| Summary: | Canopy height is an important attribute that allows characterizing the forest vertical structure and analyze changes in vegetation cover over time. The objective of this study is to develop an approach for a spatio-temporal analysis of the tropical forest canopy using multi-temporal photogrammetric images. The datasets based on film and digital cameras are used to generate canopy height models and extract structural variables (tree height, relative variance between tree heights, and density of higher trees in the upper canopy). The combination of these variables is used in the analysis. Each variable is segmented into ordinal categorical classes in its respective dataset with temporal class gradients being obtained between the periods of the multi-temporal datasets. Experiments were conducted in a tropical forest under regeneration and with diversity of tree species in different successional stages. Three sets of images (years 1978, 2010, and 2017) were used for analyzing canopy cover changes. A classification based on histograms of gradient classes indicated and quantified the most frequent behavior of the canopy over time. The results showed that the most significant variations in cover changes could be explained by 13 classes of temporal gradients, which described 88% of the canopy. This classification was validated with field data collected in sample plots. From the results, it can be concluded that the proposed approach provides accurate assessments of the spatio-temporal canopy cover changes for forest management. © 2021 The Author(s)
|
|---|
