Using Optimality Principles to Predict Spatio-Temporal Patterns of Vegetation-Atmosphere Fluxes at Leaf to Global Scales

A predictive understanding of biological variation in space and time -- from spatial gradients of light within plant canopies, seasonal fluctuations in temperature and water availability during a growing season, to geographic variation in climate and soil nutrient availability across the land surface -- is a central but challenging goal in biospheric sciences. Functional attributes of vegetation, such as the capacities to exchange carbon, water and energy with the atmosphere, can be assessed based on thermodynamic and aerodynamic properties of the canopy-atmosphere system, however many of these properties cannot be directly measured at the global scale. In lieu of direct measurement, optimization methods based on simplifying theories of the underlying processes, including Maximum Entropy Production (MEP) and economic theories of plant carbon and water relations, are needed to provide sufficient constraint to estimate the required parameters. Using theories of functional coordination in which it is assumed that plants maintain a balance between the supply and demand of a variable (e.g. absorbed radiation, CO2, water) consistent with MEP in complex source-sink physiological systems, it is possible to predict spatial patterns of leaf photosynthetic capacity within plant canopies as well as their temporal variation throughout the growing season. When combined with satellite remote sensing observations of canopy light absorptance (fAPAR), these same theories can be used to predict seasonal variations in leaf and canopy photosynthesis and transpiration, and global spatio-temporal patterns of productivity and evapotranspiration. Predictions using this approach are consistent with observations at leaf to landscape scales based on leaf gas exchange and eddy covariance measurements in arctic to tropical ecosystems.