Full text
| |
| Author(s): |
Catia Urbanetz
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Campinas, SP. |
| Institution: | Universidade Estadual de Campinas (UNICAMP). Instituto de Biologia |
| Defense date: | 2010-04-29 |
| Examining board members: |
Luiza Sumiko Kinoshita;
George John Shepherd;
Ivan Schiavini da Silva;
Natalia Macedo Ivanauskas;
Ricardo Ribeiro Rodrigues
|
| Advisor: | Luiza Sumiko Kinoshita; Fernando Roberto Martins |
| Abstract | |
The aim of this study was to investigate the influence of soil nutrients and microrelief on the variation in the spatial structure of the Atlantic Dense Ombrophilous Forest vegetation. Three blocks of 40 plots were established in a steep slope forest, a clay soil, at three altitudinal levels. Three other blocks of 20 plots were established in a lowland forest, in a poor sandy soil. Area "D" is 1,5 km from the steep slope, and areas "E" and "F" at further 1,5 km intervals in the direction of the coast. All the trees individual with PBH ? 15 cm were collected. The total area sampled was 1.8 ha. Classification and ordination analysis were applied to an abundance matrix in order to analyse the similarities among samples, patch structure or gradients. Floristic dissimilarities among plots (Bray-Curtis index computed with species abundance in each plot) were correlated with environmental distances among plots (Euclidian distance index computed with nutrients soil data, altitude and declivity in each plot) and geographic distances among plots (Euclidian distance index) through Mantel and partial Mantel tests. Canonical Correspondence Analysis (CCA) was applied to species abundance and abiotic data to detect species-environment relations. It was found 3415 individuals of 201 species and 55 families. A clear separation among plots of the steep slope forest and the lowland forest was found, although, there were species common to both vegetation types. The CCA and partial Mantel test results suggested that part of this spatial structure differentiation seems to be related to sand percentage, Magnesium and pH of the soil, declivity and altitude. Nevertheless, a large part of the inertia could not be explained by the abiotic data. The species common to both areas probably have a broad ecological plasticity and can tolerate the extreme condictions of poor nutrients avaiability and flooding of the lowland. The lowland forest flora was not uniform and two communities were recognised in that area. The CCA results indicate that these could be related with the texture gradient in the lowland soil. The floristic composition and the structure of the steep slope forest was more regular than the lowland forest. There were no correlation among species abundance and abiotic factors when geographic space was taken into account. Possibly stochastic events such as dispersion and human impacts or other unmeasured factors could determine the spatial structure found, together with to unexplained inertia. Ecological determinism seems to play a more important role at the landscape scale than at the local scale in this study. Studies focusing other aspects such as dispersion, competition or other unmeasured abiotic data could be helpful in clarifying the main causes of the unexplained patterns found in this study. (AU) | |