Estimating particle exposure in the Mexico City metropolitan area

Abstract

Study question: We examined whether methods for measuring exposure to airborne particles less than 10 m in aerodynamic diameter ( PM 10 ) in the Mexico City metropolitan area give different estimates of PM 10 levels, and the nature of these differences, and developed a model for estimating missing PM 10 data for one measurement method. Methods: Government PM 10 measurements using two different technologies at five sites ( the Sierra -Anderson PM 10 HighVolume Air Sampler System, Hi -Vol ) ( every sixth day ) and the Rupprecht and Patashnik Tapered Element Oscillating Microbalance ( TEOM ) monitor ( daily ), as well as Harvard Impactor ( HI ) data collected for research purposes from one monitoring station were matched by day and monitoring site, then compared visually and with basic descriptive statistics. We fit linear regression models with airport visual range measurements, meteorological data, and information on other air pollutants to predict the Hi -Vol measured PM 10 levels for those days when direct measurements were not available. Results: We found relatively low correlations ( r ranging from 0.46 to 0.63 ) between PM 10 measured with the TEOM and Hi -Vol methods, and highly variable differences ( 0 -70 g / m 3 ) between the means of these measurements, depending on monitoring site. The HI measurements had a relatively high correlation with the Hi -Vol measurements ( r = 0.90 ). The models developed for the missing Hi -Vol measurements provided a series of estimated values similar to the actual Hi -Vol measurements, although the estimated series did not have high values in the range observed in the measured data. Conclusions: The differences we observed in the PM 10 measurements across methods in Mexico City may be important when studying health effects associated with particle exposure, evaluating method performance under conditions and operating protocols similar to those in Mexico City, and determining compliance with air quality standards. The estimated series of PM 10 measures may be a useful index of exposure for use in studies of the effects of air pollution on health.

A Case-Crossover Analysis of Out-of-Hospital Coronary Deaths and Air Pollution in Rome, Italy

“…PNC data were not directly available for the study period, but were retrospectively estimated from one monitoring station from 2001-2002. The statistical model predicts PNC well, and our estimation procedure was certainly superior (in terms of R 2 ) to what has been done to estimate PM 10 or PM 2.5 in previous studies (49)(50)(51)(52)(53)(54)(55). However, PNC is derived from existing air pollutants, and it is highly correlated to some of them ( Table 2).…”

Air Pollution and Mortality in Latin America

“…We excluded the Xalostoc station in Mexico City. This monitor is near industrial sources (Retama, personal communication, January 2005), as confirmed by systematically high concentration readings and relatively lower correlations with other monitor sites, 28 and is thus not likely to be representative of wider population exposure. Pearson correlations among the selected monitors within each city were all at least 0.50.…”

Do associations between airborne particles and daily mortality in Mexico City differ by measurement method, region, or modeling strategy?

“…Data from three different PM 10 metrics were available for comparisons: two from the government of Mexico City (one with daily concentrations, one with every sixth day measures, from 1994 to 1998), and one PM 10 metric from an academic research effort (daily concentrations in 1994 and 1995). In a previous analysis (O'Neill et al, 2002), we found low correlations between the two methods used by the Mexico City government to measure and regulate PM 10 , as well as between one of the government methods and the research group's data. We also found large variation in the mean levels measured by the two government methods, depending on the monitoring site (O'Neill et al, 2002).…”

“…In a previous analysis (O'Neill et al, 2002), we found low correlations between the two methods used by the Mexico City government to measure and regulate PM 10 , as well as between one of the government methods and the research group's data. We also found large variation in the mean levels measured by the two government methods, depending on the monitoring site (O'Neill et al, 2002). We speculated that the observed differences could be important for epidemiology studies of the health impacts of particulate air pollution.…”

“…Few reports, however, have addressed the differences that the PM 10 measurement technique may make in epidemiologic effect estimates. Although some studies have examined comparability of PM 10 measurement methods, the focus has not been on the associations with health outcomes (Allen et al, 1997;O'Neill et al, 2002).…”