Estimation of scalar source/sink distributions in plant canopies using Lagrangian dispersion analysis: Corrections for atmospheric stability and comparison with a multilayer canopy model

Source/sink distributions of heat, water vapour and CO2 within a rice canopy were inferred using an inverse Lagrangian dispersion analysis and measured mean profiles of temperature, specific humidity and CO2 mixing ratio. Monin-Obukhov similarity theory was used to account for the effects of atmospheric stability on sigma(w)(z), the standard deviation of vertical velocity and tau(L)(z), the Lagrangian time scale of the turbulence. Classical surface layer scaling was applied in the inertial sublayer (z > z(r)uf) using the similarity parameter zeta = (z - d)/L, where z is height above ground, d is the zero plane displacement height for momentum, L is the Obukhov length, and z(r)uf approximate to 2.3h(c), where h(c) is canopy height. A single length scale h(c), was used for the stability parameter 3 = h(c)/L in the height range 0.25 < z/h(c) < 2.5. This choice is justified by mixing layer theory, which shows that within the roughness sublayer there is one dominant turbulence length scale determined by the degree of inflection in the wind profile at the canopy top. In the absence of theoretical or experimental evidence for guidance, standard Monin-Obukhov similarity functions, with zeta = h(c)/L, were used to calculate the stability dependence of sigma(w)(z) and tau(L)(z) in the roughness sublayer. For z/h(c) < 0.25 the turbulence length and time scales are influenced by the presence of the lower surface, and stability effects are minimal. With these assumptions there was excellent agreement between eddy covariance flux measurements and deductions from the inverse Lagrangian analysis. Stability corrections were particularly necessary for night time fluxes when the atmosphere was stably stratified. The inverse Lagrangian analysis provides a useful tool for testing and refining multilayer canopy models used to predict radiation absorption, energy partitioning and CO2 exchanges within the canopy and at the soil surface. Comparison of model predictions with source strengths deduced from the inverse analysis gave good results. Observed discrepancies may be due to incorrect specification of the turbulent time scales and vertical velocity fluctuations close to the ground. Further investigation of turbulence characteristics within plant canopies is required to resolve these issues.