Trophic relationships in the eastern English Channel: how to simplify food web structure description for trophic niche determination?

Determining the organisation of natural communities as well as their ability to face environmental changes are at the core of many ecological studies. In that aim, many studies tried to describe trophic interactions between species and quantify their strength in ecological networks, the multiplicity of trophic links being assumed to promote the stability and persistence of communities. In marine ecosystems, the study of trophic relationships has extensively benefited from the development of analytical tools like stable isotopes analysis (SIA) as dietary tracers. Particularly, δ15N values are useful in estimating intra- and inter-specific trophic level variations while δ13C values will indicate variation in the source of organic matter. Here, we proposed to use stable isotope data to establish relevant functional groups in order to simplify food web description and to further determine species’ isotopic and trophic niche. SIA signatures of marine species from the eastern English Channel were used as a case study. Particulate organic matter, zooplankton, invertebrates and fish were collected during two scientific surveys conducted in October 2009 and January 2010 in the area. Trophic positions were determined using stable isotopes of δ15N and δ13C and were then aggregated into functional groups using clustering analysis coupled to a bootstrap procedure. The food web of the eastern English Channel is characterized by 10 functional groups: one group of primary producers, 4 groups of primary consumers, and 5 groups of higher-order consumers. It is composed of two main trophic pathways, one pelagic and one benthic, and forms a continuum of four tropic levels. This procedure allows us to produce a simplified description of the food web structure that could be directly used as input in trophic modelling and isotopic niche determination (i.e. SIAR mixing model and SIBER niche width calculation) for which association of species were necessary to reduce the number of potential source pools and to constrain model outputs