Krupa S V, Legge A H
Department of Plant Pathology, University of Minnesota, 495 Borlaug Hall, 1991 Buford Circle, St. Paul, MN 55108, USA.
Environ Pollut. 2000 Jan;107(1):31-45. doi: 10.1016/s0269-7491(99)00154-2.
During some past two decades there has been a growing interest among air pollution-vegetation effects-scientists to use passive sampling systems for quantifying ambient, gaseous air pollutant concentrations, particularly in remote and wilderness areas. On the positive side, excluding the laboratory analysis costs, passive samplers are inexpensive, easy to use and do not require electricity to operate. Therefore, they are very attractive for use in regional-scale air quality assessments. Passive samplers allow the quantification of cumulative air pollutant exposures, as total or average pollutant concentrations over a sampling duration. Such systems function either by chemical absorption or by physical adsorption of the gaseous pollutant of interest onto the sampling medium. Selection of a passive sampler must be based on its known or tested characteristics of specificity and linearity of response to the chemical constituent being collected. In addition, the effects of wind velocity, radiation, temperature and relative humidity must be addressed in the context of absorbent/adsorbent performance and sampling rate. Because of all these considerations, passive samplers may provide under- or overestimations of the cumulative exposures, compared to the corresponding data from co-located continuous monitors or active samplers, although such statistical variance can be minimized by taking necessary precautions. On the negative side, cumulative exposures cannot identify short-term (<few hours) pollutant episodes or regulatory non-compliance, where appropriate. Equally important, cumulative exposures (e.g. with ozone, a non-accumulating pollutant in plants) cannot account for the stochasticity and the dynamics of air pollutant exposure and plant (particularly deciduous vegetation) response. Although still being tested, statistical methods appear to be available to reconstruct passive sampler data to mimic the dynamics of the co-located continuous monitoring data. Regional level air quality and ecological risk assessments (multi-point models) as a whole should consider this type of approach in the future, using co-located passive samplers, with continuous monitors at selected locations as calibration points for mapping the air quality distribution on temporal and spatial scales.
