After a late flight into Foz de Iguacu in Brazil, a city known for spectacular waterfalls that are actually split between not two but three countries, I kicked things off early at AGU with a shot of pre-sweetened coffee and a press briefing suggesting that carbon dioxide releases from rivers in the temperate regions are non-trivial, if poorly understood.

The unpublished work by David Butman, a graduate student under Peter Raymond at the Yale University’s School of Forestry and Environmental Studies in New Haven, Connecticut, is based off of long-term monitoring data from the US Geological Survey. Using such factors as temperature and alkalinity, the researchers calculated carbon dioxide concentrations for US rivers, which results in emissions of about 0.2 gigatonnes of carbon annually. The general assumption in the carbon-cycle models, he says, has been that all of that carbon is merely exported to the ocean.

The next step is to extrapolate out, and Butman’s calculations suggest that total emissions for temperate regions add up to about 1.7 gigatonnes carbon per year. For perspective, this is higher than the total carbon emissions from fossil fuels in the United States in 2008 (which ring in around 1.5 gigatonnes after converting from carbon dioxide to carbon).

—

Butman then piggybacked on an earlier paper published in Nature (see also News and Views) in 2002 to come up with a more global estimate. That paper, which was based on measurements in the Amazon, suggested that rivers and streams in the humid tropics could be responsible for 0.9 gigatonnes of carbon emissions each year. In sum, these estimates suggest that this process could pump roughly 2.7 gigatonnes of carbon into the atmosphere annually, and Butman says the numbers for lakes and reservoirs could be even higher.

“Inland freshwater systems are definitely going to contribute carbon dioxide on a magnitude that is going to make a difference,” Butman says.

The last question is perhaps the biggest: where does this CO2 come from? And on that point, Butman says the data simply isn’t in yet. Some of it could come from plants within the river ecosystem in rivers like the Amazon and perhaps some major rivers in the north, such as the Mississippi in the United States. Alternatively, much of it could come from soil erosion, in which case it is merely carbon emissions that are already counted under terrestrial cycle being shifted into freshwater systems. “This is really the beginning of a larger research project,” he says.