Correction for water vapor in the measurement of atmospheric trace gases

The presence of water vapor in a sample of air reduces the concentration of a trace gas measured from the sample. We present a methodology to correct for this effect for those cases when the concentration of the trace gas has already been measured from a wet sample. The conversion or correction factor that takes the wet mole fraction to a dry mole fraction is determined by the mixing ratio of water vapor inside the sampling canister. For those samples where the water vapor is saturated inside the canister, the water vapor mixing ratio is largely determined by laboratory conditions; for the unsaturated samples, the mixing ratio is determined by station conditions. If the meteorology at the sampling station is known, the equations presented here can be used directly to calculate the appropriate correction factor. For convenience, we use climatological data to derive average monthly correction factors for seven common global sampling sites: Barrow, AK, US (71degreesN, 157degreesW); Cape Meares, OR, US (45degreesN, 124degreesW); Mauna Loa, HI, US (19degreesN, 155degreesW); Ragged Point, Barbados (13degreesN, 59degreesW); American Samoa (14degreesS, 171degreesW); Cape Grim, Tasmania, Australia (41degreesS, 145degreesE); South Pole (90degreesS). These factors adjust wet mole fractions upwards within a range of 0.002% for the South Pole to over 0.8% for saturated sites. We apply the correction factors to wet nitrous oxide (N2O) mole fractions. The corrected data are more consistent with our understanding of N2O sources. (C) 2002 Published by Elsevier Science Ltd.