A farm-scale basis for predicting nitrous oxide emissions from dairy farms

The IPCC methodology was used to provide farm-scale estimates of N2O emissions from 2 dairy farms in S.W. England. Emissions were 16.5 and 15.9 kg N2O-N per ha from Farm A and B, respectively, but a large degree of uncertainty was associated with these estimates (range 5.5–50.1 and 4.6–42.5 kg N2O-N per ha for Farm A and B, respectively). The generalised assumptions and emission factors employed in this methodology can be refined at a farm scale, where more detailed information is available. Two alternative methodologies were therefore developed. The first was an improved IPCC approach using emission factors based on current literature and approaches and incorporating predictions of leached N from an existing UK model (NCYCLE). The second used the process-based model DNDC to provide estimates of N2O emission from soil. Using the improved IPCC approach, total emission was 5.1 and 8.9 kg N2O-N per ha from Farm A and B, respectively. Emission from the soil sector was decreased by 64% and 23% for Farm A and B, respectively, relative to the IPCC method. The decrease in the soil sector was largely due to a reduction in emission from grazing animals and applied animal manures. The use of NCYCLE-based estimates of nitrate leaching in the improved IPCC approach resulted in a 77% and 61% reduction in indirect emission at Farms A and B, respectively, reducing both the total emission and the proportion of the total that was due to the indirect sector. The large effect of components of the indirect sector calculations on IPCC estimates was demonstrated in a sensitivity analysis of the methodology. Data on which to base estimates of emission from indirect sources remain scarce. Preliminary measurements of indirect losses on a farm taken over 6 months confirm that indirect sources make a substantial contribution to the total emission. Estimates from the DNDC method of emission from the soil sector were larger than those of the other methods (8.2 and 7.0 kg N2O-N per ha from Farm A and B, respectively) . Use of a dynamic model such as DNDC for the estimation of emissions at a farm scale would provide a greatly improved capability for scenario testing and hence the development of mitigation strategies. However, some calibration and development of DNDC would be required before confident estimates of emissions from all sectors could be made.