The results of atmospheric CO₂, CH₄ and CO measurements are presented. The measurements were made in air samples collected at heights of 4, 25, 100, 200 and 300 m above ground, and in the atmospheric column in Obninsk, Russia (55.11 N, 36.57  E, 183 m asl).

Simulations with a regional transport model are evaluated in order to determine to which extend the indirect fossil fuel combustion tracer CO or the purely anthropogenic tracer SF₆ can be used to retrieve the contribution of fossil fuel emissions in the atmospheric CO₂ signal.

Based on simulations with high-resolution transport models we investigate the detectability of surface flux signals in the atmospheric CO₂ concentration and infer some general guidelines for the sampling of the continental troposphere for the purpose of constraining mid-latitude land carbon sinks.

Upwelling of high-nutrient waters in the equatorial Pacific gives rise to a band of enhanced primary production around the equator that stretches from Peru almost to Indonesia. It has been suggested that this oceanic region accounts for a large part of global net production. The equatorial Pacific is also thought to be the largest oceanic CO₂ source and makes an important contribution to the atmospheric CO₂ budget.

We examine the growth-rate of atmospheric CO₂ in 2002 and 2003. Observations show consecutive increases of greater than 2 ppmv per year for the first time on the Mauna Loa record. We use a statistical regression to show that increasing anthropogenic emissions and ENSO activity are unable to account for the CO₂ growth-rates of 1992 and 1993 following the Pinatubo volcanic eruption, or the anomalously high growth-rate of 2003. Increased forest fires in the northern hemisphere, consistent with remote-sensing and carbon monoxide measurements, seem likely to have contributed significantly to the 2003 anomaly. We hypothesise that the hot and dry Eurasian summer of 2003 led to an increase in forest fire emissions from Siberia, and may also have directly suppressed land-carbon uptake. Model results lead us to expect a steady increase in airborne fraction as climate change weakens the natural carbon sink and accelerates CO₂ rise.

In recent years our knowledge of gas exchange across the air-sea interface at the process level has improved as a consequence of new instrumentation and novel use of injected and natural tracers.  However, there remains significant uncertainty in the extrapolation of these results to larger scales, especially for studies focusing on global-scale processes such as the earth's carbon cycle.

We have constructed an estimate of annual-mean surface fluxes of carbon dioxide for the period 1992-6 using observational constraints from the atmosphere and from the ocean interior. The method interprets in situ observations of carbon dioxide concentration in the ocean and atmosphere using transport estimates from global circulation models.

High-quality data of total inorganic carbon (TCO₂) and other oceanographic parameters have been acquired repeatedly between 1994 and 2003 along 137ºE (WOCE P9) in the western North Pacific. They indicate the significant increase in TCO₂, apparent oxygen utilization (AOU) and preformed TCO₂ in the water columns between 20ºN and 30ºN, in particular, in the North Pacific Subtropical Mode Water (NPSTMW). The increase in the preformed TCO₂ suggests the 0.9 to 1.1 mol m^-2 yr^-1 accumulation of the anthropogenic CO₂ in this region. However, the change in the preformed TCO₂ associated with the change in the formation region and/or advection of NPSTMW is also suggested.

To examine vertical distributions of the O₂/N₂ ratio in the stratosphere, air samples were collected using a cryogenic sampler over Sanriku, Japan and Syowa, Antarctica. It was clearly seen that d(O₂/N₂), as well as simultaneously measured d¹⁵N of N₂ and d¹⁸O of O₂, decreased gradually with increasing height in the stratosphere. The observed profiles of stratospheric ï€ d¹⁵N and d¹⁸O were in good agreement with those calculated using a steady state 1-dimensional eddy-diffusion/molecular-diffusion model suggesting that the upward decrease of stratospheric d(O₂/N₂) is caused by O₂ and N₂ molecules fractionated differently by gravity. The stratospheric d(O₂/N₂) corrected for the gravitational separation indicated that the average value at heights above 20-25 km over Sanriku was always higher than the upper tropospheric d(O₂/N₂) value over Japan at the corresponding time, and that it has decreased secularly, as was found in the troposphere.