LAUSR

Abstract Inputs of nitrogen to agricultural production systems are necessary to produce food, feed and fibre, but nitrogen (N) losses from those systems represent a waste of a resource and a threat to both the environment and human health. The nitrogen use efficiency (NUE) of an agricultural production system can be seen as an indicator of the balance between benefits and costs of primary food, feed and fibre production. Here, we used modelling to follow the fate of the virgin N input to different production systems (ruminant and granivore meat, dairy, arable), and to estimate their NUE at the system scale. We defined two ruminant meat production systems, depending on whether the land places constraints on farming practices. The other production systems were dairy, granivore and arable production on land without constraints. Two geographic regions were considered: Northern and Southern Europe. Measures to improve NUE were identified and allocated to Low, Medium and High ambition groups, with Low equating to the current situation in Europe for production systems that are broadly following good agricultural practice. The NUE of the production systems was similar to or higher in Southern than Northern Europe, with the maximum technical NUEs if all available measures are implemented were for North and South Europe, respectively, 82% and 92% for arable systems, 71% and 80% for granivores, 50% and 36% for ruminant meat production on constrained land, 53% and 55% for dairy production on unconstrained land and 46% and 62% for ruminant meat production on unconstrained land. The values for NUE found here tend to be higher than reported elsewhere, possibly due to the accounting for long-term residual effects of fertiliser and manure in our method. The greatest increase in NUE with the progressive implementation of higher ambition measures was in unconstrained granivore systems and the least was in constrained ruminant meat systems, reflecting the lower initial NUE of granivore systems and the larger number of measures applicable to confined livestock systems. Our work supports use of NUE as an indicator of the temporal trend in the costs and benefits of existing agricultural production systems, but highlights problems associated with its use as a sustainability criteria for livestock production systems. For arable systems, we consider well-founded the NUE value of 90% above which there is a high risk of soil N depletion, provided many measures to increase NUE are employed. For systems employing fewer measures, we suggest a value of 70% would be more appropriate. We conclude that while it is feasible to calculate the NUE of livestock production systems, the additional complexity required reduces its value as an indicator for benchmarking sustainability in practical agriculture.