Name: Adrielle Luiza Vieira de Melo
Type: MSc dissertation
Publication date: 21/02/2020
Advisor:

Name	Role
Jane Meri Santos	Advisor *

Examining board:

Name	Role
Elisa Valentim Goulart	Co advisor *
Jane Meri Santos	Advisor *
Luis Fernando Alzuguir Azevedo	External Examiner *
Neyval Costa Reis Jr.	Internal Examiner *

Summary: Peak concentrations of pollutants can occur over a very short period of time due to the effects of atmospheric turbulence. However, these concentration peaks are not predicted in Gaussian models of atmospheric dispersion of pollutants that are normally based onaverages of 1 hour, 8 hours and 24 hours. To solve this limitation, the power law is commonly used, which relates maximum average concentration values in averagingtimesless than one hour (peak) and hourly average concentration values (mean), known as peak-to-meanratio (P/M). Although, the power law describes the P/M ratio as a function of the ratio between the respective high averagingtimes to an exponent p, has not been fully investigated, especially in urban environments WHERE buildings directly influence the flow and dispersion of pollutants. The pexponent can vary with distance, source, land occupation, meteorological conditions, among other factors. The main objective of this work is to evaluate the influence of the wind direction and the source position on the P/M ratio in urban scenarios (represented by a uniform building arrayand anotherwith a tall building embedded in a regular array). For this purpose, experimental wind tunnel data obtained fromscientific literature were used. High frequency concentration data monitored on several sensors with different heightand distance from the source were studied. In general, when the wind direction is perpendicular to the largest face of the obstacles (0°), p-values decrease with distance, with a more pronounced trendwhen the source is located in the region of circulation between two obstacles (position S1). For the wind directions of 45° (impingingon the edge of the obstacle) and 90° (impingingon the smallest face of the obstacle), the same trend was observed in the sensors located near the central line of the plume. In the same direction as the wind, sources located in the urban canyon produced different results compared to the source S1. The p-values increase with height mainly in all evaluated scenarios. The position of the source, wind direction, urban scenarioand sensor positionwere determinant in the values of turbulenceintensity, intermittency, percentile 99thand p-values. In addition, the scenario with tall building, in general, at half the height of uniform buildings presented p-values lower than the scenario of uniform height, however when evaluating p-values in the middle of the tall building thep-values are mostly higher in the scenario with the tall building.Keywords:atmospheric turbulence, peak-to-mean concentration, maximum concentration, averaging time, urban environment.