Thirty-three analytical laboratories providing diffusion tube analysis for UK NO₂ Network participants were invited to take part in the 1999 field intercomparison exercise.

Seven nitrogen dioxide diffusion tubes (six exposure tubes and one travel blank) were supplied by each laboratory, for exposure over a period of one month, August 1999. As in the 1998 study, diffusion tubes were exposed simultaneously, upon purpose made exposure racks located immediately adjacent to the automatic chemiluminescent NO_x monitoring equipment installed at the DETR's Automatic Urban Network (AUN) site at Walsall Alumwell. The exposure time was a 28-day period from 4^th August to 1^st September 1999.

Upon completion of exposure, diffusion tubes were capped and the exposure time noted. The exposed samplers and travel blanks were then returned to the supplying laboratory for analysis. Travel blanks accompanied exposure tubes to and from the test site. They were isolated in sealed sample bags, and refrigerated throughout the exposure period.

The participating laboratories sent their analytical results to AEA Technology for collation. Results reported in micrograms per cubic metre (µg m^-3) were converted to parts per billion (ppb) using a conversion factor of 0.523.

Prior to the interpretation of the measurement data supplied by the laboratories, outlying data within the datasets for each laboratory were removed using the Grubb's Test. This statistical test was used in an iterative process, at a probability level of P=0.05. It was assumed that the sample distribution was normal. In order to enable a comparison of 1999 data with similar data from the 1998 field intercomparison, outliers from the 1998 dataset were removed retrospectively using the same test.

The decision to remove outliers was taken in order to provide better average estimates of laboratory performance and also take into account the levels of screening employed within the UK NO₂ Network whereby abnormally low or high results are investigated and deleted from the Network's dataset where appropriate. The presence of this level of screening was not recognised in the summary report for the 1998 intercomparison exercise.

Chapter 1         Chapter 3