Estudo da camada limite atmosférica em regiões metropolitanas costeiras com simulações de brisa marítima

• Autor(a) principal: Salvador, Nadir
• Data de Publicação: 2014
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal do Espírito Santo (riUfes)
• Texto Completo: http://repositorio.ufes.br/handle/10/1221
• topic: Sea-breeze; Local wind circulations; Brise marine; Circulations locales; WRF; Modelo WRF; Circulações locais; Engenharia Sanitária; Previsão numérica do tempo (Meteorologia); Brisa marítima; Circulação atmosférica; Camada limite planetária; 628

The main objective of this work was to identify and characterize the daily evolution of the Atmospheric Boundary Layer in the Great Region of Vitória (RGV), state of Espírito Santo, Brazil, and the Region of Dunkerque (RD), department of Nord Pas-de-Calais, France, evaluating the accuracy of parameterizations used in Weather Research and Forecasting (WRF) model to detect the formation and attributes of Thermal Internal Boundary Layer (CLI) formed by sea breezes. The RGV has complex relief in a coastal region of rugged topography and a chain of mountains parallel to the coast. The RD has a simple relief in a coastal region with small peaks not higher than 150 meters, all along the domain of study. To evaluate the results of the predictions made by the model, the results of two campaigns were used: one held in Dunkerque-FR, in July 2009, using a light detection and ranging (LIDAR) system, sonic detection and ranging (SODAR) and a surface meteorological station (EMS) data; another one held in Vitória-BR, in July 2012, also using a LIDAR, SODAR and EMS data. The simulations were performed using three PBL parameterizations schemes, two nonlocal closure, Yonsei University (YSU) and Asymmetric Convective Model 2 (ACM2), and a local closure, Mellor Yamada Janjic (MYJ) and two land surface schemes (CLS), Rapid Update Cycle (RUC) and Noah. As per the RGV as for RD, simulations with the six possible combinations were made for the periods in which the campaigns were made, using four nested domains, with the three largest square with 1863 km, 891 km and 297 km of side dimensions, grids 27 km, 9 km and 3 km, respectively, and the study domain, with dimensions 81 km in North-South direction and 63 km in the East-West grid 1 km, with 55 vertical levels up to approximately 13,400 m, more concentrated near the ground. The results of this study showed that: a) depending on the configuration adopted, the computational effort may increase too, though without a large increase in the accuracy of the results; b) for the RD, the simulation using the MYJ and Noah parameterizations produced the best estimation for CLI. Simulations using the ACM2 and YSU parameterizations, inferred the sea breeze entry with a maximum delay of three hours; c) for the RGV, the simulation that used the YSU and Noah parameterization made the best inferences about the CLI. The results show that it is necessary to evaluate in advance the computational effort required for certain settings and the accuracy of specific sets of parameterizations for each region. The differences are associated with the ability of different parameterizations capturing surface data from global, essential information for determining the intensity of vertical turbulent mixing and surface soil temperature, suggesting that a better representation of land use is crucial to improve estimations of the CLI and other parameters as input in models of dispersion of air pollutants.