To solve traffic congestion in Bangkok, the Bangkok Mass Transit system (BTS) constructed an overhead rail system with 24 stations. The BTS train station, S2, in this study area covers Silom road and obstructs the air pollutant dispersion in a congestion area. The 1: 200 physical model of the buildings along Silom road with the train station, S2, was simulated in this research to determine the air pollutant dispersion in the train station area. A tracer gas (CO₂) was emitted from a simulated line source with emission rates of 0.383, 0.681, 1.293, 2.586, 5.177 and 10.77 mg/min to simulate actual pollutant emission rates. The CO₂ gas was sampled at 55 locations in the model. The Kriging method was used to interpolate the data in the study area. . Emission rates were used to make the difference between measured CO₂ in the model area and ambient CO₂ large enough to be differentiated. Regression Analysis was used to relate analytically the mass emission rate to the CO₂ concentration. The results indicate that the maximum CO concentrations exceed the 30 ppm Bangkok standard along the Southeast side of Silom Road at the passenger platform level. Drivers will acquire more harmful levels of CO than pedestrians at street level, especially near the Southwest end of the train station. NO₂ concentrations do not exceed the standard (0.17 ppm) at street level. The highest predicted VOC is 1.05 ppm. These results may be used in the future for numerical modeling study. / Master of Science

Made available in DSpace on 2014-08-20T14:25:46Z (GMT). No. of bitstreams: 1 dissertacao_marieli_sallet.pdf: 231023 bytes, checksum: f445526b62fbfcde40fb1bc5ca90923a (MD5) Previous issue date: 2007-02-23 / Currently, the search for analytical solutions for the dispersion problems is one of the main research subjects in the pollutant dispersion modeling. These solutions become important due to the intention to obtain dispersion models that generate reliable results in a small computational time, which are of great interest for regulatory air quality applications. Lagrangian particle models are an important and effective tool to simulate the atmospheric dispersion of airborne pollutants. These models are based on the Langevin equation, which is derived from the hypothesis that the velocity is given by the combination between a deterministic term and a stochastic term. In this work is presented a new Lagrangian particle model to simulate the pollutant dispersion in low wind speed conditions. During low wind speed, the diffusion of a pollutant in the planetary boundary layer (PBL) is indefinite and it has been observed that the plume is subject to a great deal of horizontal undulations, which are called plume meandering. The method proposed leads to a stochastic integral equation whose solution has been obtained through the Method of Successive Approximations or Picard s Iteration Method. The integral equation is written in terms of the real and imaginary parts of the complex function before performing the multiplication of the integrating factor, expressed by the Euler formula, inside and outside of the integral solution. To take account the meandering effect, the Frenkiel s Eulerian autocorrelation functions for low wind conditions is included naturally in the model. The new approach has been evaluated through the comparison with experimental data and other different dispersion models. Particularly, the results obtained by the model agree very well with the experimental data, indicating the model represents the dispersion process correctly in low wind speed conditions. It is also possible to verify that the new model results are better than ones obtained by the other models. The analytical feature of the technique and the natural inclusion of the Frenkiel s Eulerian autocorrelation function become the model more accurate than other models. / Atualmente, a busca por soluções analíticas para os problemas de dispersão é um dos principais assuntos de pesquisa na modelagem da dispersão de poluentes. Estas soluções tornam-se importantes devido à intenção de obter modelos de dispersão que geram resultados confiáveis em um tempo computacional pequeno, que são de grande interesse para aplicações no controle da qualidade do ar. Modelos de partícula Lagrangeano são uma ferramenta importante e eficaz para simular a dispersão atmosférica de poluentes do ar. Esses modelos são baseados na equação de Langevin, que é derivada da hipótese que a velocidade é dada por uma combinação entre um termo determinístico e um termo estocástico. Neste trabalho é apresentado um novo modelo de partícula Lagrangeano para simular a dispersão de poluentes em condições de velocidade de vento fraco. Durante a velocidade de vento fraco, a difusão de um poluente na Camada Limite Planetária (CLP) é indefinida e tem sido observado que a pluma está sujeita a grandes ondulações horizontais, que são chamadas meandro do vento. O método proposto leva a uma equação integral estocástica cuja solução é obtida através do Método das Aproximações Sucessivas ou Método Iterativo de Picard. A equação integral é escrita em termos das partes real e imaginária da função complexa antes de realizar a multiplicação do fator integrante, expresso pela fórmula de Euler, dentro e fora da solução integral. Para considerar o efeito do meandro, as funções de autocorrelação Euleriana de Frenkiel para condições de vento fraco são incluídas naturalmente no modelo. A nova aproximação foi avaliada através da comparação com dados experimentais e outros diferentes modelos de dispersão. Particularmente, os resultados obtidos pelo modelo concordam muito bem com os dados experimentais, indicando que o modelo representa o processo de dispersão corretamente em condições de velocidade de vento fraco. Também é possível verificar que os resultados do novo modelo são melhores do que os obtidos pelos outros modelos. A característica analítica da técnica e a inclusão natural da função de autocorrelação Euleriana de Frenkiel tornam o modelo mais exato que os outros modelos.