Now a team from France, the US and the UK has used unique atmospheric carbon dioxide measurements collected over a vertical profile by aircraft from the US National Oceanic and Atmospheric Administration/Earth System Research Laboratory (NOAA/ESRL) to estimate the land carbon sink over the US. The comprehensive data meant that the team was able to estimate sink size by an analysis using only wind measurements, rather than having to model transport of the gas.

"Our approach concentrates on sampling the full 3D field and avoiding as much as possible deficiencies of transport models," Cyril Crevoisier of the Pierre Simon Laplace Institute, France, told environmentalresearchweb. "To that end, we use the unique aircraft data available through the NOAA/ESRL Carbon Cycle Group Aircraft Program (www.esrl.noaa.gov/gmd/ccgg/aircraft/), which largely solve the need to accurately solve the vertical transport by making 12 measurements from the surface to 25,000 feet at 20 sites over North America."

According to Crevoisier, thanks to this unique dataset the team is able to focus on the most conservative atmospheric inverse-modelling approach, which is to balance in- and out-flow of carbon into the troposphere. "We use the same air-mass transport fields as used by traditional atmospheric inverse approaches, but our method is substantially less sensitive to vertical transport biases because the method focuses on the vertical integral of atmospheric carbon dioxide," he said.

The researchers found that the mainland US, excluding Alaska, absorbed roughly 0.5 Petagrammes of carbon per year between 2004 and 2006. The figure agreed well with the estimate based on forest inventory of the first North American State of the Carbon Cycle Report. The Midwest and Southeast took up the most carbon; this was probably due to crops in the Midwest and secondary forest regrowth in the Southeast.

Crevoisier says that more than half of the apparent atmospheric sink was located in the Midwest states. But he feels it's important to stress that this does not imply a long-term sequestration of carbon in that region, since the carbon stored in the crops will be transported and consumed elsewhere, for example in the pastureland of western and southern regions.

"Our method provides a completely independent estimate of carbon fluxes in North America, which is corroborated by on-ground measurements," he added. "Due to its simplicity and robustness, and to the use of a unique set of data giving access to the 3D distribution of carbon dioxide over the continent, our method may thus been seen as a benchmark for flux estimate and suggests a direction forward in the field."

The researchers reckon that a relatively small increase in aircraft profile-sampling frequency could provide substantially more accurate flux estimates. An increase in spatial coverage would also be beneficial.

"Our work also provides a means to partition the sink in different regions and to thus link variations in carbon uptake to vegetation, climate and human drivers, a well-defined priority of the North American Carbon Program," said Crevoisier. "Such approaches are also attractive for other land regions of the globe where carbon budget is poorly known (e.g. the Amazon basin)."

The aircraft profiles could also be used to validate carbon dioxide observations made from spaceborne instruments, which typically measure a single column average of the gas over several locations each day. Using the vertical gradient captured in the aircraft profiles may even enable conversion of space observations into fluxes.

"Higher-altitude measurements are needed to have access to the full column of carbon dioxide and address several questions for atmospheric transport, such as the vertical convection, or the tropospheric/stratospheric air exchange, which is still poorly known and one of the major flaws in current climate models," said Crevoisier. "To this end, the NOAA/ESRL Carbon Cycle Group Aircraft Program has plans to not only put instruments on commercial aircraft that regularly fly to 42,000 feet but also balloons where we can get measurements up to 90,000 feet."

The researchers reported their work in PNAS.