Landscape structure analysis of basin Cuiabá, Petrópolis, Rio de Janeiro, Brazil
ISBN 978-85-88783-11-9
Authors
1Cintra, D.; 2Colman, T.; 3Uchôa, D.; 4Fernandes, M.
1UFRJ Email: danilabgis@gmail.com
2UFRJ Email: thiagocolman@gmail.com
3UFRJ Email: danuzzauchoa@gmail.com
4UFRJ Email: manoel.fernandes@ufrj.br
Abstract
The intense anthropogenic pressure on forest areas over time, beyond the natural dynamics of clearings itself, results in several changes of the landscape, such as hydrological and erosive dynamics, which may cause catastrophic events. In this sense, it becomes necessary to perform studies on the structure and distribution of forest remnants, aiming mainly conservation as one of the factors to minimize these events. The use of satellite imagery as well as the process of their classification was able to enhance the ability to monitor these areas. Besides that, it is necessary to have a good cartographic basis that needs to be systematically update. The present study is located in Vale do Cuiabá, Petrópolis, Rio de Janeiro, Brazil, and aims to analyze the landscape structure from the knowledge of the forest remnants characteristics after the occurrence of landslides in February 2011. In order to understand the quality of these forest remnants were created maps involving different variables such as slope, aspect, solar radiation and landform. Besides that, different vegetation indices were compared with the purpose of delimitate forest remnants with different characteristics: NDVI, NDVINIR2/Red-edge, NIRNDVI and YNDVI. The NDVI, NDVINIR2 and YNDVI indices could detect objects belonging to vegetation class with high spectral response, indicating a healthier vegetation, with thresholds between 0.4 and 0.7. The detection of forest fragments was clearer with the application of NDVI and YNDVI with thresholds ranging between 0.3 and 0.4 and 0.2 and 0.3, respectively. The NDVINIR2/Red-edge index showed a distinct threshold in the classification of vegetated areas, however presenting confusion between other classes, such as rock and dark roofs, as well as NIRNDVI. In this study, the NIR2 and yellow bands stood out, unlike the Red-edge band. Still, it should be emphasized that the application of vegetation indices is highly dependent upon the purpose of the research, and the region in which it is being applied. After that were selected six fragments to visit on the field and evaluate the vegetation structure. From this was confirmed that the forest remnants with better vegetation structure are mainly those present in the south facing slope with concave shapes and a mean to high slope. This occurs primarily due to increased moisture retention, because in the South Hemisphere, the North facing slopes receive more sunlight than facing the South which affects the forest development in various forms, including differences in the floristic composition.
Keywords
Forest Fragmentation; Spatial Analysis; Geoecology