Accidental release of hazardous gases: modelling and assessing risk

The renewed concern in assessing risks and consequences from technological hazards in industrial and urban areas continues emphasizing the development of local-scale consequence analysis (CA) modelling tools able to predict shortterm pollution episodes and exposure effects on humans and the environment in case of accident with hazardous gases (hazmat). In this context, the main objective of this thesis is the development and validation of the EFfects of Released Hazardous gAses (EFRHA) model. This modelling tool is designed to simulate the outflow and atmospheric dispersion of heavy and passive hazmat gases in complex and build-up areas, and to estimate the exposure consequences of short-term pollution episodes in accordance to regulatory/safety threshold limits. Five main modules comprising up-to-date methods constitute the model: meteorological, terrain, source term, dispersion, and effects modules. Different initial physical states accident scenarios can be examined. Considered the main core of the developed tool, the dispersion module comprises a shallow layer modelling approach capable to account the main influence of obstacles during the hazmat gas dispersion phenomena. Model validation includes qualitative and quantitative analyses of main outputs by the comparison of modelled results against measurements and/or modelled databases. The preliminary analysis of meteorological and source term modules against modelled outputs from extensively validated models shows the consistent description of ambient conditions and the variation of the hazmat gas release. Dispersion is compared against measurements observations in obstructed and unobstructed areas for different release and dispersion scenarios. From the performance validation exercise, acceptable agreement was obtained, showing the reasonable numerical representation of measured features. In general, quality metrics are within or close to the acceptance limits recommended for ‘non-CFD models’, demonstrating its capability to reasonably predict hazmat gases accidental release and atmospheric dispersion in industrial and urban areas. EFRHA model was also applied to a particular case study, the Estarreja Chemical Complex (ECC), for a set of accidental release scenarios within a CA scope. The results show the magnitude of potential effects on the surrounding populated area and influence of the type of accident and the environment on the main outputs. Overall the present thesis shows that EFRHA model can be used as a straightforward tool to support CA studies in the scope of training and planning, but also, to support decision and emergency response in case of hazmat gases accidental release in industrial and built-up areas.

Alteração climática dos índices de estabilidade na Peninsula Iberica

Este trabalho tem como objetivo a comparação da intensidade, frequência e distribuição de um conjunto de índices de estabilidade atmosférica simulados entre o clima histórico (1986-2005) e um cenário climático (2081-2100) na Península Ibérica. Considerou-se o cenário de emissão de gases RCP8.5. Estes índices avaliam a instabilidade atmosférica que é um elemento fundamental e percursor no desenvolvimento de tempestades. Através dos seus valores limite, é possível estimar alterações na probabilidade de ocorrência de eventos extremos que se poderão desenvolver no clima futuro, relativamente ao histórico. Primeiro, utilizou-se um conjunto de simulações do WRF com dois forçamentos: reanálises do ERA-Interim e um modelo do Max Planck Institute. De seguida, foram calculados diferentes índices de estabilidade. A validação do modelo consistiu no cálculo das médias sazonais, da sua diferença e das respetivas PDFs dos índices simulados pelo WRF-MPI e WRF-ERA. Verifica-se uma sobrestimação do CAPE, SHR6km (vento de corte) e SWEAT simulados pelo WRF-MPI. No entanto, nos campos dos índices simulados pelos dois forçamentos para o período histórico, verifica-se que os padrões espaciais são semelhantes apesar das diferenças na intensidade. Como as alterações climáticas dos índices são avaliadas através de diferenças, estas discrepâncias não invalidam a utilização do modelo no futuro. Posteriormente foram estudadas as alterações climáticas dos índices através da comparação entre o clima histórico e futuro. Estima-se um aumento da intensidade do CAPE e uma diminuição (aumento) da frequência de eventos com intensidade reduzida (elevada). Estas alterações são robustas no verão e outono. Também é esperado um aumento da intensidade do SHR6km na primavera e inverno tal como da frequência de SHR6km elevado nestas estações e uma redução da intensidade e da frequência de eventos com SHR6km elevado nas restantes. Haverá um possível aumento robusto da intensidade do SWEAT no verão e outono, bem como da frequência destes valores. Concluindo, será provável um aumento da frequência dos ambientes favoráveis ao desenvolvimento de tempestades, devido a uma maior intensidade e probabilidade de ocorrência de valores extremos do CAPE e do SWEAT. No entanto, a redução do SHR6km, poderá diminuir a organização das tempestades e o seu tempo de vida.

Calibration and validation of hydrodynamic, salt and heat transport models for Ria de Aveiro lagoon (Portugal)

Ria deAveiro is a very complex shallow water coastal lagoon located on the northwest of Portugal. Important issues would be left unanswered without a good understanding of hydrodynamic and transport processes occurring in the lagoon. Calibration and validation of hydrodynamic, salt and heat transport models for Ria de Aveiro lagoon are presented. The calibration of the hydrodynamic model was performed adjusting the bottom friction coefficient, through the comparison between measured and predicted time series of sea surface elevation for 22 stations. Harmonic analysis was performed in order to evaluate the model's accuracy. To validate the hydrodynamic model measured and predicted SSE values were compared for 11 stations, as well as main flow direction velocities for 10 stations. The salt and heat transport models were calibrated comparing measured and predicted time series of salinity and water temperature for 7 stations, and the RMS of the difference between the series was determined. These models were validated comparing the model results with an independent field data set. The hydrodynamic and the salt and heat transport models for Ria de Aveiro were successfully calibrated and validated. They reproduce accurately the barotropic flows and can therefore adequately represent the salt and heat transport and the heat transfer processes occurring in Ria deAveiro.