| The general problem of urban pollution and its relation to the city population is examined in this dissertation. A simple model suggests that pollutant concentrations should scale approximately with the square root of city population. This model and its experimental evaluation presented here serve as important guidelines for urban planning and attainment of air quality standards including the limits that air pollution places on city population.;The model was evaluated using measurements of air pollution. Optical properties of aerosol pollutants such as light absorption and scattering plus chemical species mass concentrations were measured with a photoacoustic spectrometer, a reciprocal nephelometer, and an aerosol mass spectrometer in Mexico City in the context of the multinational project "Megacity Initiative: Local And Global Research Observations (MILAGRO)" in March 2006. Aerosol light absorption and scattering measurements were also obtained for Reno and Las Vegas, NV USA in December 2008-March 2009 and January-February 2003, respectively. In all three cities, the morning scattering peak occurs a few hours later than the absorption peak due to the formation of secondary photochemically produced aerosols. In particular, for Mexico City we determined the fraction of photochemically generated secondary aerosols to be about 75% of total aerosol mass concentration at its peak near midday.;The simple 2-d box model suggests that commonly emitted primary air pollutant (e.g., black carbon) mass concentrations scale approximately as the square root of the urban population. This argument extends to the absorption coefficient, as it is approximately proportional to the black carbon mass concentration. Since urban secondary pollutants form through photochemical reactions involving primary precursors, in linear approximation their mass concentration also should scale with the square root of population. Therefore, the scattering coefficient, a proxy for particulate matter mass concentration, is also expected to scale the same way.;Experimental data for five cities: Mexico City, Mexico; Las Vegas and Reno, NV, USA; Beijing, China; and Delhi, India (the data for the last two cities were obtained from the literature); are in reasonable accord with the model. The scaling relation provided by the model may be considered a useful metric depending on the assumption that specific city conditions (such as latitude, altitude, local meteorological conditions, degree of industrialization, population density, number of cars per capita, city shape, etc.) vary randomly, independent of city size. While more detailed studies (including data from more cities) are needed, we believe that this relatively weak dependence of the pollution concentration on the city population might help to explain why the worsening of urban air quality does not directly lead to a decrease in the rate of growth in city population. |
