Modeling Plant Canopies in the Earth System: Moving Beyond the Big-Leaf Paradigm to Multilayered Canopies

The central paradigm of land surface models since the 1980s has been to represent plant canopies as a homogeneous "big leaf" without vertical structure, though with separate fluxes for vegetation and soil. A critical advancement was to analytically integrate leaf physiological processes over profiles of light and nitrogen in the canopy and to represent sunlit and shaded portions of the canopy. In the Community Land Model (CLM5), the sunlit/shaded big-leaf canopy structure has remained essentially unchanged over some 20 years of model development while soil physics has expanded to 20 soil layers up to 8.5 m deep and an additional 5 bedrock layers to a depth of 50 m. Although the model is highly resolved in its soil vertical structure, vertical structure in the canopy is ignored. In this study, we compare a multilayer canopy model with the equivalent single-layer canopy to examine the importance of non-linear vertical profiles within the canopy when modeling surface fluxes. Comparison with flux tower measurements show that for the particular simulation, sensible heat flux, latent heat flux, gross primary production, friction velocity, and radiative temperature for the one-layer canopy are degraded compared to a benchmark 42-layer canopy. The solution converges to that of the 42-layer canopy with 5-10 layers. Key non-linearities that cause the scaling error are the vertical profiles of solar radiation, air temperature, vapor pressure, and wind speed. The scaling error arising from radiative transfer relates to the scattering of diffuse radiation in the canopy and increases with greater downwelling diffuse radiation incident on the canopy. The vertical profile of photosynthetic capacity (Vcmax) also contributes to scaling error and can be used to tune the single-layer canopy to more closely match the 42-layer canopy (but only for latent heat flux and gross primary production). The vertical profile of leaf water potential, in which the upper canopy is water stressed on dry soils, also contributes to the scaling error. The results of this study harken back to a 30 to 40-year-old debate about how to model plant canopies and whether big-leaf models are adequate. Our results, with substantial error between big-leaf and multilayer canopies, suggest that it is time to move beyond the big-leaf paradigm to embrace a multilayer framework.