| Riassunto: | Despite its importance in ecology, forestry and resource management the relationship between remote sensing data and forest inventory data are still poorly understood, in particular regarding the structural complexity of woody canopy elements. In this paper we evaluate by multivariate techniques and Generalized Linear Models (GLM), the influence of 4 woody canopy structure variables by strata (height, DBH, crown diameter and density) on the Landsat TM reflectance data. We analyzed three forest types in the Cofre de Perote National Park. Biplots suggest differences related to the type of forest and the metric expression of structural attributes. GLM indicate that the total reflectance shows differential response that can be associated to canopy structural complexity and the type of forest with some limitations related to the sensitivity of the sensors. Landsat bands 1-3, 5-7 showed a better association with forest complexity variables, in particular for Pinus hartwegii forest and mixed forest. Apparently forest structural attributes and limitations of Landsat sensor sensitivity itself prevents the identification of reliable association between reflectance and complexity variables in Abies religiosa forest.
|
|---|
