Metodologia para avaliação ambiental em bacias hidrográficas, utilizando técnicas de geoprocessamento e sensoriamento remoto
Carregando...
Arquivos
Data
2017-08-18
Autores
Souza, José Carlos de [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A espacialização de dados ambientais através de técnicas de geoprocessamento e o monitoramento do comportamento espectral da vegetação por meio de imagens orbitais são instrumentos importantes nos estudos ambientais de análise da paisagem, em especial os aplicados a bacias hidrográficas. O objetivo da presente pesquisa foi desenvolver uma metodologia de avaliação ambiental aplicada à bacia hidrográfica, através da estimativa de parâmetros biofísicos, gerados por imagens Landsat 8 e determinação de parâmetros físico-químicos dos solos. O estudo foi aplicado na Bacia Hidrográfica do Rio Una, localizada em Ibiúna, no Estado de São Paulo, Brasil. Foram estimados o índice de vegetação por diferença normalizada – NDVI; o índice de vegetação ajustado do solo – SAVI; o índice de água por diferença normalizada – NDWI e o índice de ressecamento da vegetação por temperatura – TVDI. O estudo foi desenvolvido utilizando imagens do satélite Landsat 8 datadas de janeiro (período úmido) e agosto (período seco) do ano de 2015. O processamento digital das imagens foi realizado no software ArcGis 10.3. Foram geradas matrizes de correlação com base na combinação pixel a pixel para os índices NDVI, SAVI e NDWI utilizando o software Matlab para a análise espacial e temporal da cobertura vegetal. Além disso, foram coletados amostras de solo para determinação dos teores de matéria orgânica e carbono orgânico total – COT e levantamento in situ, dos teores de umidade e pH da área de estudo. Os resultados levantados foram interpolados pelo método do inverso do quadrado da distância - IQD. Os índices estimados apresentaram padrão espacial com forte relação com a sazonalidade climática, com ênfase ao regime de chuvas. Os resultados obtidos mostraram uma redução dos índices NDVI e SAVI, em 87,5% e 92% respectivamente, do período úmido para o seco. No NDWI foi identificado perda em conteúdo de água na vegetação em 73,03% dos pixels. As reduções nos valores dos pixels, nos três índices analisados, evidenciam a influência direta das condições climáticas no vigor e no teor de umidade da vegetação. O TVDI também apresentou influencia da sazonalidade climática. No mês de janeiro predominaram valores entre 0.2 e 0.4, indicando umidade normal, e em agosto predominaram valores entre 0.4 e 0.8, indicando condições ligeiramente seca ou estresse hídrico moderado. O TVDI se mostrou adequado para avaliar o déficit hídrico em diferentes coberturas do solo, com potencial para monitorar as perdas de umidade do solo e da vegetação. A umidade do solo, com medição in loco, registrou valores maiores nos meses com maior total de chuva, em todas as amostras, com exceção de algumas áreas agrícolas que ocorriam irrigação por aspersão, no período seco. Os solos apresentaram fraca acidez em todas as amostras, registrando pequena variação do pH entre os períodos, indicando condições de menor acidez no período de estiagem. Os teores de COT e matéria orgânica dos solos, indicaram as áreas de pastagens com teores mais elevados e as culturas temporárias com teores mais reduzidos, em relação aos demais tipos de cobertura do solo da bacia. A prática de cultivo e o processo de intemperismo explicam a redução nos teores de TOC e MO nos solos de culturas temporárias. Já as pastagens com suas estruturas radiculares, explicam a potencialização no armazenamento do carbono. As imagens Landsat 8 se mostraram eficientes na estimativa de parâmetros biofísicos da vegetação e as técnicas de geoprocessamento contribuíram substancialmente nas análises espaciais, se apresentando como ferramentas indispensáveis em estudos ambientais aplicados a bacias hidrográficas.
The spatialization of environmental data through geoprocessing techniques and the monitoring of vegetation spectral behavior through orbital images are important tools in environmental studies of landscape analysis, in special those applied to watershad. The objective of the present research was to develop an environmental assessment methodology applied to the watershad, through the estimation of biophysical parameters, generated by Landsat 8 images and determination of chemical-physical parameters of the soils. The study was applied in the Una Watershad, located in Ibiúna, in the State of São Paulo, Brazil. Were estimated the Normalized Difference Vegetation Index - NDVI; The Soil Adjusted Vegetation Index - SAVI; Normalized Difference Water Index - NDWI and the Temperature Vegetation Dryness Index - TVDI. The study was developed using Landsat 8 satellite images dated January (wet period) and August (dry period) of the year 2015. The digital process of images was performed in ArcGis 10.3 software. Correlation matrices were generated based on the pixel by pixel combination for the NDVI, SAVI and NDWI indexes using the Matlab software for the spatial and temporal analysis of the vegetation cover. In addition, soil samples were collected to determine the organic matter and total organic carbon (TOC) and survey in loco of soil moisture content and pH of the study area. The results obtained were interpolated by the Inverse Distance Weighting - IDW. The estimated indices presented a spatial pattern with strong relation with climatic seasonality, with emphasis on the rainfall regime. The results showed a reduction of the NDVI and SAVI indices, in 87.5% and 92%, respectively, from the wet to the dry period. In the NDWI, loss of water content in the vegetation was detected in 73.03% of the pixels. The reductions in the values of the pixels, in the three indices analyzed, show the direct influence of the climatic conditions on the vigor and the moisture content of the vegetation. TVDI also had an influence of climatic seasonality. In January values between 0.2 and 0.4 predominated, indicating normal humidity, and in August, values between 0.4 and 0.8 were predominant, indicating conditions slightly dry or moderate water stress. The TVDI proved to be suitable to evaluate the water deficit in different soil coverages, with potential to monitor soil and vegetation moisture losses. Soil moisture, with in loco measurement, recorded higher values in the months with the highest total rainfall, in all samples, except for some agricultural areas that were irrigated by sprinkling in the dry period. Soils presented low acidity in all samples, registering a small pH variation between the periods, indicating conditions of lower acidity in the dry season. The TOC and organic matter contents of the soils indicated the pasture areas with higher contents and the temporary crops with lower levels, in relation to the other types of soil cover in the watershed. The cultivation practice and the weathering process explain the reduction in TOC and OM contents in soils of temporary crops. On the other hand, pastures with their root structures explain the potential of carbon storage. Landsat 8 images were efficient in estimating biophysical parameters of the vegetation and geoprocessing techniques contributed substantially to spatial analysis, presenting themselves as indispensable tools in environmental studies applied to watersheds.
The spatialization of environmental data through geoprocessing techniques and the monitoring of vegetation spectral behavior through orbital images are important tools in environmental studies of landscape analysis, in special those applied to watershad. The objective of the present research was to develop an environmental assessment methodology applied to the watershad, through the estimation of biophysical parameters, generated by Landsat 8 images and determination of chemical-physical parameters of the soils. The study was applied in the Una Watershad, located in Ibiúna, in the State of São Paulo, Brazil. Were estimated the Normalized Difference Vegetation Index - NDVI; The Soil Adjusted Vegetation Index - SAVI; Normalized Difference Water Index - NDWI and the Temperature Vegetation Dryness Index - TVDI. The study was developed using Landsat 8 satellite images dated January (wet period) and August (dry period) of the year 2015. The digital process of images was performed in ArcGis 10.3 software. Correlation matrices were generated based on the pixel by pixel combination for the NDVI, SAVI and NDWI indexes using the Matlab software for the spatial and temporal analysis of the vegetation cover. In addition, soil samples were collected to determine the organic matter and total organic carbon (TOC) and survey in loco of soil moisture content and pH of the study area. The results obtained were interpolated by the Inverse Distance Weighting - IDW. The estimated indices presented a spatial pattern with strong relation with climatic seasonality, with emphasis on the rainfall regime. The results showed a reduction of the NDVI and SAVI indices, in 87.5% and 92%, respectively, from the wet to the dry period. In the NDWI, loss of water content in the vegetation was detected in 73.03% of the pixels. The reductions in the values of the pixels, in the three indices analyzed, show the direct influence of the climatic conditions on the vigor and the moisture content of the vegetation. TVDI also had an influence of climatic seasonality. In January values between 0.2 and 0.4 predominated, indicating normal humidity, and in August, values between 0.4 and 0.8 were predominant, indicating conditions slightly dry or moderate water stress. The TVDI proved to be suitable to evaluate the water deficit in different soil coverages, with potential to monitor soil and vegetation moisture losses. Soil moisture, with in loco measurement, recorded higher values in the months with the highest total rainfall, in all samples, except for some agricultural areas that were irrigated by sprinkling in the dry period. Soils presented low acidity in all samples, registering a small pH variation between the periods, indicating conditions of lower acidity in the dry season. The TOC and organic matter contents of the soils indicated the pasture areas with higher contents and the temporary crops with lower levels, in relation to the other types of soil cover in the watershed. The cultivation practice and the weathering process explain the reduction in TOC and OM contents in soils of temporary crops. On the other hand, pastures with their root structures explain the potential of carbon storage. Landsat 8 images were efficient in estimating biophysical parameters of the vegetation and geoprocessing techniques contributed substantially to spatial analysis, presenting themselves as indispensable tools in environmental studies applied to watersheds.
Descrição
Palavras-chave
Índices de vegetação, Paisagem, Uso do solo e cobertura vegetal, Sazonalidade climática, SIG, Vegetation indexes, Landscape, Soil use and vegetation cover, Climatic seasonality, GIS