Nitrous oxide is a powerful greenhouse gas and it plays a key role in atmospheric chemistry and ozone destruction. Atmospheric levels of the gas have been rising as a result of man's activities, including growing crops and grazing animals.

The Intergovernmental Panel on Climate Change (IPCC) recommends calculating nitrous oxide emissions from grasslands by assuming that output of the gas increases linearly with grazing density because more animals supply more nitrogen in their waste and in decomposing plant residues. But now a team from Germany, China and the UK has found that, at least in grasslands in semi-arid, cool temperate regions, this approach could overestimate nitrous oxide emissions by up to 72%.

"In contrast to general assumptions, grazing was found not to increase nitrous oxide emissions in grasslands of cool temperate climates," Klaus Butterbach-Bahl of Karlsruhe Institute of Technology, Germany told environmentalresearchweb. "There is a need to revise and supplement current methodology to calculate effects of livestock on N₂O emissions in steppe/prairie systems."

Butterbach-Bahl and colleagues measured nitrous oxide emissions over a period of one year at 10 grassland sites with various levels of grazing in Inner Mongolia, China, at time intervals of three hours and one week. Other studies, in contrast, have tended to focus on the growing season and have taken relatively infrequent measurements over short time periods with.

"Steppe/prairie systems cover huge parts of the terrestrial surface, but are hardly studied outside of North America," said Butterbach-Bahl. "The investigated steppe region is typical for huge parts of the Eurasian steppe belt."

The team found that ungrazed steppe emitted a large pulse – about 72% of its annual emissions – of nitrous oxide during the spring thaw. Steppe grazed more heavily by livestock emitted a smaller pulse of the gas in spring. The most heavily grazed sites emitted just 8% of their annual emissions during the thaw.

As a result of this spring-thaw pulse, grasslands supporting more livestock emitted less nitrous oxide over the course of a year, even though they had a higher output of the gas during the growing season, in line with the theory behind current calculation techniques. The researchers believe that existing approaches may overestimate nitrous oxide emissions for semi-arid cool, temperate grasslands by up to 72%. Such grasslands probably account for one-third of temperate grasslands worldwide, an area of around 10 million sq km, according to Stephen Del Grosso of the US Department of Agriculture.

Butterbach-Bahl and colleagues from Karlsruhe Institute of Technology, the Chinese Academy of Sciences, and the UK's Center for Ecology and Hydrology reckon that a higher stock density crops vegetation lower to the ground, which in turn allows snow to erode more quickly in spring, cuts soil temperature and reduces the amount of soil moisture. This provides less favourable conditions for the microbes that respire anaerobically and cause denitrification in water-logged surface soil above the frozen layer beneath during the spring-thaw.

"We need to focus research on transition periods such as freeze-thaw since major trace-gas fluxes may occur then, in our case approximately 80% of the annual fluxes over 3–4 weeks," said Butterbach-Bahl. "We need to better constrain sources to be able to develop sound mitigation strategies."

Grasslands cover about one-fifth of temperate land surface and are widely used as pasture. Writing in Nature, the team suggests that livestock grazing has the potential to reduce natural background nitrous oxide fluxes, as might cutting and hay-making.

"That possibility must be qualified because it is unclear what portions of temperate arid grasslands are currently grazed at levels that minimize N₂O emissions," said Del Grosso. "In addition to N₂O, the effects of grazing intensity on other factors (such as plant productivity, vegetation community structure, soil erosion, levels of soil organic matter, and methane emissions) must be taken into account."

Now Butterbach-Bahl and colleagues are establishing a new measurement network, analysing the effect of grassland systems on greenhouse gases, such as methane and carbon dioxide as well as nitrous oxide, and looking to model possible options for reducing emissions.