Field performance evaluation of Maxxam passive samplers for regional monitoring of ambient SO2, NO2 and O3 concentrations in Alberta, Canada

A field performance evaluation of Maxxam passive samplers was carried out for ambient concentrations of sulfur dioxide (SO2), nitrogen dioxide (NO2) and ground-level ozone (O3). Monthly passive and hourly continuous air monitoring data from four regional air monitoring networks in Alberta, Canada were evaluated over a 5-year period (2006-2010). Monthly concentrations were relatively low, ranging from 0.1 to 3.9 ppb, 0.2-18.1 ppb and 10.1-56.1 ppb for SO2, NO2 and O3, respectively. From duplicate passive sampling, geometric mean precision values were 17.9%, 14.8% and 4.7% for SO2, NO2 and O3, respectively. Geometric mean of the relative error (as a measure of accuracy) was 30% (median = 33%, interquartile range, IQR 15-63%) for SO2 and 32% (median = 40%, IQR = 25-64%) for NO2. O3 measurements had a better measure of accuracy with a geometic mean relative error of 12% (median = 17%, IQR = 8-27%) and met the acceptable level recommended by United States National Institute of Safety and Health (NIOSH) and the European Union (EU) Directive (±25%). From reduced major axis (RMA) analysis, bias (systematic error) is apparent in the Maxxam passive samplers in the field resulting in overestimation of ambient SO2 and O3 concentrations and underestimation of NO2 concentrations relative to continuous analyzers. Seasonal influences were observed for accuracy of passive SO2 and O3 measurements. A poor association was found between passive versus continuous concentrations for SO2 and O3 during the winter and the summer, respectively.