Remote sensing and digital images are techniques that have the potential to assist in the analysis of terrestrial ecosystems in spatial scale. It is known that a forest monitoring is more accurate when you know the ecosystem features at different spatial scales. Relate measurements with remote sensing and obtained in the field (scales of detail) is challenging since several physical interference (biotic and abiotic) can provide errors in these relationships. One of the difficulties in obtaining estimates of parameters and vegetation biophysical variables with remote sensing are the spatial resolution is often low and this complicates the relationship between the regional data, location and detail. Thus, this study aims to evaluate the similarity and the correlation between canopy reflectance and leaf and propose a model that improves the estimates in leaf scale. We used cluster analysis, t-test, Pearson correlation and regression to analyze the similarity between samples, the degree of correlation and regression finally adopted for the model. The results were quite satisfactory and showed that there are significant differences between the reflectance in the leaf and canopy in Savannah for a significance level of 0.01. The model developed here can be using with high significance for high spatial resolution images such as IKONOS.
