Predicting the effects of climate change on marine communities and the consequences for fisheries

Climate effects on the structure and function of marine communities have received scant attention. The few existing approaches for predicting climate effects suggest that community responses might be predicted from the responses of component populations. These approaches require a very complex understanding of ecological interactions among populations. An alternate and informative parallel process is to ask whether it is possible to make predictions about community level responses to climate that are independent of knowledge about the identity and dynamics of component populations. We propose that it is possible to make such predictions, based on knowledge of the processes that determine the size–structure of communities. We suggest that theory that relates metabolic scaling, predator–prey interactions and energy transfer in size-based food webs, allows the size–structure and productivity of communities across a range of trophic levels to be predicted, provided that predictions of the effects of climate on primary production are available. One simple application of the community-focused predictions is to ask whether predictions of the size composition and abundance of populations for alternate climate scenarios are compatible with predictions for the size composition and relative abundance of communities. More sophisticated treatments could predict the effects of climate scenarios on multiple interacting populations and compare their combined size-abundance structure and production with that predicted for the community under the same climate scenario. The main weakness of the community approach is that the methods predict abundance and production by size-class rather than taxonomic group, and society would be particularly concerned if climate driven changes had a strong effect on the relative production of fishable and non-fishable species in the community. The main strength of the community approach is that it provides widely applicable ‘null’ models for assessing the biological effects of climate change and a baseline for model comparisons.