Abstract
Techniques for forest restoration have been widely developed over the past decades, allowing the reestablishment of vegetation in extreme cases such as surface mining. However, there are still issues related to management and monitoring that require further understanding, especially concerning comparisons with reference ecosystems. In this study, hierarchical agglomerative clustering (HAC) with uncertainty estimation is proposed as a methodology for forest restoration assessment. For this purpose, analysis was made of phytosociological variables for 27 plots located in reforested closed mines and in the Amazon forest reference ecosystem. The technique grouped the reference ecosystem separately from the reclamation sites. The HAC was affected by dependency among the analyzed variables, and heterogeneity was observed for all the phytosociological parameters in the cluster groups formed by the mining locations. However, each group showed specific characteristics related to the different environmental conditions and the forest restoration performance. The results demonstrated that HAC with uncertainty estimation was more suitable for defining groups, compared to the classical approach, offering a promising methodology for evaluation of the outcomes of forest restoration and for guiding management actions in disturbed tropical forests.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call