Chronic exposure to ambient PM2.5 concentrations is one of the leading worldwide mortality risk factor, which has seen an increase over the last decade. Sensitivity analysis is an essential approach in attributing societal impacts to emissions sources in any geographic scale.Source attribution of the societal impacts of PM2.5 pollution offers great potential for informing policy development and implementations. Assessing the location-specific societal benefits of reducing emissions from a regional to hemispheric scale is the major motivation of this work because exposure to ambient PM2.5 concentration is not only a regional concern but also global. This thesis employs adjoint sensitivity analysis, integrating demographic, epidemiological, and economic data, in a full-complexity modeling approach to link the sources of emissions to societal impacts.This study uses the adjoint of U.S. EPA's Community Multiscale Air Quality (CMAQ) model (CMAQ-ADJ) to provide location-specific source attribution of the societal burden of PM2.5 pollution. CMAQ-ADJ model is extended from the regional to the hemispheric platform, including boundary transport of the societal impacts and updates to the chemical representation. The monetized health impacts of coal phase-out regulation in Ontario were retrospectively assessed, as well as those for planned phase-out across Canada. Backward boundary conditions in adjoint model were implemented for the first time for the estimation of the health burden over Canada.Our findings suggest that the coal phase-out had substantial, albeit lower than previously predicted, health benefits within the province, and that the choice of the epidemiological model has an impact on the estimated health benefits. We also evaluate the transboundary impact of US coal-fired electricity generation and its emission control measures over the same period, and find the benefits from U.S. emission reductions to be larger than those from Ontario coal phase-out.iii Using hemispheric CMAQ-ADJ, location-specific premature avoided deaths and monetized health impacts are attributed to the emissions over the Northern Hemisphere. Countryspecific value of statistical life (VSL) estimates are applied to evaluate benefit-per-ton at a hemispheric scale. The highest pollution-emitting countries experience larger hemispheric health burdens and the majority of these countries with larger health burdens are the countries with a larger proportion of low-income populations. Consistent with past studies we find China and India to carry the majority of the hemispheric burden. Further, sector-specific health impacts reveal that the benefits of emission control vary greatly from sector to sector, and region to region.Further exploring country-level inequalities in PM2.5 exposure, we employed hemispheric CMAQ adjoint to quantify inter-national air pollution health inequalities. A single SES parameter (country-level income) is used to estimate location-specific health inequalities for countries. As expected, low-income countries with high leve...