Forest vertical structure from multibaseline interferometric radar for studying growth and productivity

This paper describes the first quantitative attempt to estimate the vertical profile characteristics of forests from multibaseline interferometric radar (INSAR) data. In order to calculate the accuracy with which vertical characteristics can be estimated with INSAR, a simple, two-layer (sucanopy and canopy) forest is assumed to be observed by 4 baselines, most of which can be realized by current airborne systems. Calculations suggest that the height of the subcanopy layer, the total height of the forest, and the altitude of the underlying surface can be estimated with accuracies of 2-4 meters, with multiple baselines at C-band (wave length=0.056 cm). Extinction coefficients of each layer can be estimated with accuracies of 0.1 db/m. The estimation of height and extinction parameters from INSAR data is nonlinear, and accuracies therefore may depend heavily on the actual characteristics of the observed forest. This paper also discusses the application of vertical profile information estimated from INSAR to the identification of old-growth forests and their characteristics of growth and productivity. Because old-growth forests are characterized by layered vertical structures, the subcanopy and canopy characteristics discussed in this paper may enable identification and monitoring of old-growth stands, which in turn would facilitate ecosystem management and assessment of primary production on a global scale.