The use of spectral data to estimate the structural parameters of vegetation has been considered as one of the most important applications of the remote sensing of ecosystems. In this paper, spectral enhancements and transformations such as vegetation indices, vegetation fraction from mixture analysis from TM/LANDSAT data were used to estimate structural parameters of 'Floresta Nacional do Araripe', in the State of Ceará. The image was calibrated at ground radiance and had the outliers removed before any further processing. Green vegetation fraction as well as vegetation indices and the spectral response of individual bands were all positively and significantly correlated with estimates of the mean height of the sample plot, basal area at the breast height, and basal area at the ground. The correlation coefficients varied from 0.33 to 0.60. The best linear relationships obtained by regression between simple ratio and the structural indices of the vegetation are presented.
remote sensing; spectral mixture analysis; vegetation indices; vegetation structural variables; types of vegetation