Natural disaster-related hazards such as flooding, tsunamis and earthquakes have very visible impacts. The physical environment is demonstrably changed and the dead and injured can be seen and identified. The impacts of air pollution are usually quite different however. In developed countries today we rarely get short high-pollution events like those experienced in Meuse Valley in 1930, Donora in 1948, and London in 1952. Instead we get lower levels of pollution which we know have the potential to harm health. However, this means that illness and death is not immediate and not always easily attributable to specific temporal event. This fact makes the science highly contested, and is the basis for three of the papers in this issue of the Journal.