Verifying inventory predictions of animal methane emissions with meteorological measurements

The paper examines the strengths and weaknesses of a range of meteorologica l flux measurement techniques that might be used to verify predictions of g reenhouse gas inventories. Recent research into emissions of methane (CH4) produced by enteric fermentation in grazing cattle and sheep is used to ill ustrate various methodologies. Quantifying this important source presents s pecial difficulties because the animals constitute moving, heterogeneously distributed, intermittent, point sources. There are two general approaches: one, from the bottom up, involves direct measurements of emissions from a known number of animals, and the other, from the top down, infers areal emi ssions of CH4 from its atmospheric signature. A mass-balance method proved successful for bottom-up verification. It permits undisturbed grazing, has a simple theoretical basis and is appropriate for flux measurements on smal l plots and where there are scattered point sources. The top-down methodolo gies include conventional flux-gradient approaches and convective and noctu rnal boundary-layer (CBL and NBL) budgeting schemes. Particular attention i s given to CBL budget methods in both differential and integral form. All t op-down methodologies require ideal weather conditions for their applicatio n, and they suffer from the scattered nature of the source, varying wind di rections and low instrument resolution. As for mass-balance, flux-gradient micrometeorological measurements were in good agreement with inventory pred ictions of CH4 production by livestock, but the standard errors associated with both methods were too large to permit detection of changes of a few pe r cent in emission rate, which might be important for inventory, regulatory or research purposes. Fluxes calculated by CBL and NBL methods were of the same order of magnitude as inventory predictions, but more improvement is needed before their use can be endorsed. Opportunities for improving the pr ecision of both bottom-up and top-down methodologies are discussed.