The world's ecosystems are currently exposed to several globally important anthropogenic stressors, which may trigger a massive wave of species' extinctions as well as a homogenization of species assemblages by introductions. Accordingly, concern has been rising that this species loss may be associated with reduced ecosystem functions and the associated services they provide to human societies. In this vein, the relationship between biodiversity and ecosystem functioning has become a prominent research topic (~5300 publications mentioned at ISI web of science until March 2015) but one that has been dominated by controlled experiments for two decades. The management of ecosystems and policy decisions, however, have to cope with multi-trophic communities in dynamic ecosystems across larger spatial and temporal scales. In response to this, BEF research has recently been increasingly integrated in field studies and mechanistic modelling approaches, which has led to novel insights on real-world consequences of biodiversity change. However, this information is scattered over different types of communities and ecosystems. Therefore, we propose an Organized Oral Session (OOS) including novel modelling concepts and empirical field studies across ecosystems that bring together (1) research approaches from different scientific fields (e.g., terrestrial versus marine research) along with (2) empirical and theoretical developments. Thereby, this OOS will explore new ways of understanding how species loss will impact ecosystem functioning under non-equilibrium conditions characterized by environmental fluctuation, spatial connectivity and disturbances. Specifically, we solicit contributions that (1) address variation in diversity across trophic levels amending the present focus of BEF research on variation in single trophic groups, especially plants, (2) analyze BEF relationships in non-equilibrium communities under external disturbances and varying environmental conditions, (3) address changes in BEF relationships in meta-communities across larger spatial scales relevant to the management and conservation of natural ecosystems, (4) study BEF relationships across long temporal scales including evolutionary signatures, and (5) integrate research from terrestrial, marine and freshwater ecosystems, thereby seeking a synthesis of patterns and processes across ecosystems. This OOS will advance the field by providing a novel conceptual and theoretical basis to obtain a general understanding of BEF in real landscapes. The session will include contribution on the effects of co-evolution and adaptation, the link between functional traits and mechanisms of coexistence in communities, and the impacts of resource supply, network structure and spatial movements (e.g. dispersion or dispersal) on biodiversity-ecosystem functioning relationships.