A hundred years of food web aggregation: the case study of the Barents Sea

The difference between various food web aggregation approaches is discussed. We review the history of food web aggregation (e.g. using trait combinations), lumping data (e.g. temporal food webs) and the biological relevance of food web resolution (e.g. larval stages). The effects of aggregation approaches on food web structure (e.g. the number of links between species at intermediate trophic levels) and the needs for standards (e.g. testing predictions) are briefly discussed. Finally, a case study of the Barents Sea food web is used for illustration, where we compare the outcome of biological and mathematical approaches to food web aggregation. Future research directions are also suggested and discussed, including the role of experimental food web research and better relating interaction networks to co-occurrence networks.