Potencial reprodutivo e regenerativo de espécies arbóreas em florestas secundárias na Amazônia Central

Abstract

The main objectives of this study were to: (1) evaluate the relationship between seed weight, seed number, fruit weight, fruit number, and seed number per fruit using a conceptual model termed "seed packaging"; (2) evaluate the relationship of seed and fruit variables (weight and number), duration of fruiting and germination with seedling and adult tree density; and (3) evaluate the effects of topography and micro-site conditions on the seed bank and establishment of tree species. The study was conducted in the experimental area of the Biological Dynamics of Forest Fragment Project, located 80 km North of Manaus, Amazonas. For objectives 1 and 2, flowering and fruiting phenology were monitored monthly amog 12 pioneer species over three years. The weight of seeds and fruits, number of fruits and seeds and germination rate were determined for each species. The density of seedlings and adult individuals was estimated using sampling plots. For objective 3, 21 artificially-created gaps in 20-year old secondary forests were established in three topographic levels (plateau, slope and bottomland), where litter (with and without litter removal) and soil treatments (soil turned and soil unturned) were allocated in four sub-plots at the center of each gap. Although the 12 pioneer species belonged to the same ecological guild, there was high variation in both phenological patterns (ranging from continuous to supra-annual) and reproductive traits, for which the weight and number of seeds and fruits varied widely among species. The number of seeds was partly related to the weight of seeds, but when incorporating number and weight of fruit into the same model, seed number was almost completely explained by these variables. The number of seeds per fruit was positively correlated with fruit weight and seed number and negatively correlated with seed weight and fruit number. The density of seedlings and adults of the 12 pioneer species were best explained by fruit weight and number. After 20 years of natural regeneration in abandoned pastures, the seed bank was still dominated by a few tree species typical of early succession. There was a strong positive effect of litter removal and a moderate positive effect of soil turning on seedling recruitment. Moreover, recruitment and initial growth were also favored in the bottomlands, probably due to the greater availability of water and fertile soils. The results of this study provide important tools for the management of secondary forests, for which the management of seed bank by removing litter coupled with soil turning are useful tools for activating the seed bank and accelerate seedling emergence.