Emissions of greenhouse gases methane and nitrous oxide underestimated, research suggests

Kroon recently obtained her PhD degree for this much more accurate method, which also partly solves the problem of this systematic underestimation.

When it comes to greenhouse gases many of us think first of CO2. But a large proportion of global greenhouse gas emissions are actually other gases, such as N2O (nitrous oxide) and CH4 (methane). In the Netherlands the contributions of methane and nitrous oxide to the total emission of greenhouse gases are estimated to be 8% and 6% respectively. Worldwide these figures are 14% and 9%.

The emission of methane and nitrous oxide is largely the result of agricultural activities; nitrous oxide from fertilisers and methane mostly from cows. In peat pasture areas these emissions are particularly prevalent. PhD student Petra Kroon carried out measurements of methane and nitrous oxide for ECN and TU Delft on an intensively managed peat pasture, but the measuring techniques she used can also be used in other ecosystems.

Given the large contribution these two gases make to total greenhouse emissions, it is important to chart their emission levels accurately. According to Kroon, however, there are considerable disadvantages to the emission measurement technique most commonly used for these gases, the so-called 'chamber measurement' method. Put simply, this involves regularly measuring the concentration of the gases emitted from the soil and trapped in a sealed box. Rising concentrations of these gases are then analysed in order to draw conclusions about the amount of gas being given off by a given surface area. The problem with this method is that methane and nitrous oxide emissions fluctuate strongly in time and space. "If you do the same measurements ten metres away, or ten days later, the results can be totally different. To reduce uncertainty you would have to do an almost impossible number of measurements," explains Kroon. The result is a high uncertainty in the measured annual emission values: about 50%.

An improved measuring method, the so-called eddy covariance method, recently became available. The eddy covariance technique has long been in use for that other greenhouse gas, CO2, but was inapplicable for methane and nitrous oxide until recently, partly because the concentrations of methane and nitrous oxide are so much lower. Petra Kroon tested the new technique, which can measure emission values at a single point for 24 hours a day, 365 days a year and for several hectares at the same time. She concluded that when used properly, uncertainty in measured emission levels of methane and nitrous oxide could be reduced from about 50% to less than 10%.

Besides the greater uncertainty of the results, according to Kroon, the chamber measurement method also suffers from another potential problem. It involves certain assumptions, for instance that concentrations in the chamber rise in a linear fashion. "These assumptions may make analysis easier, but they don't always fit the facts, and this leads to underestimates in emission levels." Her findings are currently the subject of international discussions, and in the months to come much work will be devoted to calculating the impact they will have on national and international emission estimates.

Kroon calculated that the annual greenhouse gas emissions in a peat pasture area used for intensive dairy farming were 16000 kilo of CO2 equivalent per hectare. 70% of this was caused by methane and nitrous oxide emissions. Because the innovative eddy covariance technique yields an average across a hectare or more, it cannot reveal which parts of that area are responsible for most of the emissions; but a colleague, PhD student Arina Schrier-Uijl of the Wageningen University and Research Centre, succeeded in doing so. By making extremely careful chamber measurements she was able to show that these methane emissions arose principally from the ditches and ditch sides. This is a most surprising finding.