There is an increasing demand for environmental assessments of the marine environment to include ecosystem function. However, existing schemes are predominantly based on taxonomic (i.e. structural) measures of biodiversity. Biodiversity and Ecosystem Function (BEF) relationships are suggested to provide a mechanism for converting taxonomic information into surrogates of ecosystem function. This review assesses the evidence for marine BEF relationships and their potential to be used in practical monitoring applications (i.e. operationalized).Five key requirements were identified for the practical application of BEF relationships: 1) a complete understanding of strength, direction and prevalence of marine BEF relationships, 2) an understanding of which biological components are influential within specific BEF relationships, 3) the biodiversity of the selected biological components can be measured easily, 4) detail which ecological mechanisms are the most important for generating marine BEF relationships, e.g. identity effects or complementarity, and 5) establish what proportion of the overall functional variance is explained by biodiversity, and hence BEF relationships.Many positive and some negative BEF relationships were found within the literature, although many reproduced poorly the natural species richness, trophic structures or multiple functions of real ecosystems. Null relationships were also reported. The consistency of the positive and negative relationships was often low that compromised the ability to generalize BEF relationships and confident application of BEF within marine monitoring. Equally, some biological components and functions have received little or no investigation.Expert judgement was used to attribute biological components using spatial extent, presence and functional rate criteria. This approach highlighted the main contributing biological components to the ecosystem functions, and that many of the particularly influential components were found to have received the least amount of research attention.The need for biodiversity to be measureable (requirement 3) is possible for most biological components although difficult within the functionally important microbes. 3Identity effects underpinned most marine BEF relationships (requirement 4). As such, processes that translated structural biodiversity measures into functional diversity were found to generate better BEF relationships.The analysis of the contribution made by biodiversity, over abiotic influences, to the total expression of a particular ecosystem function was rarely measured or considered (requirement 5). Hence it is not possible to determine the overall importance of BEF relationships within the total ecosystem functioning observed. In the few studies where abiotic factors had been considered, it was clear that these modified BEF relationships and have their own direct influence on functional rate.Based on the five requirements, the information required for immediate 'operationalization' of BEF relationships within marin...

a b s t r a c tThe European Union's Marine Strategy Framework Directive (MSFD) seeks to achieve, for all European seas, "Good Environmental Status" (GEnS), by 2020. Ecological models are currently one of the strongest approaches used to predicting and understanding the consequences of anthropogenic and climate-driven changes in the natural environment. We assess the most commonly used capabilities of the modelling community to provide information about indicators outlined in the MSFD, particularly on biodiversity, food webs, non-indigenous species and seafloor integrity descriptors. We built a catalogue of models and their derived indicators to assess which models were able to demonstrate: (1) the linkages between indicators and ecosystem structure and function and (2) the impact of pressures on ecosystem state through indicators. Our survey identified 44 ecological models being implemented in Europe, with a high prevalence of those that focus on links between hydrodynamics and biogeochemistry, followed by end-to-end, species distribution/habitat suitability, bio-optical (remote sensing) and multispecies models. Approximately 200 indicators could be derived from these models, the majority of which were biomass and physical/hydrological/chemical indicators. Biodiversity and food webs descriptors, with ∼49% and ∼43% respectively, were better addressed in the reviewed modelling approaches than the non-indigenous species (0.3%) and sea floor integrity (∼8%) descriptors. Out of 12 criteria and 21 MSFD indicators relevant to the abovementioned descriptors, currently only three indicators were not addressed by the 44 models reviewed. Modelling approaches showed also the potential to inform on the complex, integrative ecosystem dimensions while addressing ecosystem fundamental properties, such as interactions between structural components and ecosystems services provided, despite the fact that they are not part of the MSFD indicators set. The cataloguing of models and their derived indicators presented in this study, aim at helping the planning and integration of policies like the MSFD which require the assessment of all European Seas in relation to their ecosystem status and pressures associated and the establishment of environmental targets (through the use of indicators) to achieve GEnS by 2020.

No abstract