Influência de características edáficas na seletividade ambiental de espécies arbóreas ao longo de um gradiente topográfico em floresta de terra firme na Amazônia Central

Abstract

The Amazon forest is distinguished by being the largest natural reservoir of plant wealth of the planet, where different each of their forest habitats has very rich and wide variety of species, often unique to a particular environment. Understanding how plant species are organized in their environment and how they use available resources is an important tool for the management and conservation of biodiversity. Thus, this study aimed to group tropical tree species according to environmental heterogeneity of the area, focusing on chemical and physical variables soil. The study was conducted in two forest sites in Central Amazonia: (a) forest management company Mil Precious Woods (PWA), the municipality of Silves, and (b) Experimental Field of Embrapa, located in the Agricultural District Suframa (CEDAS) Rio Preto da Eva. They were inventoried eight (08) continuous hectares at each site. All individuals with DAP≥ 10 cm were marked and identified in 2014 and had their registered UTM coordinates. 200 soil samples were collected at a depth of 0-20 cm per area, which were submitted to chemical and particle size analysis. They were surveyed 5,040 individuals distributed among 356 species on the site and give in 4,279 individuals represented by 203 species in the PWA site, with 191 species exclusive to the site and give in 69 exclusive PWA site. Among the five most abundant species on the site CEDAS are Eschweilera bracteosa, Protium amazonicum, Eschweilera coriacea, oblongifolia Licania and Pouteria sp, corresponding to 21% of individuals. In the PWA site, are Trattinnickia burserifolia, Eschweilera coriacea, Ocotea neesiana, Eschweilera collina and Trymatococcus amazonicus, which together account for 22% of all individuals. Although they are the same forest type, the two sites have distinct floristic composition, showing no floristic and structural homogeneity in upland forest in Central Amazon. The canonical correspondence analysis (CCA) indicated a significant correlation between the distribution of tree species along the topographic gradient and soil variables, with more pronounced effect of soil texture. The results suggest that it is possible to separate groups of species on the basis of soil characteristics, and texture is the variable which best describes the distribution of species throughout the soil gradient, mediated by the topography variations.