Summary
We present measurements and analysis of the isotopic composition (ä15Nbulk, ä15Ná, ä15Nâ, and ä18O) of nitrous oxide in 50 samples of tropospheric air collected at Cape Grim, Tasmania between 1978 and 2005 and in 11 samples of firn air from Law Dome, Antarctica, with mean ages dating from 1940 to the 1990s. Over this 65 year period, the isotopic compositions show overall decreasing trends, consistent with previous firn air measurements and with the prediction that the increase in atmospheric N2O is a result of agricultural and other human activities that produce N2O that is isotopically lighter than the natural sources. For the Cape Grim archived air, the trends in ä15Nbulk, ä15Ná, ä15Nâ, and ä18O are -0.036±0.003, -0.028±0.022, - 0.038±0.022, and -0.023±0.006‰ yr-1, respectively, while those for the Law Dome firn air are -0.033±0.004, -0.031±0.014, -0.034±0.017, and -0.023±0.003‰ yr- 1, respectively. Although not statistically significant, there is a slightly larger negative trend in ä15Nâ versus that for ä15Ná15Ná and ä15Nâ trends among these archived air datasets is real atmospheric temporal and/or spatial variability is likely to have implications for the regional or global N2O isotope budgets, since box model results suggest that the relative trends in ä15Ná and ä15Nâ depend on the isotopic compositions of the natural and anthropogenic surface sources, as well as implications for constraining the relative contributions of nitrification and denitrification to the atmospheric N2O burden. In addition to the decreasing trends, time series analyses of the Cape Grim archived air demonstrate for the first time that small but statistically significant mean seasonal cycles in ä15Nbulk, ä15Ná, and ä18O can occur at the surface, quite remarkable for a trace gas with a 120-year lifetime. Box model simulations show the timing of these seasonal cycles is consistent with seasonally varying fluxes of N2O from the stratosphere, thermal outgassing from the ocean, and Southern Ocean ventilation. Interannual variations are also apparent at Cape Grim, particularly a large change in the amplitude of the seasonal cycles during the large ENSO in 1997-1998. Overall, these data provide important new constraints on the N2O budget and how N2O concentrations have changed on annual, interannual, and decadal time scales, and should thus contribute to a more detailed mechanistic understanding of what environmental variables control the biogeochemical cycling of atmospheric N2O.