Development of a Ray-Tracing Based Canopy Radiation Model

Solar radiation is the main driver in the global water and energy cycles and a primary component of the interaction between plant canopies and the atmosphere. How it is attenuated, reflected, absorbed, and re-emitted from the canopy plays an important role in determining the distribution of temperature and water vapor within the canopy and how energy in the form of latent and sensible heat fluxes is exchanged between the canopy and the atmosphere. For plants, this energy exchange is of great importance in driving growth, changes in leaf temperatures, and gas exchange. Modeling radiation transfer in plant canopies is difficult due to their intricate geometry and the complex interactions that form as radiation is absorbed, reflected, and re-emitted by randomly distributed elements. Ray tracing algorithms borrowed from the computer graphics community have become a versatile tool in modeling radiation exchange. This class of models allows for efficient parallel calculation of incoming solar and emitted longwave radiation exchange.

The present research involves the development of a portable radiation transfer routine to calculate the short and longwave radiation components of the surface energy budget in a complex plant canopy. This tool is designed to couple with both complex high-resolution and high-efficiency reduced-model computational fluid dynamics (CFD) codes. Our ultimate interest is to characterize the role that radiation and canopy geometry play in land-atmosphere coupling and the overall canopy energy budget.. The ray tracing-based radiation code is built on NVIDIA's highly efficient OptiX framework and can be executed on one or more graphics processing units (GPUs), each of which consist of a large number of parallel processing elements. The radiation routine is designed to be embedded within a CFD code that solves for the conservation heat, water vapor, and momentum in the canopy. This strategy results in two-way coupled, physically realistic plant and ground surface conditions which can be calculated in parallel with the mass, momentum, moisture and temperature solutions and accessed in discrete intervals as needed.