Simulation of micrometeorology of crops, some methods and their problems, and a few results

Abstract

Three different models for the energy exchange of the canopy-soil system with the atmosphere are described. A first approximation is the so-called “big leaf” model. In this model, the canopy surface is characterized by a single crop resistance and by an atmospheric boundary layer resistance. A second approximation, and a major improvement, is a “greenhoused canopy” model which has stratification with respect to radiation and wind velocity, but not with respect to air temperature and humidity. In the third approximation, gradients in air properties inside the canopy are also included. A scheme with distributed resistances (first-order methodology) is used to describe the in-canopy profiles of vapour pressure deficit, D. The steady state solution for this scheme can be found by matrix solutions (Waggoner et al., 1969), reworked by Chen (1984) into a much faster and more elegant method. For the calculation of total fluxes, the greenhoused canopy model was found to be sufficient. In complicated models, numerical instabilities remain lingering problems, probably because of the multitude of feedbacks. Examples of different types of numerical instabilities and of methods to solve them are discussed. Under conditions of water shortage, stomatal closure will occur, reducing the rates of both transpiration and CO 2 assimilation. Canopy temperature is increased, but for proper interpretation of canopy temperatures measured by remote sensing, the structure of the canopy should be taken into account. The simulated gradients of aerial properties were found to be much stronger in short and dense canopies than in taller and aerodynamically rougher canopies. Soil evaporation was potentially larger under broad-leafed species than under grass. Countergradient transport can be understood by second-order modelling, but it can also appear in the simpler first-order methodology by considering sudden wind gusts alternating with periods of low wind. Mean gradients of scalars are dominated by the long periods of low wind and mean fluxes by the brief gusts. This phenomenon undermines the validity of incanopy flux measurement on the basis of observed mean gradients.