A Synoptic Climatological Approach to Separate Weather- and Pollution-Induced Impacts on Human Mortality

Considerable research indicates that human mortality is significantly affected by both ambient meteorological conditions on a particular day, as well as atmospheric pollutant levels. However, it is difficult to segregate the impact of each of these factors, and to identify any possible synergistic effects.

The goal of this study is to describe a climatological procedure that facilitates the separation of pollution and meteorological effects on human mortality. Our research into the segregation of these impacts involves use of the synoptic climatological approach. Each day at a particular location is classified into one of several air masses, or weather types, based on ambient meteorological conditions. In this study, the Spatial Synoptic Classification (SSC) has been utilized, and a number of air mass types and transitional situations can then be ascribed to each day.

When analyzed synoptically, the heat-mortality relationship usually identifies one or two air masses that are associated with mean mortality statistically significantly greater than the mean. It is these oppressive air masses that serve as the basis for our division into “offensive” days and “non-offensive” days, based on meteorological conditions. Once divided, the pollutant-mortality relationship is then evaluated within each of these groups. Days are then classified by quintiles of suspended particulates (PM10, PM2.5, TSP) and ozone (O3) within each of these two groups.

Results vary across the several cities analyzed. In some cases, important differentials are noted. For example, on “oppressive” days, there is no correlation between pollutant levels and mortality; mortality remains high irrespective of pollution concentration. It is only during “non-oppressive” conditions that mean mortality increases along with pollution levels. In other locales, pollution seems to play some role during oppressive meteorological conditions.

This suggests that, while there is considerable interurban variability, in some cases the meteorological impact seems to overwhelm any pollution signal. These responses seem to be robust on an intraregional level.