Fatty acid patterns of Southern Ocean shelf and deep sea peracarid crustaceans and a possible food source, foraminiferans Laura W ¨ urzberg a,n , Janna Peters b , Angelika Brandt a a Biozentrum Grindel und Zoologisches Museum, Universit¨ at Hamburg, Martin-Luther-King Platz 3, 20146 Hamburg, Germany b Institut f¨ ur Hydrobiologie und Fischereiwissenschaften, Universit¨ at Hamburg, Grosse Elbstrasse 133, 22767 Hamburg, Germany article info Available online 27 May 2011 Keywords: Deep sea Fatty acid Diet composition Peracarida Foraminifera Southern Ocean benthos abstract In order to investigate the diversity of diet composition in macrobenthic peracarid crustaceans from the Antarctic shelf and deep sea, the fatty acid (FA) composition of different species belonging to the orders Isopoda, Amphipoda, Cumacea and Tanaidacea was analysed. Multivariate analyses of the FA composition confirmed general differences between the orders, but also distinct differences within these orders. To gain information on the origin of the FAs found, the potential food sources sediment, POM and foraminiferans were included in the study. Most of the analysed amphipod species displayed high 18:1(n 9)–18:1(n 7) ratios, widely used as an indicator for a carnivorous component in the diet. Cumaceans were characterised by increased phytoplankton FA markers such as 20:5(n 3) (up to 29% of total FAs), suggesting a diet based on phytodetritus. High values of the FA 20:4(n 6) were found in some munnopsid isopods (up to 21% of total FAs) and some tanaidacean species (up to 19% of total FAs). 20:4(n  6) also occurred in high proportions in some foraminiferan samples (up to 21% of total fatty acids), but not in sediment and POM, possibly indicating the ingestion of foraminiferans by some peracarid crustaceans. & 2011 Elsevier Ltd. All rights reserved. 1. Introduction Peracarid crustaceans belong to the most abundant and most diverse groups inhabiting the Southern Ocean shelf and deep sea floor (Hessler and Thistle, 1975; Wilson, 1998; Brandt et al., 2007a, b). A possible mechanism responsible for high peracarid diversity could be resource partitioning via specialisation for certain types of food. At a first glance, however, selective feeding seems unlikely as the deep sea is considered to be a relatively homogenous habitat with limited food availability (MacArthur and Pianka, 1966; Lehmann, 1976). However, one characteristic feature specific for peracarid crustaceans is very good chemor- eceptors (aesthetascs), which can be of particular help in finding preferred food items (Rehm, 2009). While most cumaceans and tanaidaceans are generally regarded as non-selective deposit-feeders (Kudinova-Pasternak, 1991; Cartes et al., 2001; Blazewicz and Ligowski, 2002), many deep sea amphipods are carnivorous or scavenging (Thurston, 1990; Dauby et al., 2001). The feeding modes of isopods are very diverse and include the whole spectrum from detritus-feeding to carnivory and necro- phagy (Wolff, 1962; Luxmoore, 1982). Also foraminiferivory has been described in some deep sea isopods (Svavarsson et al., 1993; Gujmundsson et al., 2000). However, information on the specific diets of Antarctic per- acarid crustaceans as well as their role in the food web, especially in deep sea settings, still remains scarce. This is at least partly due to low sampling effort and, in case of deep sea organisms, the difficulties of in vivo experiments. Additionally, the often empty guts or homogenous material within the guts (e.g. Iken et al., 2001; Br ¨ okeland et al., 2010) of most macrobenthic invertebrates make a clear identification of the dietary composition difficult. For small abyssal invertebrates alternative methods to identify their diet are needed to shed light on feeding strategies and food web relationships. One established approach to identify diet compo- nents is to analyse their FA composition (e.g. Sargent and Whittle, 1981; Graeve et al., 1997; Dalsgaard et al., 2003). FAs are not degraded during digestion and are accumulated in the tissues of consumers. Specific FAs are characteristic for certain food sources, and can thus be used as trophic biomarkers. Examples are the diatom markers 20:5(n  3), and 16:1(n  7) (Volkman et al., 1989; Graeve et al., 1994; Sargent et al., 1995) or 18:1(n  9), indicative of carnivorous feeding, if accompanied by low 18:1(n  7) levels (e.g. Falk-Petersen et al., 1998; Auel et al., 2002). For example, in arctic shelf invertebrates, Graeve et al. (1997) found a clear gradient from low 18:1(n  9)–18:1(n  7) ratios in suspension feeders, via predatory decapods, to higher ratios in scavenging amphipods. The 18:1(n  7) isomer is derived by chain elongation from 16:1(n  7), typical for phytoplankton (especially diatoms), and higher proportions of this FA in consumers are presumably indicative of herbivory (e.g. Graeve et al., 1997, Falk-Petersen Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/dsr2 Deep-Sea Research II 0967-0645/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr2.2011.05.013 n Corresponding author. E-mail address: laura.wuerzberg@helmholtz-muenchen.de (L. W ¨ urzberg). Deep-Sea Research II 58 (2011) 2027–2035