Análise integrada de elementos da paisagem da Serra do Tepequém por meio de dados de sensoriamento remoto e de campo

Abstract

The northern state of Roraima is characterized in that distinction from other regions flaps peculiar characteristics of landscape elements that compose it. In the surroundings of this region, the municipality of Amajarí stands out with the Tepequém saw presenting diverse fitopaisagem, high lands and vast scenic beauty. This study aims to analyze the relationship of the typological distribution of vegetation type and vegetation type x x relief pedology at the top of the Sierra, through the use of remote sensing data, geomorphological subdivision, soil classes and field data. To achieve these goals we used a picture of OLI sensor onboard Landsat 8, and performed to correct the interference of atmospheric constituents through the dark pixel subtraction method. Enhancement operations were performed, linear expansion of contrast and conversion to RGB color composite. Images related to 6R5G8B colorful composition and band 5, served as reference for the supervised classification by pixel classifier pixel maximum likelihood (MAXVER). In the training phase, were identified in the corresponding image pixels to each of the required phytophysionomic classes. To achieve this study was conducted fieldwork, chemical, physical and soil organic matter, which analyzed in an integrated manner resulted in a map of soil and vegetation type of the Sierra. Data processing was carried out in PCI Geomatica software version 10.2, Spring 5.2 and ArcGIS version 10. The soil analysis showed chemically poor soils, acids, with high content of aluminum saturation. In evaluating the textural class of profiles, a predominance of Franco-Clay-Sandy texture. The characterization of the geomorphology used for this study was a previous study database. With the development of this work was possible the structure of the vegetation type map with 3 vegetation classes, with areas of forest, wooded savanna and savanna graminosa. It was also designed the top of the soil map of Tepequém saw distributed in seven soil classes including Cambisol, Udorthent, Quartzipsamment, Plinthosol Petrico, Oxisol, Epiaquic Haplustult and red Argisoil. The integrated analysis of these products relief and soil x the top of the mountain vegetation presented important results, even a visual analysis of overlap of maps you can see significant relationship between landscape elements. In the statistical treatment of data between the relationships of the elements involved, we used the simple ratio of the observed frequency and expected frequency (Fo / Fe) obtained from the intersections of classes. We used the X² test (chi-square) and submitted the analysis to Yates. Statistical results confirm significant relationship between vegetation and vegetation x x Geomorphology Pedology, then determining the rejection of the hypothesis H0 (no relationship between landscape elements) and acceptance of the hypothesis H1 (there is a relationship between landscape elements). The forest area showed higher with Cambisols and deep valleys. The tree savannah showed affinity with Neossolos and slope while graminosa Savannah had a greater relationship with Gleysol and reliefs lined hhills and plains.