CONTEXT: More nutrient cycling may be achieved by using less external inputs (feed, fertilisers) and reduce losses to the environment, especially in intensive farming systems. Yet, changes in on-farm management may have unintended consequences at higher aggregation scales due to potential trade-offs. OBJECTIVE: The objective of this study was to develop a multi-indicator and multi-level model which operates at farm and regional level to evaluate scenarios for improved nitrogen cycling. METHODS: A new model, based on nitrogen flow analysis, was used to compare five scenarios with the current situation as reference. The model was applied to a case study region, the Dutch province Drenthe including typical arable, pig, poultry, and dairy farms. In the scenarios, the proportion of manure digested for biogas production, and imported amounts of synthetic fertiliser and feed into the region were varied, as single measures or in combination. RESULTS AND CONCLUSIONS: Modelling results showed that digestion of manure for biogas production reduced total regional nitrogen losses and produced renewable energy. A 20 % decrease in synthetic nitrogen fertiliser application reduced crop yields only slightly and improved the regional nitrogen use efficiency and nitrogen cycling, as manure availability in Drenthe was sufficient to meet a large proportion of the total crop nutrient requirements. Combining multiple measures was most effective in increasing nitrogen cycling (+65 %), leading to reduced greenhouse gas emissions (-49 %) and an improved net energy balance (+84 %) from agriculture in Drenthe, with the largest contribution coming from restricting feed import (resulting in a reduction of the total livestock herd in the region). However, when livestock was reduced, more synthetic nitrogen fertiliser was needed to maintain crop yields. Our study also highlighted trade-offs: positive effects on nitrogen cycling, greenhouse gas emissions and nitrogen losses coincided with reduced food production and organic matter inputs to soils, with consequences for carbon stocks. Furthermore, results for the whole region were not always representative for each farm type. SIGNIFICANCE: The results demonstrate that our systems approach, quantifying multiple indicators simulta- neously at farm and region level, can provide a better understanding of benefits and trade-offs when aiming for an agricultural system which is productive, but with reduced emissions to the environment. The developed model is generic and can be applied to evaluate alternative nitrogen cycling scenarios in other European regions with only little parameterisation needed from publicly available data.
