Estimation of Vegetation Aerodynamic Roughness of Natural Regions Using Frontal Area Density Determined from Satellite Imagery

Parameterizations of the frontal area index and canopy area index of natural or randomly distributed plants are developed, and applied to the estimation of local aerodynamic roughness using satellite imagery. The formulas are expressed in terms of the subpixel fractional vegetation cover and one non-dimensional geometric parameter that characterizes the plant's shape. Geometrically similar plants and Poisson distributed plant centers are assumed. An appropriate averaging technique to extend satellite pixel-scale estimates to larger scales is provided. ne parameterization is applied to the estimation of aerodynamic roughness using satellite imagery for a 2.3 sq km coniferous portion of the Landes Forest near Lubbon, France, during the 1986 HAPEX-Mobilhy Experiment. The canopy area index is estimated first for each pixel in the scene based on previous estimates of fractional cover obtained using Landsat Thematic Mapper imagery. Next, the results are incorporated into Raupach's (1992, 1994) analytical formulas for momentum roughness and zero-plane displacement height. The estimates compare reasonably well to reference values determined from measurements taken during the experiment and to published literature values. The approach offers the potential for estimating regionally variable, vegetation aerodynamic roughness lengths over natural regions using satellite imagery when there exists only limited knowledge of the vegetated surface.