Nitrate leaching from outdoor pig production is a long-standing environmental problem for surface and groundwater pollution. In this study, the effects of inclusion of poplar trees in paddocks for lactating sows on nitrogen (N) balances were studied for an organic pig farm in Denmark. Vegetation conditions, soil water and nitrate dynamics were measured in poplar and grass zones of paddocks belonging to main treatments: access to trees (AT), no access to trees (NAT) and a control without trees (NT), during the hydrological year April 2015 to April 2016. Soil water drainage for each zone, simulated by two simulation models (CoupModel and Daisy), was used to estimate nitrate leaching from the zones in each paddock. N balances (input minus output) for the treatments were computed and compared. The results showed that, in terms of annual water balance and regardless of treatment, simulated evapotranspiration of poplar was 560-569 and 489-498 mm for CoupModel and Daisy, respectively, and corresponding evapotranspiration of grass-clover was 250 and 400 mm, against precipitation of 1076 mm. Simulated drainage below the root zone varied as 620-723 mm for Daisy and 568-958 mm for CoupModel, the higher end of the latter being probably overestimated. Annual nitrate leaching ranged from 32 kg N ha in the poplar zone of NAT up to 289 kg N ha in the control grass zone of NT. The poplar zone showed significantly lower nitrate leaching, by 75-80%, compared to the grass zone. For the control NT treatment, nitrate leaching was approximately 50% higher in the grass zone closest to the hut compared to the grass zone further away. NT treatment also had the largest surface N balance of 468 kg N ha compared to 436 and 397 kg N ha for AT and NAT, respectively. When N losses by leaching and volatilisation were included, soil N balances were 118, 157 and 113 kg N ha for AT, NAT and NT, respectively. Overall, the two simulation models were found useful tools for analyses of water balance for complex agroforestry systems. The findings collectively suggest that it is possible to decrease nitrate leaching from outdoor pig production on sandy soils by inclusion of poplar trees. Additional measures are nevertheless needed to reduce N losses on a mean area basis in paddocks with 20% tree cover.