ALBERT

Notes: Abstract Increasing atmospheric concentrations of ‘greenhouse gases’ are expected to result in global climatic changes over the next decades. Means of evaluating and reducing greenhouse gas emissions are being sought. In this study an existing simulation model of a tropical savanna woodland grazing system was adapted to account for greenhouse gas emissions. This approach may be able to be used in identifying ways to assess and limit emissions from other rangeland, agricultural and natural ecosystems. GRASSMAN, an agricultural decision-support model, was modified to include sources, sinks and storages of greenhouse gases in the tropical and sub-tropical savanna woodlands of northern Australia. The modified model was then used to predict the changes in emissions and productivity resulting from changes in stock and burning management in a hypothetical grazing system in tropical northeastern Queensland. The sensitivity of these results to different Global Warming Potentials (GWPs) and emission definitions was then tested. Management options to reduce greenhouse gas emissions from the tropical grazing system investigated were highly sensitive to the GWPs used, and to the emission definition adopted. A recommendation to reduce emissions by changing burning management would be toreduce fire frequency if both direct and indirect GWPs of CO2, CH4, N2O, CO and NO are used in evaluating emissions, but toincrease fire frequency if only direct GWPs of CO2, CH4 and N2O are used. The ability to reduce greenhouse gas emissions from these systems by reducing stocking rates was also sensitive to the GWPs used. In heavily grazed systems, the relatively small reductions in stocking rate needed to reduce emissions significantly should also reduce the degradation of soils and vegetation, thereby improving the sustainability of these enterprises. The simulation studies indicate that it is possible to alter management to maximise beef cattle production per unit greenhouse gases or per unit methane emitted, but that this is also dependent upon the emission definition used. High ratios of liveweight gain per unit net greenhouse gas emission were found in a broadly defined band covering the entire range of stocking rates likely to be used. In contrast, high values of liveweight gain per unit ‘anthropogenic’ greenhouse gas emission were found only at very low stocking rates that are unlikely to be economically viable. These results suggest that policy initiatives to reduce greenhouse gas emissions from tropical grazing systems should be evaluated cautiously until the GWPs have been further developed and the implications of emission definitions more rigorously determined.