Modelling Health Impacts of Air Pollution and Their Valuation: an Application to Santiago, Chile

This chapter describes how estimates of air pollution emissions are transformed into ambient concentration of pollution, further translated into exposure and health impacts on a population, and finally expressed in damages. Different trade and environmental policy menus lead to different levels of pollution emissions and income. For each menu the emissions levels translate into specific mortality and morbidity impacts on the population. The impacts have different values because health risks and income levels are specific to the policy menu. The willingness to pay (WTP) to avoid or reduce air pollution increases with increasing wealth and pollution levels. We go through these processes—from emissions to valuation of mortality and morbidity impacts—for the case study of the Chilean economy described in Chapter 6, with a focus on air pollution in Santiago. The emissions at the national level coming from the economy-wide model are calibrated to reproduce existing data on emissions of air pollution in Santiago. Then, a simple dispersion model is constructed using information on pollution concentration and emission inventory for Santiago. Dispersion modelling allows us to link emissions of effluents to ambient concentration of the effluents in the air. Once concentrations are estimated, they are applied to dose-response functions, which express the changes in incidence of morbidity and mortality induced by changes in pollution concentrations. Morbidity focuses on specific nonfatal illnesses or symptoms caused by air pollution, whereas mortality focuses on changes in the risk of death. The air pollutants of interest are small particulates, sulphur dioxide (SO2), nitrogen dioxide (NO2), carbon monoxide (CO), ozone (O3), and lead. Finally, the health incidences are applied to Santiago’s population to obtain the health impact in Santiago corresponding to national emissions estimated by the model (number of deaths, and morbidity measures, such as number of cases of bronchitis).