Agroforestry (AF), “the deliberate integration of woody vegetation (trees and/or shrubs) as an upper storey on land, with pasture (consumed by animals) or an agricultural crop in the lower storey” (Mosquera-Losada et al., 2009) has risen to prominence as a land use to help address climate change and provide environmental, economic, and social benefits. The tradition of separating between science and practice in agriculture and in forestry has led to untapping opportunities for a functional use of trees in agroecosystems in order to sustain food, fibre and timber production. Trees have great potential to play an important role in the sustainable management of agricultural landscapes increasing ES generated by both traditional and innovative integrated cropping systems. ES are benefits that the human population derive directly or indirectly from ecosystem functions. Supplying the increasing demand for provisioning services (food, energy, minerals) often comes at the cost of decreasing cultural, regulating or supporting services (MEA 2005). The ES concept proved applicable for the comprehensive assessment of agroecosystems, highlighting trade-offs among services. However, it needs to be further developed to consider the relationship between the application of agricultural inputs and the resulting negative externalities which may derive from them, e.g. soil losses, greenhouse gas emissions, energy consumption, pesticide resistance, biodiversity losses, contamination of soil or water. In the EU, the area of AF systems is approx. 8.8% of the agricultural land. It is grouped in three main agroforestry categories, with specific subcategories: arable AF; livestock AF; AF with high value trees. Compared to conventional agriculture, AF may contribute significantly to carbon sequestration, increase several regulating ES, e.g. soil conservation, fertility, enhance biodiversity. Positive effects on provisioning ES (food, fibre. timber production) by AF systems seem to be globally unclear compared to conventional agriculture. In tropical regions, AF has been widespread as a traditional land use developed by subsistence farmers and as an important livelihoods option to tackle several challenges addressed in the global development agenda. In the last decades, land-sparing to facilitate the expansion of intensive agriculture has led to decreasing of tree presence on farmland. Intensification of tree crop cultivation, e.g. cocoa (Theobroma cacao L.) is associated with deforestation and loss of biodiversity and ES. Sustainability assessment of AF systems is commonly focused on a small number of indicators and ES. Until today stakeholder involvement in the assessment process has been poor; this should be increased to complete a holistic framework in which biophysical measurements are complemented by monetary and socio-cultural data. There is a need to build a shared, comprehensive, strengthened methodology for the assessment of ES in agroforestry systems to define new sustainable and efficient practices for land management. Novel research pathways should be designed with an inter-sectoral multi-actor approach to create a holistic methodological framework, such is the objective of UNDERTREES: 1.Go beyond biophysical and monetary evaluation towards a common method for ES assessment at various temporal and spatial scales; 2.Involve stakeholders to map and estimate ES; 3.Adopt multiscale approaches to inform regional, national and EU policymakers.