Remote sensing applied to the Monitoring of the Largest Artificial Lake in Europe (Alqueva reservoir)

Abstract

This doctoral thesis demonstrates the potential of satellite remote sensing to monitor both water quality and atmospheric phenomena over the Alqueva reservoir, highlighting its relevance in supporting adaptive water management in regions vulnerable to climate change. The work focuses on assessing the spatial and temporal dynamics of key water quality parameters using multi-sensor satellite data from ESA (MERIS, OLCI, MSI) and NASA (MODIS), spanning two decades. To ensure reliable estimates of water surface reflectances, a robust atmospheric correction method was implemented and validated. Essential biophysical indicators — such as chlorophyll-a, phycocyanin concentrations, and water turbidity — were analysed across different regions of the reservoir. Empirical algorithms were developed and validated using data from MSI and OLCI sensors. MSI proved particularly effective for nearshore and small-scale analysis due to its good spatial resolution. However, since Alqueva shows limited spatial variability in water quality throughout most of the year, the higher spectral resolution of MERIS and OLCI was more advantageous for long-term trend monitoring and for assessing the risk of microalgae presence. These sensors enabled the detection of phycocyanin — a key pigment in cyanobacteria — and supported the development of a robust Optical Water Type (OWT) database, useful for water quality alert systems. In parallel, this thesis presents an original contribution to the monitoring and analysis of fog events over the Alqueva reservoir, based on geostationary satellite data. Observations from the SEVIRI sensor onboard the Meteosat satellites, combined with cloud classification products and multispectral approaches, enabled the characterisation of fog formation, evolution and dissipation, as well as an improved distinction between fog and low-level clouds. This approach highlights the potential of geostationary remote sensing to support nowcasting systems and to monitor surface atmospheric processes in aquatic environments. Overall, the results indicate that the Alqueva reservoir exhibits good water quality during most of the year, characterised by low surface reflectance values, with less favourable conditions observed mainly in the northern sector during summer and early autumn. The thesis thus demonstrates the usefulness of integrating multi-sensor and multi-scale data for the characterisation of complex This doctoral thesis demonstrates the potential of satellite remote sensing to monitor both water quality and atmospheric phenomena over the Alqueva reservoir, highlighting its relevance in supporting adaptive water management in regions vulnerable to climate change. The work focuses on assessing the spatial and temporal dynamics of key water quality parameters using multi-sensor satellite data from ESA (MERIS, OLCI, MSI) and NASA (MODIS), spanning two decades. To ensure reliable estimates of water surface reflectances, a robust atmospheric correction method was implemented and validated. Essential biophysical indicators — such as chlorophyll-a, phycocyanin concentrations, and water turbidity — were analysed across different regions of the reservoir. Empirical algorithms were developed and validated using data from MSI and OLCI sensors. MSI proved particularly effective for nearshore and small-scale analysis due to its good spatial resolution. However, since Alqueva shows limited spatial variability in water quality throughout most of the year, the higher spectral resolution of MERIS and OLCI was more advantageous for long-term trend monitoring and for assessing the risk of microalgae presence. These sensors enabled the detection of phycocyanin — a key pigment in cyanobacteria — and supported the development of a robust Optical Water Type (OWT) database, useful for water quality alert systems. In parallel, this thesis presents an original contribution to the monitoring and analysis of fog events over the Alqueva reservoir, based on geostationary satellite data. Observations from the SEVIRI sensor onboard the Meteosat satellites, combined with cloud classification products and multispectral approaches, enabled the characterisation of fog formation, evolution and dissipation, as well as an improved distinction between fog and low-level clouds. This approach highlights the potential of geostationary remote sensing to support nowcasting systems and to monitor surface atmospheric processes in aquatic environments. Overall, the results indicate that the Alqueva reservoir exhibits good water quality during most of the year, characterised by low surface reflectance values, with less favourable conditions observed mainly in the northern sector during summer and early autumn. The thesis thus demonstrates the usefulness of integrating multi-sensor and multi-scale data for the characterisation of complex aquatic systems, contributing to a better understanding of water quality variability and associated atmospheric phenomena - Resumo: Deteção Remota Aplicada à Monitorização do Maior Lago Artificial da Europa (Reservatório de Alqueva) Esta tese de doutoramento demonstra o potencial da deteção remota por satélite para monitorizar tanto a qualidade da água como fenómenos atmosféricos sobre a albufeira de Alqueva, realçando a sua relevância no apoio à gestão adaptativa da água em regiões sensíveis às alterações climáticas. O trabalho centra-se na avaliação das dinâmicas espaciais e temporais de parâmetros-chave da qualidade da água, recorrendo a dados multissensor provenientes da ESA (MERIS, OLCI, MSI) e da NASA (MODIS), ao longo de duas décadas. Para garantir estimativas fiáveis de refletância à superfície na água, foi aplicado e validado um método robusto de correção atmosférica. Indicadores biofísicos essenciais — como clorofila-a, concentrações do pigmento ficocianina e turbidez da água foram analisados em várias regiões da albufeira. Foram desenvolvidos e validados algoritmos empíricos com dados dos sensores MSI e OLCI, sendo o MSI particularmente eficaz para análises junto à margem e em pequenas escalas devido à sua excelente resolução espacial. No entanto, dado que Alqueva apresenta, na maior parte do ano, uma variabilidade espacial limitada da qualidade da água, a maior resolução espectral dos sensores MERIS e OLCI revelou-se mais vantajosa para o acompanhamento de tendências de longo prazo e avaliação do risco de presença de microalgas. Estes sensores permitiram detetar a presença de ficocianina — pigmento característico de cianobactérias — e suportaram a construção de uma base de dados robusta de Tipos Ópticos de Água, com utilidade em sistemas de alerta. Paralelamente, a tese apresenta um contributo original na monitorização e análise de eventos de nevoeiro sobre a albufeira de Alqueva, recorrendo a dados de satélite geoestacionário. A utilização de observações do sensor SEVIRI, a bordo dos satélites Meteosat, em combinação com produtos de classificação de nuvens, permitiu caracterizar a formação, evolução e dissipação do nevoeiro, bem como melhorar a sua distinção face a nuvens baixas. Esta abordagem evidencia o potencial da deteção remota geoestacionária para apoiar sistemas de nowcasting e para a monitorização de processos atmosféricos de superfície em ambientes aquáticos. De forma global, os resultados indicam que a albufeira de Alqueva apresenta, na maior parte do ano, boa qualidade da água, caracterizada por baixas refletâncias à superfície, observando-se condições menos favoráveis sobretudo na zona Norte, no verão e início do outono. A tese demonstra assim a utilidade da integração de dados multissensor e multiescala na caracterização de sistemas aquáticos complexos, contribuindo para uma melhor compreensão da variabilidade da qualidade da água e dos fenómenos atmosféricos associados.