Family-level AMBI is valid for use in the north-eastern Atlantic but not for assessing the health of microtidal Australian estuaries

Analysis of benthic macroinvertebrate samples at a higher taxonomic level than species, e.g. family, potentially provides a more cost-effective protocol for environmental impact assessments and monitoring as it requires less time, funds and taxonomic expertise. Using the AMBI database, species ecological group scores are shown to be coherent within families. Faunal data from a wide range of environmental impact scenarios in the north-eastern Atlantic demonstrate that AMBI, calculated from mean values for families, exhibits a strong linear relationship with species-level AMBI, the correlation improving by using square-root transformed rather than untransformed abundances. In many regions of the world, however, the sensitivity of benthic macroinvertebrates to environmental perturbations is unknown, precluding the use of AMBI for environmental assessments. Yet the families are essentially the same as in the AMBI database. The utility of family-level AMBI is tested using data for four south-western Australian estuaries previously subjected to environmental quality assessments, but where only 17 species of the 144 taxa are included in the AMBI database. Although family-level AMBI scores reflect differences in environmental quality spatially and temporally within an estuary, they do not follow variations in environmental quality among estuaries. Indeed, south-western Australia estuaries are numerically dominated by families with high AMBI scores, probably due to the detrimental effects of natural accumulations of organic material in estuaries with long residence times. As taxonomic distinctness follows trends in environmental quality among estuaries, as well as temporally and spatially within a system, it provides an appropriate substitute for assessing the 'heath' of microtidal estuaries.

Making modelling count - increasing the contribution of shelf-seas community and ecosystem models to policy development and management

Marine legislation is becoming more complex and marine ecosystem-based management is specified in national and regional legislative frameworks. Shelf-seas community and ecosystem models (hereafter termed ecosystem models) are central to the delivery of ecosystem-based management, but there is limited uptake and use of model products by decision makers in Europe and the UK in comparison with other countries. In this study, the challenges to the uptake and use of ecosystem models in support of marine environmental management are assessed using the UK capability as an example. The UK has a broad capability in marine ecosystem modelling, with at least 14 different models that support management, but few examples exist of ecosystem modelling that underpin policy or management decisions. To improve understanding of policy and management issues that can be addressed using ecosystem models, a workshop was convened that brought together advisors, assessors, biologists, social scientists, economists, modellers, statisticians, policy makers, and funders. Some policy requirements were identified that can be addressed without further model development including: attribution of environmental change to underlying drivers, integration of models and observations to develop more efficient monitoring programmes, assessment of indicator performance for different management goals, and the costs and benefit of legislation. Multi-model ensembles are being developed in cases where many models exist, but model structures are very diverse making a standardised approach of combining outputs a significant challenge, and there is a need for new methodologies for describing, analysing, and visualising uncertainties. A stronger link to social and economic systems is needed to increase the range of policy-related questions that can be addressed. It is also important to improve communication between policy and modelling communities so that there is a shared understanding of the strengths and limitations of ecosystem models.