetically active radiation. Light absorption can easily be estimated using the Lambert–Beer law for simple homogeneous canopies composed of one layer, one species, and no canopy gaps. However, forest canopies are usually not homogenous, vertically or horizontally, and detailed tree-level models have been developed to account for this heterogeneity. These models have high input and computational demands and work on a finer temporal and spatial resolution than is often required by stand level growth models, making them impractical for this purpose. The aim of this study was to develop a stand-level “summary” light model that can account for (1) canopy gaps, (2) multiple horizontal canopy layers that may or may not overlap, and (3) multiple components (species, age classes or dominance classes). The model divides the canopy into horizontal layers that consist of a single component, or multiple components whose crowns overlap vertically. The light absorption of each layer is calculated using the Lambert–Beer law and then partitioned to each component in that layer using weightings based on the leaf area, extinction coefficients and the relative heights of each component within the layer. Canopy gaps within each layer are accounted for by assuming a Poisson-distribution of trees, while taking account of crown surface area-to-leaf area ratio and solar zenith angles, which change with latitude and season. The summary model was compared with a detailed tree-level model and performed well for stands containing up to eight components and for a wide range of stand structures, in terms of trees per ha and multiple canopy layers. For both the whole canopy and when partitioning light between individual components the summary model was nearly unbiased with low relative average errors (−0.26% and −0.30%, respectively) and high model efficiencies (0.94 and 0.87, respectively). Further improvements could be achieved by improving the ability of the model to partition light between components within a given layer. This model can be parameterised with easily obtainable information about crown sizes and extinction coefficients and could be used to examine light dynamics in complex canopies and in stand-level growth models.