A 3D model of mass and energy transfers in vegetated canopy

Exchanges of energy, water, CO2 and aerosols between vegetated landscapes and the atmosphere play a central role in the dynamics of weather, climate, surface hydrology and terrestrial ecology. They are regulated and affected by biophysical processes that interact non-linearly with environmental forcing. A great diversity of models simulates soil-vegetation-atmosphere transfers. Most of them are built on the hypothesis that vegetation is an horizontally uniform medium. However, canopy architecture greatly affects biophysical processes. This is especially due to the 3-D distribution of radiation budget. Neglect of canopy architecture can lead to very large errors (e.g. 40% for photosynthesis).

We present a newly developed 3-D model of mass and energy exchanges. It relies on a realistic description of the studied landscape. A major feature is to fully resolve the interaction between vegetation and its microclimate, thereby producing realistic solutions that satisfy both the physiological properties and their corresponding microclimate. The model combines 4 main modules: - (1) a radiative component managed by the 3-D radiative transfer DART model which simulates the 3-D distribution of absorbed radiation by the Earth surface from the visible to the thermal infrared optical domain and also the corresponding directional reflectance and brightness temperature images; - (2) a vegetation functioning component managed by the leaf photosynthesis model of Collatz which leads to a 3-D distribution of net carbon assimilation and leaf transpiration; - (3) a 3D soil functioning module that treats the heat and mass transfer and also soil respiration; - (4) a micrometeorological module that simulates turbulent transfer with a "classical" approach and diffusion transfer with the LNF theory. This last module is the only 1-D component. This approximation arises from the fact that radiative transfer is the main process that controls energy balance and all biophysical processes while turbulent transport re-distributes the mass and energy fluxes throughout the atmospheric boundary layer.

This model was successfully tested against field measurements over 2 sites: a fallow land of the MUREX experiment, Southwest France near Toulouse and a pine forest canopy of the Euroflux network, Southwest France near Bordeaux. All simulated fluxes are in good agreement with experimental data under many environmental conditions (cloudy day, humid air, dry air, hot day...). In the pine forest site, simulations indicate strong horizontal gradients of leaf and soil temperature that directly affect energy and mass fluxes. This underlines the fact that non linearity of biochemical process occur verticality and also horizontally and that oversimplified formulations may significantly reduce the accuracy of simulated fluxes.