An intercomparison of airborne PAN measurements

Abstract

As part of the NASA Tropospheric Chemistry Program a series of field intercomparisons have been initiated to evaluate state‐of‐the‐art capability for measuring key tropospheric species. These intercomparisons, designated as Chemical Instrumentation Test and Evaluation (CITE), are conducted as part of NASA's Global Tropospheric Experiment. The objectives of the second series of instrument tests, CITE 2, were to evaluate instrumentation for measuring NO2, HNO3, and peroxyacetyl nitrate (PAN) and to determine for various tropospheric environments the relative abundances and partitioning among the major nitrogen species. This paper summarizes the results from the PAN instrument intercomparisons. Other results are addressed in companion papers. Both PAN instruments use the same detection principle of electron capture gas chromatography of a cryogenically enriched sample of ambient air. NASA Ames Research Center and the National Center for Atmospheric Research were responsible for the respective instruments. The intercomparisons included three exchanges of standards and 13 intercomparison flights in which a variety of types of air masses were sampled. Nine flights were based from Ames Research Center, California, and the remaining four were ferry flights between Ames and Wallops Flight Center, Virginia (aircraft home base). Flight altitudes ranged from 150 to 5000 m above ground level. All flights but one were during daylight hours. PAN mixing ratios during the flight intercomparison periods were generally <300 parts per trillion by volume (pptv), and about 40% of the results for mixing ratios were <100 pptv. At mixing ratios of <100 pptv the two instruments agreed on the average to about 17 pptv with a 95% confidence interval of ±9 pptv. Instrument agreement at PAN levels of 100–300 pptv was of the order of 25% with a confidence interval of ±6%. These levels of agreement are within expected limits based on the stated accuracy and precision of the two instruments. However, it is noted that for some of the individual intercomparison data periods, agreement was outside expected levels.