Systematic Biases in Measurement of Urban Nitrogen Dioxide using Passive Diffusion Samplers

Measurement of nitrogen dioxide using passivediffusion tube over 22 months in Cambridge, U.K. areanalysed as a function of sampler exposure time, andcompared with NO₂ concentrations obtained from aco-located chemiluminescence analyser. The averageratios of passive sampler to analyser NO₂ at acity centre site (mean NO₂ concentration 22 ppb)are 1.27 (n = 22), 1.16 (n = 34) and 1.11 (n = 7) forexposures of 1, 2 and 4-weeks, respectively. Modellingthe generation of extra NO₂ arising from chemicalreaction between co-diffusing NO and O₃ in thetube gave a ratio (modelled/measured) of 1.31 for1-week exposures. Such overestimation is greatest whenNO₂ constitutes, on average, about half of totalNO_x (= NO + NO₂) at the monitoring locality.Although 4-week exposures gave concentrations whichwere not significantly different from analyserNO₂, there was no correlation between thedatasets. At both the city-centre site and anothersemi-rural site (mean NO₂ concentration 11 ppb)the average of the aggregate of four consecutive1-week sampler exposures or of two consecutive 2-weeksampler exposures was systematically greater than fora single 4-week exposure.The results indicate two independent and opposingsystematic biases in measurement of NO₂ bypassive diffusion sampler: an exposure-timeindependent chemical overestimation with magnitudedetermined by local relative concentrations of NO andO₃ to NO₂, and an exposure-time dependentreduction in sampling efficiency. The impact of theseand other potential sources of systematic bias on theapplication of passive diffusion tubes for assessingambient concentrations of NO₂ in short (1-week)or long (4-week) exposures are discussed in detail.