Fechar Janela

Integrating field data and remote sensing to study secondary forests in Amazonian rural settlements

Mateus Batistella, Embrapa Satellite Monitoring, mb@cnpm.embrapa.br (Presenting)
Dengsheng Lu, Indiana University/CIPEC, dlu@indiana.edu

Secondary forests in the Amazon gained importance when attention was called to processes following landscape disturbances, such as deforestation. Sharp distinctions between successional stages are often artificial, but sometimes useful to characterize selected landscapes and to estimate their role in carbon sequestration. Remote sensing and GIS have improved the capability to monitor processes of Land-Use/Land-Cover (LULC) change in the Amazon. In this paper, the results for vegetation structure in Rondônia are presented as a basis for discussing the reflectance of secondary forests when using Landsat TM images. Fieldwork was carried out during the dry seasons of 1999, 2000, 2002, and 2003. Vegetation structure data were collected through surveys encompassing land-cover classes such as initial secondary succession (SS1), intermediate secondary succession (SS2), advanced secondary succession (SS3), and mature forest. Every plot was registered with a Global Positioning System (GPS) device to allow further integration with remote sensing data. Variables analyzed included density, diameter at breast height, basal area, total height, and biomass. The results for vegetation structure analyses informed image classifications. Descriptive statistics, graphic outputs, and analysis of variance (ANOVA) were performed. The results showed that SS1, SS2, SS3, and forest were well separated when using solely the data for vegetation structure (p<0.001). However, analyses of reflectance on selected TM bands allowed the separation of only three of these classes (SS1 and SS2 mixed together, SS3, and forest). The authors are engaged in improving the performance of image classifications using more robust techniques, such as spectral mixture analyses and spatial-spectral classifiers. The implications of this kind of study surpasses the understanding of vegetation recovery processes at local scales. It allows the spatial-temporal monitoring of Amazonian landscapes regarding their land-cover dynamics, useful for integrative programs, such as Proambiente.

Submetido por Mateus Batistella em 18-MAR-2004

Tema Científico do LBA:  LC (Mudanças dos Usos da Terra e da Vegetação)

Sessão:  

Tipo de Apresentação:  Oral

ID do Resumo: 348

Fechar Janela