Atmospheric measurements at several surface stations made between 1978 and 1990 of the anthropogenic chemical compound 1,1,1‐trichloroethane (methyl chloroform, CH3CCl3) show it increasing at a global average rate of 4.4 ± 0.2% per year (1σ) over this time period. The measured trends combined with industrial emission estimates are used in an optimal estimation inversion scheme to deduce a globally averaged CH3CCl3 tropospheric (and total atmospheric) lifetime of 5.7 (+0.7, −0.6) years (1σ) and a weighted global average tropospheric hydroxyl radical (OH) concentration of (8.7 ± 1.0) × 105 radical cm−3 (1σ). Inclusion of a small loss rate to the ocean for CH3CCl3 of 1/85 year−1 does not affect the stated lifetime but lowers the stated OH concentration to (8.1 ± 0.9) × 105 radical cm−3 (1σ). The rate of change of the weighted global average OH concentration over this time period is determined to be 1.0 ± 0.8% per year (1σ) which has major implications for the oxidation capacity of the atmosphere and more specifically for methane (CH4), which like CH3CCl3 is destroyed primarily by OH radicals. Because the weighting strongly favors the tropical lower troposphere, this deduced positive OH trend is qualitatively consistent with hypothesized changes in tropical tropospheric OH and ozone concentrations driven by tropical urbanization, biomass burning, land use changes, and long‐term warming. We caution, however, that our deduced rate of change in OH assumes that current industry estimates of anthropogenic emissions and our absolute calibration of CH3CCl3 are accurate. The CH3CCl3 measurements at our tropical South Pacific station (Samoa) show remarkable sensitivity to the El Nino‐Southern Oscillation (ENSO), which we attribute to modulation of cross‐equatorial transport during the northern hemisphere winter by the interannually varying upper tropospheric zonal winds in the equatorial Pacific. Thus measurements of this chemical compound have led to the discovery of a previously unappreciated aspect of tropical atmospheric tracer transport.

We present and interpret long‐term measurements of the chemically and radiatively important trace gas nitrous oxide (N2O) obtained during the Atmospheric Lifetime Experiment (ALE) and its successor the Global Atmospheric Gases Experiment (GAGE). The ALE/GAGE data for N2O comprise over 110,000 individual calibrated real‐time air analyses carried out over a 10‐year (July 1978–June 1988) time period. These measurements indicate that the average concentration in the northern hemisphere is persistently 0.75±0.16 ppbv higher than in the southern hemisphere and that the global average linear trend in N2O lies in the range from 0.25 to 0.31% yr−1, with the latter result contingent on certain assumptions about the long‐term stability of the calibration gases used in the experiment. Interpretation of the data, using inverse theory and a 9‐box (grid) model of the global atmosphere, indicates that the N2O surface emissions into the 90°N–30°N, 30°N–0°, 0°–30°S, and 30°S–90°S semihemispheres account for about 22–34, 32–39, 20–29 and 11–15% of the global total emissions, respectively. The measured trends and latitudinal distributions are consistent with the hypothesis that stratospheric photodissociation is the major atmospheric sink for N2O, but they do not support the hypothesis that the temporal N2O increase is caused solely by increases in anthropogenic N2O emissions associated with fossil fuel combustion. Instead, the cause for the N2O trend appears to be a combination of a growing tropical source (probably resulting from tropical land disturbance) and a growing northern mid‐latitude source (probably resulting from a combination of fertilizer use and fossil fuel combustion). The exact combination of these sources which best fits the data depends on the assumed tropospheric‐stratospheric exchange rates for N2O in the northern hemisphere relative to the southern hemisphere. Accepting a theoretically‐calculated N2O lifetime of 166±16 years due to stratospheric destruction only, we deduce from the ALE/GAGE data a 10‐year average global N2O emission rate of (20.5±2.4) × 1012 g N2O yr−1, but with significant year‐to‐year variations in emissions associated perhaps with year‐to‐year variations in tropical land disturbance.

We study a mechanism of iceberg breakup that may act together with the recognized melt and wave-induced decay processes. Our proposal is based on observations from a recent field experiment on a large ice island in Baffin Bay, East Canada. We observed that successive collapses of the overburden from above an unsupported wavecut at the iceberg waterline created a submerged foot fringing the berg. The buoyancy stresses induced by such a foot may be sufficient to cause moderate-sized bergs to break off from the main berg. A mathematical model is developed to test the feasibility of this mechanism. The results suggest that once the foot reaches a critical length, the induced stresses are sufficient to cause calving. The theoretically predicted maximum stable foot length compares well to the data collected in situ. Further, the model provides analytical expressions for the previously observed "rampart-moat" iceberg surface profiles.

Thirteen years of Atmospheric Lifetime Experiment/Global Atmospheric Gases Experiment CCl3F and CCl2F2 measurements at five remote, surface, globally distributed sites are analyzed. Comparisons are made against shipboard measurements by the Scripps Institution of Oceanography group and archived air samples collected at Cape Grim, Tasmania, since 1978. CCl3F in the lower troposphere was increasing at an average rate of 9.2 ppt/yr over the period July 1978 to June 1988. CCl2F2 was increasing at an average 17.3 ppt/yr in the lower troposphere over the same period. However, between July 1988 and June 1991 the increases of CCl3F and CCl2F2 in this region have averaged just 7.0 ppt/yr and 15.7 ppt/yr, respectively. The rate of increase has been decreasing 2.4 ppt/yr2 and 2.9 ppt/yr2 over this 3‐year period. Based on a recent scenario of the global releases of these compounds and using the new calibration scale SIO 1993, the equilibrium lifetimes are estimated to be and years for CCl3F and CCl2F2, respectively. Using these lifetime estimates and a two‐dimensional model, it is estimated that global releases of these two chlorofluorocarbons in 1990 were 249±28×106 kg for CCl3F and 366±30×106 kg for CCl2F2. It is also estimated that combined releases of these chlorofluorocarbons in 1990 were 21±5% less than those in 1986.

Frequent atmospheric measurements of the anthropogenic compound methylchloroform that were made between 1978 and 1985 indicate that this species is continuing to increase significantly around the world. Reaction with the major atmospheric oxidant, the hydroxyl radical (OH), is the principal sink for this species. The observed mean trends for methylchloroform are 4.8, 5.4, 6.4, and 6.9 percent per year at Aldrigole (Ireland) and Cape Meares (Oregon), Ragged Point (Barbados), Point Matatula (American Samoa), and Cape Grim (Tasmania), respectively, from July 1978 to June 1985. These measured trends, combined with knowledge of industrial emissions, were used in an optimal estimation inversion scheme to deduce a globally averaged methylchloroform atmospheric lifetime of 6.3 (+ 1.2, -0.9) years (1sigma uncertainty) and a globally averaged tropospheric hydroxyl radical concentration of (7.7 +/- 1.4) x 10(5) radicals per cubic centimeter (1sigma uncertainty). These 7 years of gas chromatographic measurements, which comprise about 60,000 individual calibrated real-time air analyses, provide the most accurate estimates yet of the trends and lifetime of methylchloroform and of the global average for tropospheric hydroxyl radical levels. Accurate determination of hydroxyl radical levels is crucial to understanding global atmospheric chemical cycles and trends in the levels of trace gases such as methane.