Stewart, A. A., Alemu, A. W., Ominski, K. H., Wilson, C. H., Tremorin, D. G., Wittenberg, K. M., Tenuta, M. and Janzen, H. H. 2014. Whole-farm greenhouse gas emissions from a backgrounding beef production system using an observation-based and model-based approach. Can. J. Anim. Sci. 94: 463-477. Backgrounding, raising weaned beef cattle in preparation for finishing in a feedlot, is a common practice in western Canadian beef production systems. The objectives of this study were: (i) to assess the whole-farm greenhouse gas (GHG) emissions from a pasture-based backgrounding system using an observation-based and model-based approach and (ii) to compare model-based estimated emissions with observation-based emissions from the key components of the farm, in order to identify the knowledge gaps that merit further study. For the observation-based approach, emissions were garnered from a multi-disciplinary field study that examined three fertility treatments applied to the pasture grazed by beef cattle: (i) no liquid hog manure application (control); (ii) split application of liquid hog manure, half applied in fall and half in spring (split) and (iii) single spring application of liquid hog manure (single). The model-based approach used a systems-based model, adapted from Intergovernmental Panel on Climate Change algorithms, to estimate annual net farm GHG emissions from the three fertility treatments and a hypothetical synthetic fertilizer treatment. Total farm emissions included methane (CH₄), nitrous oxide (N₂O) emissions from farm components and carbon dioxide (CO₂) emissions from energy use. Net farm GHG emissions using the observation-based approach ranged from 0.4 to 2.2 Mg CO₂ eq ha^-1 and from 4.2 to 6.5 kg CO₂ eq kg^-1 liveweight gain exported; the model-based approach resulted in net farm emissions ranged from 0.6 to 3.7 Mg CO₂ eq ha^-1 and from 7.0 to 12.9 kg CO₂ eq kg^-1 liveweight gain exported. Except in the control treatment, both enteric CH₄ and soil N₂O emissions were the major contributors to total farm emissions. Emissions intensity for the hypothetical synthetic fertilizer treatment (9.4 kg CO₂ eq kg^-1 liveweight gain) was lower than for the split and single scenarios. Although individual GHG emission estimates varied appreciably, trends in emissions intensity were similar between the two approaches. Efforts to reduce GHG emissions should be directed towards components such as enteric CH₄ and soil N₂O, which have larger impacts on overall system emissions.