The abatement of ground-level ozone has been a priority of air pollution policy because of its harmful effects on human health, ecosystems, and climate. The responsiveness of ozone to emissions of its principal precursors, nitrogen oxides (NOx) and volatile organic compounds (VOCs), is known to depend nonlinearly on spatially and temporally variable factors. Given this variability, scientific understanding of ozone formation processes can facilitate the development of sensible control policies. This thesis applies a high-order sensitivity analysis technique, the Decoupled Direct Method in Three Dimensions (HDDM-3D), to examine ozone response to precursor emissions during summertime air pollution episodes in the southeastern United States. HDDM-3D is shown to accurately capture ozone response within an underlying air quality model, even over large ranges of emission perturbations. Nonlinearity of response is quantified, and nonlinear terms are applied to examine how estimates of sensitivity and source attribution respond to uncertainty in an emissions inventory. Ozone production regime is assessed using both HDDM-3D and species indicator ratios and found to be primarily NOx-limited outside urban centers. However, ozone response to region-wide emissions does not necessarily correspond to its sensitivity to local controls, hindering the usefulness of bipartite ozone regime classification. Significant heterogeneity of ozone response to NOx is found even over small spatial scales of emission origin, a potential complication often ignored in atmospheric modeling and emissions trading mechanisms. Atmospheric sensitivity analysis is linked with a comprehensive menu of potential control measures to demonstrate potential integration of scientific and economic considerations for control strategy formulation. Cost-optimized strategies are identified for attainment of federal ozone standards in Macon, Georgia, and for minimizing potential population exposure to unhealthful concentrations of ozone.

The topography of South Africa (SA) shows complex variations and is one the main factors that determine the daily weather patterns and climate characteristics. It affects for example temperature, winds and rainfall (intensity and distribution). Mesoscale numerical weather prediction (NWP) models are used to simulate atmospheric motions with high horizontal grid resolution using appropriate cumulus parameterisation schemes. They also allow users to investigate the effects of topography and surface heating on the development of convective systems. The Weather Research and Forecasting (WRF) model was applied over the complex terrain of SA to simulate extreme weather events and evaluate the influence of topography and grid resolution on the accuracy of weather simulations. This includes heavy precipitation event that lead to floods over Limpopo region of SA which was caused by the tropical depression Dando for the period 16 -18 January 2012; the heat wave events over Limpopo region for the period 22-26 October 2011 and also over Cape region for the period 15-18 January 2012. The Grell-Devenyi Ensemble (GDE) cumulus parameterization scheme was applied. The WRF model was run at a horizontal resolution of 9 km with 3 km nests, one over Limpopo and another over Cape region respectively. A total of 210 South African Weather Service (SAWS) synoptic stations data were used to verify the model, with 37 stations located over Limpopo and 88 over Cape region. The WRF model simulations are able to capture the spatial and temporal distribution of the heat wave over Limpopo and Cape regions respectively. The model verification with observational data showed that the performance statistics are in the expected range. The experiments without topography give unrealistic verification scores. The increase of model grid resolution from 9 to 3 km improved the spatial and temporal distribution and performance statistics. The above findings are in general similar for the two heat wave events, although the influence of topography over Cape region is not too pronounced. This can be attributed to different topographic variations over the Cape region as compared to the Limpopo region. The WRF model captured well the spatial and temporal distribution of rainfall patterns; verification statistics shows over-prediction of its intensity in simulation with topography. The simulation without topography shows unrealistic space and intensity of rain distribution. An increase in model grid resolution from 9 to 3 km shows improved spatial and temporal distribution of rainfall. The importance of high grid resolution and the use of non-hydrostatic equations are confirmed by the analysis of the vertical velocity distribution and moisture fluxes. The overall findings proved that topography plays a major role to weather and climate over SA. The high grid resolution allows for a better topography representation and capturing convective activities by the use of nonhydrostatic approximations. Therefore the WRF model proved to be useful forecasting tool for weather and climate simulations and can be used for operational weather forecasting over South Africa. / Dissertation (MSc)--University of Pretoria, 2014. / lk2014 / Geography, Geoinformatics and Meteorology / MSc / Unrestricted

Health impact assessments for fine particulate matter (PM2.5) often rely on simulated concentrations generated from air quality models. However, at the global level, these models often run at coarse resolutions, resulting in underestimates of peak concentrations in populated areas. This study aims to quantitatively examine the influence of model resolution on the estimates of mortality attributable to PM2.5 and its species in the USA. We use GEOS-Chem, a global 3-D model of atmospheric composition, to simulate the 2008 annual average concentrations of PM2.5 and its six species over North America. The model was run at a fine resolution of 0.5 × 0.66° and a coarse resolution of 2 × 2.5°, and mortality was calculated using output at the two resolutions. Using the fine-modeled concentrations, we estimate that 142,000 PM2.5-related deaths occurred in the USA in 2008, and the coarse resolution produces a national mortality estimate that is 8 % lower than the fine-model estimate. Our spatial analysis of mortality shows that coarse resolutions tend to substantially underestimate mortality in large urban centers. We also re-grid the fine-modeled concentrations to several coarser resolutions and repeat mortality calculation at these resolutions. We found that model resolution tends to have the greatest influence on mortality estimates associated with primary species and the least impact on dust-related mortality. Our findings provide evidence of possible biases in quantitative PM2.5 health impact assessments in applications of global atmospheric models at coarse spatial resolutions.