Emerging horizons in biodiversity and ecosystem functioning research Julia Reiss 1 , Jon R. Bridle 2 , Jose ´ M. Montoya 1, 3 and Guy Woodward 1 1 School of Biological and Chemical Sciences, Queen Mary University of London, London E1 4NS, UK 2 School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK 3 Institute of Marine Science (ICM-CSIC), Passeig Marı ´tim de la Barceloneta 37-49, 08002 Barcelona, Spain Two decades of intensive research have provided compelling evidence for a link between biodiversity and ecosystem functioning (B-EF). Whereas early B-EF research concentrated on species richness and single processes, recent studies have investigated different measures of both biodiversity and ecosystem function- ing, such as functional diversity and joint metrics of multiple processes. There is also a shift from viewing assemblages in terms of their contribution to particular processes toward placing them within a wider food web context. We review how the responses and predictors in B-EF experiments are quantified and how biodiversity effects are shaped by multitrophic interactions. Further, we discuss how B-EF metrics and food web relations could be addressed simultaneously. We conclude that addressing traits, multiple processes and food web interactions is needed to capture the mechanisms that underlie B-EF relations in natural assemblages. Why and how we measure biodiversity effects on ecosystem functioning Earth’s biota regulates numerous fluxes of energy and matter, including carbon uptake, nutrient cycling and ox- ygen production. When measured at local scales, these rates are referred to as ‘ecosystem processes.’ Attributes of the biological assemblage, in terms of the number and types of organisms and their interactions, not only deter- mine ecosystem processes but also ecosystem properties (such as resistance to invasion of exotics [1]) and how both processes and properties are sustained over time and space [1]. In combination, these three characteristics – processes, properties and their maintenance represent ecosystem functioning [2]. Although it might seem intuitive that a suite of func- tionally diverse organisms is needed to sustain ecosystem functioning, it is only in the last two decades that the relationship between biodiversity and ecosystem functioning (B-EF) has received detailed scientific interest (e.g. [3–5]). Besides its theoretical and intellectual appeal, much of B-EF research has been motivated by the alarming rates of species loss seen across the globe [6,7] and the demand to maintain the goods and services these species supply to humans [8]. The principal goals of B-EF research have been to investigate how biodiversity and ecosystem functioning are linked and to understand the mechanisms that underpin the relationship. Early B-EF studies tested whether ecosystem functioning was enhanced in species-rich versus depau- perate assemblages [9]. Since the 1990 s, a large body of work has demonstrated that biodiversity generally enhances many process rates, such as resource use or biomass production, across a wide spectrum of organisms and systems [10–12]. Biodiversity has also been shown to determine ecosystem properties, such as the ability to stay close to equilibrium in the face of environmental perturbation [1] or resistance of an assemblage to top- down control by consumers [13]. However, the evidence for positive effects of biodiversity on ecosystem function- ing is neither ubiquitous nor unequivocal [8,14]. This has stimulated considerable scientific debate [1], but the central B-EF question has now moved on from the dis- cussion of whether diversity matters to a consideration of how it matters [2]. Some recent studies have shown that the strength of biodiversity effects can increase over time (e.g. [15]). In a wider context, this finding suggests that the interpretation of B-EF relationships might be de- pendent on the level of observation. This includes temporal–spatial scales [16], but also evaluations of biotic Review Glossary Biodiversity:: refers to the extent of genetic, taxonomic and ecological diversity over all spatial and temporal scales [71]. Ecosystem function:: a synonym for ‘ecosystem process.’ Ecosystem functioning:: the joint effects of all processes that sustain an ecosystem. Some authors also use the term for ecosystem properties such as resistance to invasion [1]. Ecosystem goods and services:: are products of ecosystem functioning that are of (usually socioeconomic) value to humans. Ecosystem process:: the changes in energy and matter over time and space through biological activity, which are measured as response variables to biodiversity in B-EF experiments. These rates are also governed by the interplay of abiotic factors (physical and chemical), but the focus of B-EF research is mediation of processes (abiotic or biotic) by organisms. Examples include production of carbon, resource consumption, respiration, denitrifica- tion and nutrient uptake. Functional trait:: component of an organism’s phenotype that determines its effect on processes [19] and its response to environmental factors [2]. The term ‘trait’ should be used for the individual level only [72]. For example, body mass is a trait, biomass is not. Horizontal diversity:: the taxonomic or functional richness and evenness of entities (traits, individuals, species, genes, functional groups, etc.) within a subset of similar entities (e.g. trophic level) (after Ref. [54]). Trait:: any morphological, physiological or phenological feature measurable at the individual level [72]. Trait attribute:: a particular value or modality taken by the trait which varies both along environmental gradients and through time [72]. Vertical diversity:: a general term summarising the functional complexity of a system in the vertical (i.e. consumer–resource) dimension. Examples include food chains within the wider food web. Corresponding author: Reiss, J. (j.reiss@qmul.ac.uk). 0169-5347/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2009.03.018 505