Alometria e distribuição de palmeiras em terra-firme na Amazônia central

Abstract

The importance of tropical ecosystems in global biogeochemical cycles, in particular, the carbon cycle and its relation to climate change is still not fully understood, partly because of the diversity of ecosystems present in these regions. Studies on carbon quantification have been conducted under different approaches, however uncertainties in estimates persist and justifications are recurrent. In these evaluations, three aspects are fundamental: stocking, flow and distribution among the different compartments (trees, trees, shrubs, vines, seedlings and palm trees). Among these compartments the palm trees have been neglected once that their evaluation is sometimes disregarded; being evaluated using equations developed for trees, conversion factors or global equations. In the Amazon, palm trees constitute such an important component that the IBGE uses its dominance to distinguish forest types in the Brazilian System of Classification of Plants. Thus, in the present study, quantitative and qualitative approaches were developed to relate the contribution of palm trees to forest dynamics in the long term, to evaluate the accuracy of global allometric equations in estimating palm biomass, and to verify possible relationships between the forest, environmental variables and palm trees density and their potential to influence forest biomass dynamics. The results are presented in chapters. In Chapter 1, for the allometric studies, twenty individuals of the Euterpe precatoria Mart. were used and allometric models were developed to estimate biomass components such as stem mass Ms, foliage mass MI, root mass Mr and mass above ground Ma, were developed for this palm tree specie which is the most abundant in the Amazon. The best predictors were D2Hρ for aerial and total biomass. Comparisons with several models suggest that variations in the D-H and ρ ratio should be taken under consideration for the development of allometric models to estimate biomass in palm tree species. In Chapter 2, forest inventory data from 2002 and 2012 were used to evaluate the dynamics and effects of the topographic variation on biomass and its long term growth. Six inventory plots were installed in the plateau and bottomland area. The aerial biomass in the plateau was estimated at 315.4 t.ha-1 ± 33.2 (SE), and in the shallow in 224.2 t.ha-1 ± 20.3, these different stocks being influenced by canopy height in both toposequences. The palm tree biomass stock in the bottomland was 2.7 t.ha-1, significantly higher than in the plateau where the stock was evaluated at 0.24 t.ha-1. In Chapter 3, the relationship between environmental variables and forest structure was evaluated in order to identify potential explanatory relationships for palm tree distribution. For this evaluation, data were analyzed using Principal Component Analysis (PCA), Non-metric Multidimensional Scaling (NMDS), cluster analysis, discriminant analysis and multiple and simple linear regressions. It was verified that the environmental variables (soil texture and presence of water) strongly influence the distribution of the palm trees, making it possible to identify the potential influence of this group on the composition of the forest structure.