Greenhouse gas methane in the Arctic atmosphere has not been accurately reported for 1900–1980 from either direct observations or ice core reconstructions. By using trace gas data from firn (compacted snow layers above ice sheet), air samples at two Greenland sites, and a firn air transport model, this study suggests a likely range of the Arctic methane reconstruction for the 20th century. Atmospheric scenarios from two previous studies are also evaluated for consistency with the firn data sets.

We present O₂/N₂ and Ar/N₂ records from the Dome Fuji ice core through the bubbly ice, bubble–clathrate transition, and clathrate ice zones without gas-loss fractionation. The insolation signal is preserved through the clathrate formation. The relationship between Ar/Ν₂ and Ο₂/Ν₂ suggests that the fractionation for the bubble–clathrate transition is mass independent, while the bubble close-off process involves a combination of mass-independent and mass-dependent fractionation for O₂ and Ar.

Observing the minimal long-term change in atmospheric O₂ molar fraction combined with CO₂ observation enables us to estimate terrestrial biospheric and oceanic CO₂ uptakes separately. In this study, we firstly identified the span offset between the laboratory O₂ scales using our developed high-precision standard mixtures, suggesting that the result may allow us to estimate terrestrial biospheric and oceanic CO₂ uptakes precisely.

The surface Ar / N2 ratio showed not only secular increasing trends, but also interannual variations in phase with the global ocean heat content (OHC). Sensitivity test by using a two-dimensional model indicated that the secular trend in the Ar / N2 ratio is modified by the gravitational separation in the stratosphere. The analytical results imply that the surface Ar/N2 ratio is an important tracer for detecting spatiotemporally integrated changes in OHC and stratospheric circulation.

Air in polar ice cores provides information on past atmosphere and climate. We present a new method for simultaneously measuring eight gases (CH₄, N₂O and CO₂ concentrations; isotopic ratios of N₂ and O₂; elemental ratios between N₂, O₂ and Ar; and total air content) from single ice-core samples with high precision.

Atmospheric O₂ and CO₂ concentrations, along with CO₂ flux, have been observed in a megacity, Tokyo, Japan. The O₂ : CO₂ exchange ratio for net turbulent O₂ and CO₂ fluxes (OR_F) between the urban area and the overlaying atmosphere was obtained, and we applied it to estimate the diurnal cycles of CO₂ fluxes from gas and liquid fuel consumption separately. We found simultaneous observations of OR_F and CO₂ flux are useful in validating CO₂ emission inventories from statistical data.

The velocity of stratospheric circulation is often measured by the time since the air entered the stratosphere. This study tries to understand its vertical profile in the tropics by comparing observational data and model simulations. Our interpretation mutually consistent among them is encouraging, while some limitations such as the treatment of seasonal variation of CO₂ and mesospheric loss of SF₆ are reconfirmed stressing a need of using multiple variables in the future.

Observation of atmospheric O₂ requires highly precise standard gas mixtures with uncertainty of less than 1 ppm for the O₂ mole fraction or 5 per meg for O₂ / N₂. The uncertainty had not been achieved due unknown uncertainty factors in mass determination of the filled source gases. We first developed the primary standard mixtures with 1 ppm for the O₂ mole fraction or 5 per meg by identifying and reducing the unknown uncertainty factors.

Isotope measurements are useful for separating different methane sources. However, the lack of widely accepted standards and calibration methods for stable carbon and hydrogen isotopic ratios of methane in air has caused significant measurement offsets among laboratories. We conducted worldwide interlaboratory comparisons, surveyed the literature and assessed them systematically. This study may be of help in future attempts to harmonize data sets of isotopic composition of atmospheric methane.

This is the first research that shows concrete evidence of gravitational separation in the tropical stratosphere. This implies that gravitational separation occurs within the entire stratosphere, which gives us new insight into atmospheric dynamics.

By analysis of whole air samples collected by balloon-borne compact cryogenic samplers, we found that apparent isotope effect for stratospheric N₂O between 25 and 30 km over the Equator is larger than that observed in other latitudes and that it is almost equal to the effect predicted by laboratory simulation experiments. These results suggest that equatorial middle stratosphere can be treated as an isolated region when we consider the decomposition of N₂O by photochemical processes.

We report a comparison of the stratospheric mean-meridional circulation and eddy mixing in the stratospheric Brewer-Dobson circulation (BDC) among the six reanalysis products. Overall, discrepancies between the different variables and trends therein as derived from the different reanalyses are still relatively large, suggesting that more investments in these products are needed in order to obtain a consolidated picture of observed changes in the BDC and the mechanisms that drive them.

We developed an atmospheric N2O isotopocule model based on a chemistry-coupled atmospheric general circulation model and a simple method to optimize the model, and estimated the isotopic signatures of surface sources at the hemispheric scale. Data obtained from ground-based observations, measurements of firn air, and balloon and aircraft flights were used to optimize the long-term trends, interhemispheric gradients, and photolytic fractionation, respectively, in the model.

Atmospheric CH4 increased from 900ppb to 1800ppb during the period 1900–2010 at a rate unprecedented in any observational records. We use bottom-up emissions and a chemistry-transport model to simulate CH4. The optimized global total CH4 emission, estimated from the model–observation differences, increased at fastest rate during 1940–1990. Using δ13C of CH4 measurements we attribute this emission increase to biomass burning. Total CH4 lifetime is shortened by 4% over the simulation period.