Comparison of Particulate Mercury Measured with Manual and Automated Methods

A study was conducted to compare measuring particulate mercury (Hg-P) with the manual filter method and the automated Tekran system. Simultaneous measurements were conducted with the Tekran and Teflon filter methodologies in the marine and coastal continental atmospheres. Overall, the filter Hg-P values were on the average 21% higher than the Tekran Hg-P, and >85% of the data were outside of +/-25% region surrounding the 1:1 line. In some cases the filter values were as much as 3-fold greater, with <5% of the points falling on the 1: 1 line. A common characteristic in all seasons was that the Tekran only yielded a total of 6 data points above 1 part per quadrillion by volume (ppqv) (i.e., similar to 4% of the observations), and had similar to 25% of its measurements as below the limit of detection (<0.1-0.2 ppqv). In comparison, the filter always had Hg-P detectable above the blank level of 0.05 ppqv. The aerosol size distribution of Hg-P did not appear to be a major factor in the discrepancies between the two methods. The peaks in filter Hg-P were always concomitant with enhanced mixing ratios of selected hydrocarbons, halocarbons, and oxygenated compounds. Backward trajectories suggested that the peaks in all chemical compounds were primarily anthropogenic, and tracers indicated a combustion signature. Since the Tekran was typically unresponsive to these pollution episodes, detailed investigation of aerosol passing efficiency and the instrument response to different aerosol types should be investigated.