In the Amazon, several mining companies are located in protected areas. Ecological restoration after mining should consider several attributes. Among these, there is the need of the restored ecosystem to contain a set of key species that occur in the reference ecosystem, and thus, provide an appropriate structure of the community, and as much as possible a greatest number of native species. The objective of this work was to analyze a phytosociological index and socioeconomic status (IFSE), obtained by multivariate analysis. The model included six quantitative variables (abundance, frequency, dominance, biomass, commercial value of timber and amount of non-timber forest products) and an equal number of qualitative variables. In a rain forest area of 1,321 ha, 315 plots of 0.25 ha were sampled where all trees with DBH > 10 cm were identified and measured. It was identified 36,298 individuals, distributed in 898 species, and comprising 62 families. Using the proposed index the species, they were ranked in three categories of priority- phytosociological and socioeconomic (high, medium and low). The appropriateness of factor analysis was determined by KMO and Bartlett tests. The Bartlett test evaluated the general significance of the correlation matrix at the level of 1% probability. The KMO test showed that the variables are correlated and factor model presented a very good fit to the data. The use of multivariate analysis was statistically validated using the IFSE. Among the 898 species analyzed, 25 were selected as priorities (high and medium categories) and are recognized as key species. It was estimated the planting density of these species for recovery of mined areas in the Amazon, with biological and environmental conditions similar to those reported in this study. The diameter distribution of key species was analyzed and compared with the forest inventoried. The successional stage of key species was also analyzed.
Key specie; Multivariate analysis; Land restoration
