Re-parameterization of a quasi analytical algorithm and phycocyanin estimation for a tropical reservoir

• Ano de defesa: 2014
• Autor(a) principal: Igor Ogashawara
• Orientador(a): José Luiz Stech, José Galizia Tundisi
• Banca de defesa: Deepak Ranjan Mishra
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Instituto Nacional de Pesquisas Espaciais (INPE)
• Programa de Pós-Graduação: Programa de Pós-Graduação do INPE em Sensoriamento Remoto
• País: BR
• Link de acesso: http://urlib.net/sid.inpe.br/mtc-m21b/2014/02.13.12.56

The eutrophication of aquatic systems is a worldwide environmental problem. One of its main outcome is the algal bloom, including the potentially toxic algal which can affect human health. Because of the toxicity of the harmful algal blooms, environmental monitoring is needed, mainly in aquatic systems near to urban centers. The use of remote sensing for monitoring the algal blooms uses bio-optical modeling, which is based on the spectral behavior of the optically active components in the water to estimate their concentrations as well as their inherent optical properties (IOPs). The detection of cyanobacteria, one of the main phylum of harmful algal, occurs by the identification of a unique pigment in inland waters cyanobacteria, the phycocyanin (PC). Remote sensing techniques, such as the quasi-analytical algorithms (QAA) - a type of bio-optical model - have been used to the estimation of IOPs in aquatic systems using in situ hyperspectral data and satellite multispectral data. However there is not any QAA developed or evaluated in tropical inland waters. Therefore the goal of this research was to evaluate the need for the re-parameterization of a QAA and to re-parameterize one for tropical eutrophic inland waters. Radiometric, limnological and IOPs data were collected in the Funil Hydroeletric Reservoir, located between São Paulo and Rio de Janeiro States, Brazil. Results of the Normalized Root Mean Square Error (NRMSE) showed that for the application of QAA in tropical inland waters a re-parameterization is needed. Thus, the results of the re-parameterization showed an average NRMSE of 36\% for the retrieval of the total absorption coefficients. The colored detrital matter (CDM) absorption coefficients were retrieved with an average NRMSE of 49\%. Phytoplankton absorption coefficients were retrieved with an average NRMSE of 74\%. PC concentration estimation from the estimated IOPs showed good results (NRMSE of 24.94\%) for the in situ hyperspectral dataset. Uncertainties in the estimations are mainly due to the lack of in situ data of PC absorption coefficients to calibrate the model. The re-parameterization was also applied for a synthetic dataset of the future Ocean \& Land Color Imager (OLCI) sensor which will be part of Sentinel 3 satellite. The simulation of OLCI data was conducted using its spectral response function and it is enhanced because of its potentially use of environmental monitoring since its temporal resolution will be improved by the launch of 2 satellites working as a constellation. Overall results were encouraging since it is one of the first works to explore the estimation of IOPs in tropical inland waters through remote sensing. However, results also indicated the need for further fine tuning of the model, mainly in the estimation of total absorption coefficients. Therefore, the development of a QAA and the estimation of PC concentration in tropical inland waters are an important step for the development of a robust tool for improving water quality monitoring.