Water Availability and the Spatial Complexity of CO2, Water, and Energy Fluxes over a Heterogeneous Sparse Canopy

Abstract

The transient status of water availability to vegetation is central in shaping both the canopy-level processes as well as the spatial aspects of the turbulent exchange between the vegetation and atmosphere. This is examined by merging eddy covariance (EC) flux data with a detailed atmospheric model applied over a heterogeneous, remotely sensed surface. Eddy covariance measurements from a forested area near Mongu, Zambia, in southern Africa are used to parameterize a large eddy simulation (LES) model, which directly simulates the dynamical effects of the lower-atmospheric transport and mixing. High-resolution IKONOS satellite data are used to define the distribution of the leaf area index (LAI) and surface roughness over the 6.4 km × 6.4 km study area. Fluxes along the bottom boundary of the three-dimensional model are represented by a Penman–Monteith formulation for the canopy latent heat fluxes (LEc), and CO2 uptake by the vegetation (Ac) is added by drawing upon an observed bulk relationship from the EC data. Bare soil fluxes are modeled separately. The effect of water limitation on the vegetation fluxes is introduced by the parameter ϕc, which acts upon the canopy conductance (gc) in the form of a Jarvis-type limitation. LAI and vapor pressure deficit (D) of the canopy sublayer air are the other controls on gc. Five model simulations were conducted in which ϕc was adjusted to represent various stages of vegetation water limitation. The results reveal a two-way spatial interaction between the vegetation and the canopy sublayer air, the characteristics of which are dependent upon ϕc. For the well-watered cases, the spatial distribution of D was most predictable but its impact on the vegetation-controlled fluxes was unimportant, while in water-limiting cases the distribution of D was least predictable yet most important in terms of its effect on the fluxes. Increased scaling complexity in estimating fluxes over heterogeneous vegetation was found to be associated with the more water-limited conditions.