Integração de sensoriamento remoto

INTEGRAÇÃO DE SENSORIAMENTO REMOTO E SIG (GEOPROCESSAMENTO)
NA IDENTIFICAÇÃO DOS SOLOS PRINCIPAIS E ESTRATOS DE VEGETAÇÃO
PARA PLANEJAMENTO REGIONAL NO ESTADO DA PARAIBA

Abstract.
The main objective of this study is the better use of the natural resources for a part of Agreste region of the state of Paraiba in northeastern Brazil. Under this study, the classifications (unsupervised and supervised) were made for the interpretation of Landsat – TM Data, using ERDAS Imagine Software. The soils were classified into three major groups, such as, Luvissolos, Neossolos Litolicos and Argissolos. The Land Use and Land Cover classification was divided into four major classes, such as, Native Vegetation and Rock-outcrops, Native Vegetation,
Degraded areas and Agricultural areas. According to an average classification system, the overall classification accuracy was found approximately 86,00%. It reveals that accuracy of the classification was considered high and the results were very satisfactory. The area of each classes was calculated and the total area of digitally prepared map was approximately 629Km
2
. The classes of the system were spectrally homogeneous. The three principal land limitations encountered in the study area are: lack of water, surface rockiness and stoniness and susceptibility of erosion. It was concluded from the study that the Landsat-TM images are more effective for the detection of major soil groups, land evaluation and land use/land cover classes for the detailed regional and local planning, land development and land management for the Agreste region of the state of Paraíba. Also, such type of technology used under this study, may be used for planning, management and development of any type of climatic regions, such as
Humid, Sub-humid, Agreste, Semi-arid, Arid and Pantanal.
Key Words: Landsat-TM, Agreste Region, Image Processing, Land Use/Land Cover, Landsat-TM, Região