Light interception in plant canopies is most commonly estimated using a simple one-dimensional turbid medium model (i.e., Beer's law). Inherent in this class of models are assumptions that vegetation is uniformly distributed in space (homogeneous) and in many cases that vegetation orientation is uniformly distributed (isotropic). It is known that these assumptions are violated in a wide range of canopies, as real canopies commonly have heterogeneity at multiple scales and almost always have highly anisotropic leaf angle distributions. However, it is not quantitatively known under what conditions these assumptions become problematic given the difficulty of robustly evaluating model results for a range of canopy architectures. In this study, assumptions of vegetation homogeneity and isotropy were evaluated under clear sky conditions for a range of virtually-generated crop canopies with the aid of a detailed three-dimensional, leaf-resolving radiation model. Results showed that Beer's law consistently over predicted light interception for all canopy configurations. For canopies where the plant spacing was comparable to the plant height, Beer's law performed poorly, and over predicted daily intercepted sunlight by up to ∼115%. For vegetation with a highly anisotropic leaf inclination distribution but a relatively isotropic leaf azimuth distribution, the assumption of canopy isotropy (i.e., G=0.5) resulted in relatively small errors. However, if leaf elevation and azimuth were 1

No abstract

Abstract. Despite recent advances in the development of detailed plant radiative transfer models, large-scale canopy models generally still rely on simplified one-dimensional (1-D) radiation models based on assumptions of horizontal homogeneity, including dynamic ecosystem models, crop models, and global circulation models. In an attempt to incorporate the effects of vegetation heterogeneity or “clumping” within these simple models, an empirical clumping factor, commonly denoted by the symbol Ω, is often used to effectively reduce the overall leaf area density and/or index value that is fed into the model. While the simplicity of this approach makes it attractive, Ω cannot in general be readily estimated for a particular canopy architecture and instead requires radiation interception data in order to invert for Ω. Numerous simplified geometric models have been previously proposed, but their inherent assumptions are difficult to evaluate due to the challenge of validating heterogeneous canopy models based on field data because of the high uncertainty in radiative flux measurements and geometric inputs. This work provides a critical review of the origin and theory of models for radiation interception in heterogeneous canopies and an objective comparison of their performance. Rather than evaluating their performance using field data, where uncertainty in the measured model inputs and outputs can be comparable to the uncertainty in the model itself, the models were evaluated by comparing against simulated data generated by a three-dimensional leaf-resolving model in which the exact inputs are known. A new model is proposed that generalizes existing theory and is shown to perform very well across a wide range of canopy types and ground cover fractions.

No abstract