The Atmospheric Lifetime Experiment: 5. Results for CH₃CCl₃ based on three years of data

Abstract

We present gas chromatographic determinations of the concentrations of CH 3 CCl 3 at Adrigole (Ireland), Cape Meares (Oregon), Ragged Point (Barbados), Point Matatula (American Samoa), and Cape Grim (Tasmania) for the 3‐year period July 1978 through June 1981. The determinations involve approximately four measurements each day with on‐site calibration. The absolute values and trends for these observed concentrations are interpreted in terms of the industrial production, global circulation rate, and atmospheric lifetime of CH 3 CCl 3 by using an optimal estimation technique that incorporates a nine‐box model of the atmosphere. The globally and annually averaged trend in the lower troposphere at the midpoint of the second year of the experiment is 8.7% per year, and the inferred global atmospheric content of CH 3 CCl 3 at this midpoint is 2.58×10 9 kg. The global atmospheric lifetime deduced by using the observed trends and global content together with current estimates of industrial CH 3 CCl 3 emissions is years. This deduced lifetime is sufficiently long to imply that the observed variability in the data on seasonal and shorter time scales must be dominated by meteorological (and perhaps industrial emission) variabilities rather than by spatial and temporal variations in the rate of chemical destruction of CH 3 CCl 3 by OH. However, the observed variations on annual and longer time scales are sensitive to the spatially and temporally averaged OH concentrations. In particular, the globally averaged tropospheric OH concentration compatible with the above‐deduced CH 3 CCl 3 lifetime is (5 ± 2) × 10 5 molecule cm −3 ; in reasonable agreement, for example, with the value of ( ) × 10 5 molecule cm −3 deduced by Volz et al. (1981) from measurements of CO. Our results are sensitive to constraints imposed on uncertainties in the CH 3 CCl 3 emissions and absolute mixing ratios. If we decrease emissions by ≲8% in the years 1976–1978 and increase emissions in 1979–1981 to compensate, and if absolute mixing ratios are decreased by about 18%, then the best estimate of the lifetime from our data decreases to 6.5 years.