A sucessão ecológica em ambientes florestais em restauração: estrutura e dinâmica da regeneração natural

Ecological restoration aims to restore the self-sustainability potential of degraded environments. The successional trajectory of areas in restoration may differ from the reference ecosystem on a number of factors, especially the disturbance history of the area and management techniques applied in order to reverse the degradation scenario. In this sense, the objective of this study was to characterize the dynamics of natural regeneration and the structure of the regenerating community in an area of ecological restoration for 25 years on the right bank of the Rio Grande in the Power Station in Camargos that belongs to the Companhia Energetica de Minas Gerais - CEMIG, Nazareno, MG, in order to identify the current state of environmental regeneration. In November 2014, there was the first community inventory of seedlings existing in regeneration subplots (2 x 2 m) within the inclusion criteria (height ≥ 10 cm and DBH <5 cm), divided into 44 subplots (11 plots) in two areas, one area in restoration (REF) and a fragment of native vegetation (FRAG). In October 2015, the second sampling was carried out, in which individuals were resampled and recruits and dead individuals quantified. The parameters abundance, floristic similarity, density, dominance, importance value, equability, richness and diversity were analyzed in both samples. The dynamics was evaluated for the number of individuals and basal area in the period of 2014-2015 and individuals distributed according to height. The history of usage area provoked different behavior between the study environments; the patterns of the FRAG and REF communities along the years allowed the distinction between environments and demonstrated great environmental diversity, with differences in the abundance of species, floristic similarity, richness, diversity, vertical structure and dynamics parameters. In the REF area, there was a considerable increase in the density of individuals along the years and the establishment of different species. In this sense, the evaluation of natural regeneration in REF has shown that resilience is gradually being resumed, a necessary condition to support the potential for environmental self-sustainability. However, given the disturbance history, it can be concluded that the environment moves to an alternative state, which is possibly different from that prior to degradation.

Modelagem do crescimento e produção em povoamentos desbastados de Eucalyptus grandis

The aim of the study was to develop a system of growth and yield models for thinned stands of Eucalyptus spp.; and to assess the behavior of the growth in scenarios with 10% decrease or increase in rainfall. The probability distribution functions Weibull 2 and 3 parameters and Johnson SB for different methods were fitted. Correlation between the fitted parameters with age was evaluated. Dominant height growth behavior was evaluated to check if thinned stand changes its growth when compared to a non-thinned stands. The stand variables dominant height and basal area were projected and simultaneously predicted and projected, respectively. Individual tree equations were fitted, which were fitted as functions of stand level variables in order to decrease the error propagation. R software was used to fit all the proposed models and consequently all the fitted models were evaluated by their parameters significance (F-test) and graphs of predicted values in relation to the observed values around the 1:1 line. Thus, the prognosis system was made by two ways, first one using the full data set, and for the second one the dataset was restricted at age 7.5. Increase and decrease in 20% of rainfall were assessed by updating the site index function. Method of moments was the most precise to describe the diameter distribution for every age in eucalyptus stands for Johnson SB and Weibull 2 parameters pdfs. When observed for each pdf the correlation for their fitted parameters with age, we noticed that shape parameters for a thinned stand were no longer correlated with age, differently of non-thinned stands. Thus, thinning effect was accounted in the basal area prediction and projection modeling. This result emphasized the necessity of applying the Parameter Recovery method in order to assess differences and capture the right pattern for thinned and non-thinned stands in the future. Dominant height was not influenced by thinning intensity. Therefore the fitted Chapman-Richards model did not account for a stand being thinned or not. All the fitted equations behaved with good precision, no matter using full or precocious dataset. The prognosis system using full and/or precocious date set was evaluated for when using Parameter Recovery method for Sb and Weibull pdfs, and by then, graphical analysis and precision statistics showed appropriated results. Finally, the increase or decrease in rainfall regime were observed for eucalyptus stand yields and we may notice how important is to observe this effect, since the growth pattern is strictly affected by water.