Gap Dynamics in managed Amazonian forest: structural and ecophysiological

Abstract

This thesis examined the role of several environmental variables in determining treefall gap colonization processes. The studies were carried out in a terra-firme forest near Manaus, Amazonas State, Brazil. This was examined at three levels: individual seedlings, gaps and communities.

At seedling level, seed germination and seedling establishment were analysed in an attempt to understand the regeneration niche of the species belonging to different ecological guilds. Field trials showed that pioneer and clímax species presented different niche preferences for seed germination and ecological longeyity. Morphological and physiological attributes were compared in seedlings grown in a controlled environment, testing different levels of photosynthetically active radiation (PAR) and N-nitrate. The results showed that both these variables and their interactions influenced plant physiology, morphology and growth. The plasticity in climax and pioneer species varied differently according to the physiological and morphological traits analysed. A clear difference between the two ecological guilds was not observed for all traits.

Studies in artificially opened gaps showed that the nature of these environments played a key role in seedling establishment and growth. Of the four species examined (Cariniana micrantha, Dinizia excelsa, Dipteryx cf alata and Minquartia guianensis) all enhanced their growth rates in gaps irrespective of their ecological guilds. Clear niche differentiation was ohrved only in Minquartia and Dinizia. Gap size specialization per-se was not observed among the species during the study period, although gap sizes were loosely related to other controlling factors.

At community level, synecological studies were carried out in several gaps opened by logging in 1987. Multivariate analysis showed that original gap size was not the most important factor in controlling the composition of the secondary vegetation. Among the environmental variables studied, the interaction of soil nitrogen and light environment (direct site factor [DSF]) was indicated as the most important determinant of floristic composition. Between 4-6 years after disturbance, pioneer species dominated regeneration, but climax species were also recruited. Within each gap, the distance from the forest edges is also indicated as a controlling variable. The number of species increased in all gap sizes over time. However;no correlation between species richness and gap sizes was observed. Among the successional models, a tolerance model best explained the colonization in most of the gaps although the inhibition model is also useful in certain cases. The implications of these results for forest management are briefly examined.