Inferência espacial para mapeamento de macrófitas submersas: estudo de caso
Carregando...
Arquivos
Data
2001-02-25
Autores
Rotta, Luiz Henrique da Silva [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Este trabalho relata um estudo de caso, no qual se procedeu a um levantamento ecobatimétrico para se obter dados de profundidade do corpo de água e da altura da vegetação aquática submersa, realizado num trecho do Rio Uberaba, que faz parte do Reservatório de Porto Colômbia, abrangendo uma área de aproximadamente 360 mil m². Os dados foram processados e avaliados com o objetivo principal de desenvolver uma abordagem de inferência espacial no mapeamento de macrófitas submersas presentes na região pesquisada. Para isso, verificou-se a viabilidade do uso de interpoladores convencionais para mapear as variáveis coletadas pelo ecobatímetro, avaliou-se o potencial de imagem multiespectral de alta resolução espacial (QuickBird) na detecção da vegetação estudada, realizou-se uma análise espacial para avaliar tendência para esse tipo de ocorrência e, finalmente, estudou-se a possibilidade de se utilizar técnica de geoestatística na inferência da altura de macrófitas submersas da área de estudo. As interpolações convencionais não se mostraram satisfatoriamente adequadas para mapear a variável altura da vegetação submersa, porém, a utilização da interpolação linear por triangulação foi suficiente para representar adequadamente o relevo submerso. A imagem multiespectral se mostrou capaz de detectar aproximadamente 21% de macrófitas submersas. Com relação à análise espacial, os resultados indicaram que a dependência espacial é uma tendência que deve ser levada em consideração, daí ter sido utilizada a geoestatística na abordagem de inferência espacial da vegetação. A análise dos semivariogramas forneceu conhecimento relacionado aos métodos de levantamentos ecobatimétricos. A inferência espacial por krigagem apresentou resultado satisfatório, comparado às interpolações convencionais mencionadas anteriormente...
In this work was accomplished a case study in which an echo sounder survey was conducted to obtain data of water body depth and submerged aquatic vegetation height. The study was realized specifically on a region of Uberaba River with approximately 360.000 m². This river is part of the Porto Colombia reservoir. The data was processed and evaluated with main objective to develop an approach of spatial inference for map submersed macrophyte in the region studied. For this, it was verified the availability of using conventional interpolation to map the variables collected by the echo sounder, it was evaluated the potential of multispectral image with high spatial resolution (QuickBird) to detect the vegetation, it was done a spatial analysis and finally it was verified the possibility of using geostatistic technique to infer the submerged macrophytes height in the study area. The use of linear interpolation by triangulation was enough to represent rightly the underwater topography; however, the conventional interpolations were not satisfactorily to map the submerged vegetation height. The multispectral image was able to detect approximately 21% of submerged macrophytes. About spatial analysis, the results indicated that the spatial dependence must be considered, therefore, it was used the geostatistic in the spatial inference of submerged vegetation. Semivariograms analysis provided important knowledge about methods of echo sounder survey. The spatial inference by kriging showed satisfactory results if compared with conventional interpolations mentioned above. Finally, the approach used to mapping submerged macrophytes height can be applied in other regions and this can help the reservoirs management, allowing improve the knowledge about the local ecosystem and therewith developing an action strategy to contain the spread of submerged aquatic vegetation
In this work was accomplished a case study in which an echo sounder survey was conducted to obtain data of water body depth and submerged aquatic vegetation height. The study was realized specifically on a region of Uberaba River with approximately 360.000 m². This river is part of the Porto Colombia reservoir. The data was processed and evaluated with main objective to develop an approach of spatial inference for map submersed macrophyte in the region studied. For this, it was verified the availability of using conventional interpolation to map the variables collected by the echo sounder, it was evaluated the potential of multispectral image with high spatial resolution (QuickBird) to detect the vegetation, it was done a spatial analysis and finally it was verified the possibility of using geostatistic technique to infer the submerged macrophytes height in the study area. The use of linear interpolation by triangulation was enough to represent rightly the underwater topography; however, the conventional interpolations were not satisfactorily to map the submerged vegetation height. The multispectral image was able to detect approximately 21% of submerged macrophytes. About spatial analysis, the results indicated that the spatial dependence must be considered, therefore, it was used the geostatistic in the spatial inference of submerged vegetation. Semivariograms analysis provided important knowledge about methods of echo sounder survey. The spatial inference by kriging showed satisfactory results if compared with conventional interpolations mentioned above. Finally, the approach used to mapping submerged macrophytes height can be applied in other regions and this can help the reservoirs management, allowing improve the knowledge about the local ecosystem and therewith developing an action strategy to contain the spread of submerged aquatic vegetation
Descrição
Palavras-chave
Cartografia, Geologia - Métodos estatísticos, Sensoriamento remoto, Levantamento ecobatimétrico, Echo sounder survey, Spatial analysis of submerged macrophytes, Geostatistic inference of submerged macrophytes, Remote sensing for submerged macrophytes mapping
Como citar
ROTTA, Luiz Henrique da Silva. Inferência espacial para mapeamento de macrófitas submersas: estudo de caso. 2011. 119 f. Dissertação (mestrado) - Universidade Estadual Paulista, Faculdade de Ciências e Tecnologia, 2011.