Search for: All records

Abstract Large‐scale canopy models utilizing satellite data typically rely on simplified models to describe radiation absorption by vegetation. However, these models' accuracy can be limited when applied to heterogeneous row‐oriented canopies because of assumptions of canopy homogeneity, or oversimplification of the impacts of heterogeneity. This study evaluated existing model assumptions and developed a novel geometric binomial model for radiation absorption in discontinuous canopies with an ellipsoidal crown envelope. The simple models considered include a one‐dimensional turbid medium model (i.e., Beer's law), two models incorporating constant or variable clumping factors, and the geometric binomial model adapted to predict radiation absorption for canopies with ellipsoidal shaped crowns. Compared to Beer's law and the models with clumping factors, the proposed binomial model accounts for scattering, variable radiative path lengths through vegetation, diffuse radiation, and crown shadow overlap. The simplified models were evaluated against a sophisticated three‐dimensional (3D) leaf‐resolving radiation model (Helios) and field measurements collected in almond and olive orchards. Results indicated that Beer's law considerably overpredicted radiation absorption for a wide range of virtually generated canopies and field observations. For the model with variable clumping factor, errors increased as the radiative path length through the canopy increased. Among the simple models, the proposed binomial model resulted in small errors across all canopies (index of agreement: 0.91–0.99). With additional inputs related to the canopy geometry, this approach could be integrated within the shortwave absorbed‐radiation component of large‐scale canopy models to improve the estimation of biophysical processes such as photosynthesis and transpiration.