Emissions of methane from enteric fermentation in ruminant animals are believed to be about 23% of all anthropogenic emissions of this gas. The contribution of nitrous oxide from grazed pastures is also believed to be large, perhaps 20 to 50% of the anthropogenic emissions. The estimates for both gases are uncertain. They are based on extrapolations of "laboratory" experiments, largely chamber experiments, to the field scale. Verificarion with direct field measurements is needed. This is a difficult problem for conventional micrometeorological approaches because the animals and dung and urine patches responsible for the emissions are scattered point sources. (The animal sources are also of variable height.) The paper reports the application of a mass-balance technique for this purpose.
Gas production from sources within a test plot was calculated from measurements of the horizontal fluxes of the gas across upwind and downwind boundaries. In the applications to be described, the test plot was square, 22m x 22m, and gas concentrations were measured at 4 heights (up to 3.5m) along the length of each boundary. The concentrations were multiplied by the appropriate vector wind speeds to yield the horizontal fluxes. The difference between the integrated horizontal fluxes on downwind and upwind boundaries represented production. In one experiment, 4 cattle were grazed continuously in the test plot for 4 days. In another, 14 sheep were grazed for 6 days. Methane budgets were made for the test plot at 33- or 45-min intervals. Tests included calculation of recoveries from known gas releases, comparison with parallel measurements on discrete animals with a tracer technique, and comparisons with a conventional micrometeorological approach and a backward dispersion model. The method performed satisfactorily in all cases.
As employed by us, the mass-balance method can suffer from errors arising from the large number of gas analyses required for a flux determination and the high precision required in the concentration measurement, and it becomes unreliable when there are light winds and variable wind directions. On the other hand, it is non-disturbing, has a simple theoretical basis, is independent of atmospheric stability or the shape of the wind profile, and is apprpopriate for flux measurements in situations where conventional micrometeorological methods can not always be used. This is especially true for soil gas emissions which are notoriously spatially heterogeneous.