Estratégias funcionais de plantas arbóreas em uma cronossequência na floresta tropical úmida

Abstract

In communities assembled by deterministic processes, abiotic or biotic filters select species and functional traits with higher fitness. Based on this, the hypothesis of this work is that abiotic and biotic filters operate over a chronosequence, generating different community assembly processes. Thus, the following predictions were made: I) If functional integration exists between organs, different processes of community assembly will generate different economics spectra for the plant II) Functional strategies and traits will show different relationships with demographic rates over the chronosequence; and III) Assembly processes will generate different patterns of phylogenetic and functional structure, which will be modified over time due to mortality and recruitment. To test these predictions, data on eight functional characteristics were collected (leaf, reproductive, stem and whole plant) in 64 species distributed in nine areas, six being secondary growth forests (three with approximately 17 years and three with approximately 25 years of abandonment), and three mature forests, located within the dense ombrophilous forest domain in the northern coastal region of Pernambuco, Brazil. In each one of these areas rates of mortality, recruitment and growth of the species and communities were obtained. The existence of an economics spectrum for plants was not observed, meaning that the organs acted independently, not varying over the gradient. Some traits had a strong influence on the plant demographics, but these relationships also did not vary with chronosequence. When species were grouped based on the largest numbers of traits favouring rapid resource use (acquisitive species), or resource conservation (conservative species) or traits adapted to both strategies (intermediate species), the demographic rates showed some variation over the chronosequence, with higher mortality in acquisitive and intermediate species in areas with 25 years abandonment and higher recruitment of intermediate and conservative species at the end of the chronosequence. Furthermore, it was observed that a phylogenetic signal did not exist for the traits studied, indicting that there was no niche conservation and that the phylogenetic and functional structure showed a random pattern in the initial stages of succession, whilst at the end of succession the community was functionally dispersed and phylogenetically grouped. These patterns were maintained over time, not influenced by individuals and dead and recruited species. Thus, it is possible to say that throughout this chronosequence, the variations in filters were perceived differently when analysed by populations or communities. Traits analysed at the population level did not reflect the influence of changes observed over the chronosequence, while for the community these influences were more noticeable. Therefore, the study of functional and dynamic traits, as well as species groups, provides important information about the processes, and consequently the assembly, of communities with different resource availabilities.