Unraveling the Multiple Bottom-Up Supplies of an Antarctic Nearshore Benthic Community ⁎ ⁎ L

Home , Laevilacunaria, Phyllophora antarctica

Progress in Oceanography xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Progress in Oceanography

journal homepage: www.elsevier.com/locate/pocean

Unraveling the multiple bottom-up supplies of an Antarctic nearshore benthic community ⁎ ⁎ L. Zentenoa,c, , L. Cárdenasb,c, N. Valdiviaa,c, I. Gómeza,c, J. Höfera,c, I. Garridoa,c,d, LM. Pardoa,c, a Instituto de Ciencias Marinas y Limnológicas, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile b Instituto de Ciencias Ambientales y Evolutivas, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile c Centro de Investigación Dinámica de Ecosistemas Marinos de Altas Latitudes (IDEAL), Valdivia-Punta Arenas, Chile d Département de Biologie & Québec-Océan, La Faculté des Sciences et de Génie, Université Laval, Quebec, Canada.

ARTICLE INFO ABSTRACT

Keywords: Disentangling the bottom-up controls of natural ecosystems is key to understanding the capacity of local com- Maxwell Bay munities to resist natural and anthropogenic disturbances. Here, we used carbon and nitrogen stable isotope Trophic ecology ratios with a Bayesian multiple source mixing model to trace diverse food sources supporting the benthic trophic Ecosystem stability network in Fildes Bay (South Shetland Island, Western Antarctic Peninsula). Individuals of 16 species of con- Food web sumers and ﬁve potential food sources (e.g. inter- and subtidal macroalgae, suspended and sinking particulate Polar organic matter, and particulate organic matter from sediment) were collected during January and February Isotopic analysis Western Antarctic Peninsula 2017. The results showed that benthic organisms of Fildes Bay assimilate a broad range of available organic Benthos matter: most of the energy channeled to upper trophic consumers comes from organic matter in the surface sediment, whereas energy moving among lower trophic consumers comes largely from macroalgae and pelagic primary food sources. Overall, our evidence indicates that the present-day nearshore benthic community of Fildes Bay relies on diﬀerent primary food sources, channeling bottom-up supplies through multiple pathways, which leads to highly stable systems in the face of current scenarios of global change.

1. Introduction Antarctic ecosystems (Fischer and Wiencke,1992; Amsler et al., 1995; Corbisier et al., 2004), providing an important proportion of carbon to Pulses of phytoplankton have been long recognized as the main Antarctic benthic consumers (e.g. Reichardt, 1987; Dunton, 2001; primary food source sustaining Antarctic nearshore benthic commu- Corbisier et al., 2004). The availability of subsidies of organic matter nities (e.g. Barnes and Clarke, 1995). Nonetheless, several recent stu- for benthic consumers depends, however, on the proportion of sea dies claim that the role of pelagic primary food sources has been surface that is covered by ice. For instance, Norkko et al. (2007) found overestimated, concluding that multiple primary food sources sustain that consumption of detritus is higher in areas covered by ice, whereas these communities (Corbisier et al., 2004; Norkko et al., 2007; Gillies the consumption of fresh macroalgae is higher in areas free of ice. et al., 2013). The existence of different primary food sources implies Consequently, fresh and detrital matter play a relevant trophic role, that energy may flow through multiple pathways across the food web, increasing the availability of primary production in coastal areas of the enhancing its stability in the face of disturbances (e.g. Lawton, 1994). Antarctic Peninsula (Smith et al., 2006; Norkko et al., 2007) and in- Accordingly, unraveling the multiple bottom-up supplies may be key to fluencing the connections between the water column and benthic pro- understanding the mechanisms underpinning the stability of food webs cesses (Smith et al., 2006). off coastal Antarctica. Stable carbon and nitrogen isotope analysis is a well-established Few studies have documented the importance of primary food method for tracing the dietary sources and trophic levels of organisms sources for consumers in coastal areas off the Antarctic Peninsula (but (DeNiro and Epstein, 1981; Tieszen et al., 1983). Carbon isotopic values see Kaehler et al., 2000; Dunton, 2001; Corbisier et al., 2004; Jacob (δ13C), on the other hand, help trace trends in marine habitat use be- et al., 2006). Thus, the fate of primary food sources and their role in tween two extremes of a spatial gradient, namely inshore/benthic and sustaining benthic Antarctic communities remain uncertain. Benthic offshore/pelagic habitats (Fry and Sherr, 1984; Wada et al., 1991; micro- and macroalgae may be regarded as primary food sources in Cherel and Hobson, 2007). This gradient in marine habitat use occurs

⁎ Corresponding authors at: Instituto de Ciencias Marinas y Limnológicas, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile. E-mail addresses: [email protected] (L. Zenteno), [email protected] (L. Pardo). https://doi.org/10.1016/j.pocean.2018.10.016

Please cite this article as: Zenteno, L., Progress in Oceanography, https://doi.org/10.1016/j.pocean.2018.10.016 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx because primary producers in nearshore benthic areas are relatively and ascidians; the intertidal zone is dominated by amphipods, littorinid more 13C-enriched than those in offshore and pelagic regions. Thus, in snails, littorinid snails, crustose algae, and several small, opportunistic areas where δ13C values differ strongly among primary producers, it is macroalgae (mostly filamentous green algae) (Valdivia et al., 2014; possible to detect the relative influence that different sources have on Segovia-Rivera and Valdivia, 2016). consumer diets (Fry and Sherr, 1984; Hobson et al., 1995; Post, 2002). ff Nearshore benthic communities in Fildes Bay o King George Island 2.2. Primary food sources (South Shetland Islands) constitute a model system for assessing the ff relative importance of di erent primary food sources for consumers. To evaluate the contribution of different energy sources, samples of Overall, this ecosystem is widely heterogeneous, with trophically linked representative benthic and pelagic food sources were collected in aus- organisms (Ortiz et al., 2016), and the combination of oceanic and fresh tral summer (January–February 2017) in Fildes Bay (Fig. 1 and water contributions along with shifting sea ice distributions enhance Table 1). Benthic primary food sources included subtidal and intertidal variability in sea surface temperature, salinity, and primary pro- macroalgae and surface sediment organic matter (OM). Pelagic primary ductivity (Khim and Yoon, 2003; Moreno-Pino et al., 2016). Moreover, food sources included suspended particulate organic matter (POM) and short-term simulations in this ecosystem suggest that the dominance of sinking POM. subtidal macroalgae provides greater resistance to disturbances than Subtidal macroalgae were collected by SCUBA diving along ten found in analogous communities dominated by kelp forests and sea- transects; samples were taken using a portable underwater venture- grass meadows (Ortiz et al., 2016). All these features hint at the ex- suction device equipped with a 20-um mesh size at station C (Fig. 1), a ff fl istence of di erent bottom-up supplies owing through multiple path- rocky shore on the northern margin of Fildes Bay (depth < 30 m). ways and sustaining the heterogeneous nearshore benthic community Intertidal macroalgae were hand-picked at Station B (Fig. 1). of Fildes Bay. For surface sediment OM, five cores were collected from the upper The objective of this study is to explore the trophic links between 3 cm layer by SCUBA diving, separated by at least 4 m. These samples primary food sources (suspended and sinking particulate organic were collected at subtidal station 5 (Fig. 1), along the 20 m isobaths. fi matter, super cial sediment organic matter and macroalgae) and con- Macroalgae and surface sediment OM were washed with distilled water δ13 δ15 sumers, using stable isotope ratios ( C and N) and Bayesian sta- and frozen in a cold-room below −20 °C. tistical approaches. We tested the general hypotheses that lower and A 10-L go-flo bottle was used to sample suspended POM at 5-m upper level consumers in Fildes Bay rely on multiple energy pathways, depth, whereas sediment traps were used to collect sinking POM at 50- with a predominance of organic material derived from fresh (macro- m depth (Fig. 1). Both suspended and sinking POM were collected by algae) and surface sediment organic matter and, to a lesser extent, filtering seawater through a pre-combusted (4 h, 550 °C), 47-mm, GF/F pelagic primary food sources. glass fiber filter (0.7-μm pore size), until the fl ow of water decreased to a slow drip. Then the filters were frozen at −20 °C until the next 2. Material and methods treatment.

2.1. Study area 2.3. Consumers

This study was conducted in Fildes Bay, off the southwestern end of Invertebrates and fishes were collected from intertidal and subtidal King George Island, in the South Shetland Islands (Fig. 1). Local hy- sites (depth < 30 m) off the north coast of Fildes Bay (Fig. 1). Subtidal drography is influenced by an inflow of oceanic water from Bransfield invertebrates were collected by SCUBA diving, following the under- Strait (Khim et al., 1997) and local-origin water from melting glaciers water venture-suction explained above, whereas intertidal organisms and river runoff (Hong et al., 1991; Khim et al., 1997; Khim and Yoon, (N. concinna) were hand-picked during the diurnal low tide at Station B 2003). Despite extreme environmental conditions and the effect of (Fig. 1). Invertebrates were kept alive for 24 h in the laboratory to allow winter sea surface ice, the benthic community of Fildes Bay is char- evacuation of gut contents, then sorted and frozen (−20 °C) until acterized by highly abundant and diverse macroinvertebrates and analysis. Subtidal fishes were collected using lines fishing with different macroalge (Valdivia et al., 2014; Aghmich et al., 2016), Specifically, baits (e. g. pork, chicken, clams), and then individually sealed in plastic the subtidal zone is dominated by brown and red microalgae, along bags and frozen (−20 °C). Invertebrates were sorted and identified to with several species of fishes, scavengers, grazers, sponges, bryozoans, the lowest taxonomic classification possible. Fishes were identified to

Fig. 1. Locations of sites where food sources and consumers were sampled for stable isotopes. The solid circles show consumer sites (A = predatory ﬁshes, predatory invertebrates; B = intertidal grazer; C = subtidal grazer, scavenger amphipods, herbivore amphipods, suspensivores) and the triangles denote food source sites (1 = suspended POM; 2 = suspended POM and sinking POM; 3 = subtidal macroalgae; 4 = intertidal macroalgae; 5 = surface sediment OM).

2 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx

Table 1 δ15N and δ13C values (mean ± SE) of taxa collected from Fildes Bay.

Taxonomic group or species n δ13C(‰) δ15N(‰) Feeding strategies/Trophic group References

Sources Sinking POM 6 −21.8 ± 1.0 2.6 ± 0.8 Suspended POM 2 −19.8 ± 0.1 2.4 ± 0.0 Surface sediment POM 5 −21.0 ± 0.6 3.8 ± 0.6 Macroalgae 25 −17.8 ± 2.1 3.8 ± 1.2 Adenocystis utricularis 5 −14.3 ± 0.3 4.0 ± 1.6 Iridaea cordata 5 −20.3 ± 0.5 3.5 ± 0.2 Palmaria decipiens 5 −18.5 ± 0.5 2.4 ± 0.3 Monostroma hariotii 5 −18.5 ± 0.9 3.9 ± 0.4 Gigartina skottsbergii 5 −17.9 ± 0.8 5.5 ± 1.0

Nemertea Parborlasia corrugatus 5 −21.4 ± 0.7 6.5 ± 0.5 Deposit Feeder/Predatory invertebrate Gibson (1983)

Mollusca Bivalvia Laternula elliptica 5 −20.3 ± 0.3 3.8 ± 0.2 Suspensivore Brockington (2001) Kidderia bicolor 5 −21.7 ± 0.3 4.0 ± 0.2 Suspensivore Martens (1885)

Gastropoda Nacella concinna 5 −13.9 ± 0.7 6.9 ± 0.2 Grazer (Intertidal) Clarke et al. (2004) Laevilacunaria antarctica 4 −16.5 ± 0.2 5.5 ± 0.2 Grazer (Subtidal) Iken (1999) Margarella antarctica 5 −18.1 ± 0.3 5.9 ± 0.3 Grazer (Subtidal) Gutt and Schickan (1998)

Echinodermata Asteroidea Odontaster validus 5 −20.6 ± 1.4 7.6 ± 0.9 Deposit Feeder/Predatory Dearborn (1977)

Arthropoda Amphipoda Bovallia gigantea 5 −21.8 ± 0.4 6.6 ± 0.4 Carnivore/Predatory invertebrate Bone (1972) Cheirimedon femoratus 6 −14.7 ± 0.6 4.4 ± 0.6 Omnivore/Scavenger amphipod Seefeldt et al. (2017) Gondogeneia antarctica 3 −19.4 ± 0.2 5.9 ± 0.3 Omnivore/Scavenger amphipod Aumack et al. (2017) Djerboa furcipes 5 −16.9 ± 0.3 4.8 ± 0.2 Hervibore amphipod Nyssen et al. (2005) Prostebbingia gracilis 5 −18.7 ± 0.7 4.0 ± 0.4 Hervibore amphipod Zamzow et al. (2010) Paracerodocus miersii 5 −21.5 ± 0.6 4.8 ± 0.2 Scavenger/Benthic consumer Obermüller et al. (2013)

Chordata Fish Nothotenia rosii 16 −21.6 ± 1.2 10.0 ± 0.5 Omnivore/Predatory Fish Barrera-Oro and Winter (2008) Nothotenia coriiceps 13 −20.0 ± 0.8 10.9 ± 0.8 Omnivore/Predatory Fish Daniels (1982) Harpagifer antarcticus 8 −16.7 ± 0.7 11.1 ± 0.6 Omnivore/Predatory Fish Blankley (1982) the species level using DNA barcoding (Hebert and Gregory, 2005). undesirable variability in δ13C(Lorrain et al., 2003). Thus, each sample Genomic DNA was extracted from ﬁsh using a commercial kit (Omega was divided into two aliquots, and calcium carbonate was removed by

E.Z.N.A.), and the mitochondrial gene cytochrome c oxidase subunit I soaking subsamples in 0.5 M hydrochloric acid (HCl) until the CO2 was (COI) was amplified using the primers described by Folmer et al. completely depleted (Newsome et al., 2006). Since this HCl treatment (1994). PCR products were sent for sequencing to Macrogen, Korea adversely affects the δ15N values (Bunn et al., 1995), the remaining (www.macrogen.com), and phylogenetic reconstructions were per- aliquot was left untreated for δ15N analyses. formed using the neighbor‐joining and maximum likelihood algorithms Approximately 10–15 mg of POM with filter, 0.8–1.0 mg of algae, in Mega7 (Kumar et al., 2016). Finally, sequences of fishes used in this 0.3 mg of mollusk mantle and fish muscle, and 0.3–0.5 mg of entire study were deposited into the GenBank database. crustaceans were weighed into sterilized tin capsules and analyzed for dual stable carbon and nitrogen isotopes. Each pre-treated sample was 2.4. Stable isotope analysis combusted in a Thermo elemental analyzer integrated with an isotope ratio mass spectrometer. All the consumers and benthic primary food fi In the laboratory, filters with pelagic primary food sources, surface sources were analyzed at the Ponti cia Universidad Católica de Chile fi sediment OM, and macroalgae samples were dried at 60 °C for 36–48 h (Flash EA200 IRMS Delta Series, Thermo Scienti c, Germany), and the in an oven and then acidified for the removal of carbonate by wetting pelagic primary food sources were analyzed at the University of the filters with 1 mL 0.5 M hydrochloric acid (HCl). Consumers were California Davis, using a PDZ Europa ANCA-GSL elemental analyzer – thawed with distilled water, dried in an oven at 60 °C for 36–48 h, and interfaced to a PDZ Europa 20 20 isotope ratio mass spectrometer ground to a fine powder with a mortar and pestle. Soft tissues were (Sercon Ltd., Cheshire, UK). δ collected from sea stars, fishes, nemerteans, and mollusks. For small Stable isotope ratios were reported in standard ( ) notation using invertebrates (e.g., amphipods and some mollusks with body the following equation: length < 100 mm), the entire organisms were analyzed, and roughly δX =−[(R /R ) 1]× 1000 ten whole organisms constituted a single replicate sample. sample standard (1) Lipids were extracted from fish and invertebrate samples with a where X is either 13Cor15N, R is the ratio of heavy to light isotopes, and chloroform/methanol (2:1) solution (Bligh and Dyer, 1959). This is the standard is either PDB for 13C or atmospheric N2 for 15N. The re- because, compared with other molecules, lipids are depleted in 13C producibility, as determined by the standard deviation of the internal (DeNiro and Epstein, 1978) and their concentration in tissues may vary standards, was ± 0.18‰ and 0.30‰ for δ15N and δ13C, respectively. between and within species. On the other hand, high concentrations of inorganic carbon in some invertebrate exoskeletons may cause

3 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx

2.5. Trophic position Fig. 2), we tested normality and homoscedasticity of residuals with the Lilliefors and Levene tests, respectively. For multiple comparisons, the Nitrogen stable isotope ratios at the base of the food-web are needed data sets did not meet the normality and homoscedasticity require- to estimate the trophic position (TP) of the consumers. Because δ15N ments; thus, the non-parametrical Kruskal-Wallis test was used. In ad- values of primary producers are strongly influenced by spatial temporal dition, we used a paired-samples t-test to compare the means between parameters (Post, 2002; Iken et al., 2010; Miller and Page, 2012), our trophic groups and primary food sources considering the assumptions of baseline TPs were calculated using δ15N values of Laternula elliptica,a the t-test. These analyses were carried out with PASW Statistics (version suspensivore bivalve and primary consumer (i.e. assumed to occupy 17.0 for Windows, SPSS). Significance was assumed at 0.05. TP = 2) of phytoplankton (Brockington, 2001), and Laevilacunaria To estimate the relative contribution of primary food sources to antarctica, a herbivorous gastropod that consumes mainly macroalgae each species studied (Parnell et al., 2010), we used the upgrade func- (Iken, 1999). tion simmr from the package Stable Isotope Mixing Models within R. This Since outputs from the 1- and 2-source TP models (see Post, 2002) upgraded version of the SIAR package contains a slightly more so- were similar, the following 1-source TP model was applied for all in- phisticated mixing model that allows realistic assessments and dis- dividuals: cerning source contributions to a mixture (Parnell, 2016). Models were built over four Markov chains with 10,000 steps per chain and a burn-in TP[(NN)/ΔN]TP=−δδδ15 15 15 + consumer consumer base base (2) of 1000 iterations. Data within each group fitted a normal distribution, where the mean δ15NofL. elliptica was used as the baseline and re- as required by SIAR (Parnell et al., 2010). The fractionation factor (Δ) is a critical aspect in reconstructing presented a TLbase of 2. A value of 2.3‰ was used for trophic enrichment, Δδ15N(McCutchan et al., 2003). diets and trophic web structures (Post, 2002). This is especially true for species inhabiting Antarctic ecosystems because few controlled feeding studies exist for these organisms, often making it particularly challen- 2.6. Data analyses ging to discern metabolic issues of isotopic incorporation. On the other hand, trophic position is one of the main factors influencing Δδ15N After calculating the isotopic mean values and standard deviations values (DeNiro and Epstein, 1981; Minagawa and Wada, 1984). Thus, (SD) of the consumers and primary food or carbon sources (Table 1 and

Fig. 2. Bivariate stable isotope ratios of consumers and their potential food sources from Fildes Bay, Antarctic Peninsula.

4 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx we compared our own empirical observations with Δ calculated by Table 1), with significant differences between intertidal and subtidal meta-analyses (McCutchan et al., 2003) to obtain appropriate mea- macroalgae (paired t test; p < 0.000). Macroalgae were significantly surements of quantitative interactions between consumers and sources. more enriched in 13C than pelagic and surface sediment OM primary For this, we used linear mixing models with a fractionation factor food sources (paired t test; p < 0.001 and p < 0.01, respectively). The multiplied by the number of trophic positions (obtained by equation δ13C values of sinking POM ranged from −23.3‰ to −20.5‰, with a (2)) between the consumer and the basal resources to simulate the mean of −21.8‰; this was significantly lower than the suspended POM proportions of each food source in the consumer diet (Phillips et al., values (paired t test; p < 0.01). The δ13C values of surface sediment 2014). We assessed the differences in δ15N values between phyto- OM were quite uniform within the group (standard deviations < 1%), plankton (POM at a 5-m depth) and the primary consumer, L. elliptica, and no difference was found among surface sediment OM and pelagic to obtain the trophic shifts for δ15N values of Fildes Bay baseline. Our primary food sources (t = −0.630, p > 0.05) (Fig. 2 and Table 1). empirically observed Δ15N (mean ± SD = −1.3 ± 0.2‰) fell within The mean δ15N values of primary food sources ranged from 2.5‰ the fractionation factor range prescribed for lower trophic consumers (pelagic primary food sources) to −3.8‰ (surface sediment OM) (mean ± SD = −1.4 ± 0.2‰)byMcCutchan et al. (2003). Conse- (Fig. 2 and Table 1). The pelagic primary food sources were distinctly quently, we used their suggested Δ15N and propagated the Δ according more 15N-depleted (mean ± SD = 2.5 ± 0.7‰) than the macroalgae the trophic position obtained by equation (1) (mean ± SD; (paired t test; p < 0.001) and surface sediment OM (paired t test; TP2 = 1.4 ± 0.2‰; TP3 = 2.8 ± 0.4‰; TP4 = 4.2 ± 0.6‰; p < 0.001). Macroalgae and surface sediment OM (including micro- TP5 = 5.6 ± 0.8‰). organisms and detritus) had more similar signatures (3.5 ± 0.9‰ and The mean Δ13C obtained from the differences in δ13C values be- 3.8 ± 0.6‰, respectively); no significant differences were detected tween potential primary food sources and consumers was smaller than between groups. Δ15N values among trophic levels. For that reason, we used the general Most δ15N-depleted values were observed in the pelagic primary Δ13C (mean ± SD = 0.5 ± 0.2‰) obtained by McCutchan et al. food sources (mean ± SD = 2.5 ± 0.7‰) and were significantly (2003) for different species. lower than macroalgae and surface sediment OM values. In addition, Finally, to analyze the overall contribution of the main energy we detected a slight depth-related variation for pelagic food sources, pathways sustaining the nearshore benthic Antarctic communities with more 15N-enriched in sinking POM than suspended POM, and also (Kaehler et al., 2000; Dunton, 2001; Norkko et al., 2007; Gillies et al., a sharp variation in δ15N values between intertidal (4.0 ± 1.6‰) and 2013), we grouped the primary food sources collected in: (1) pelagic subtidal macroalgae (3.4 ± 1.0‰). primary food sources (suspended and sinking POM), (2) surface sedi- Overall, these significant differences between groups support the ment organic matter, and (3) macroalgae (intertidal and subtidal premise of multiple energy sources for macrobenthic consumers in macroalgae). Moreover, for this analysis, species were assigned trophic Fildes Bay. groups according to the literature (Table 1), their trophic position, and A total of 98 consumers were sampled in Fildes Bay, covering a wide a hierarchical cluster analysis using the mean values of each species range of taxonomic groups (Table 1 and Fig. 2). The consumers ex- (Fig. 3). hibited significant differences for both δ13C and δ15N values (Kruskal- Wallis test; δ13C: χ2 = 85.044, df = 15, p < 0.01; δ15N: χ2 = 92.186, 3. Results df = 15, p < 0.01). Diet was used to sort species into seven different feeding groups: 3.1. δ13C and δ15N of food sources and consumers predatory fishes, predatory invertebrates, scavenger amphipods, subtidal grazers, intertidal grazers, herbivore amphipods, and suspensi- The composition of both carbon and nitrogen stable isotopes dif- vores (Table 1 and Fig. 2). A cluster analysis confirmed this classifica- fered significantly among sources (Kruskal-Wallis test; δ13C: tion (Fig. 3). Scavenger amphipods, herbivore amphipods, and subtidal χ2 = 28.327, df = 4, p < 0.01; δ15N: χ2 = 10.338, df = 4, p < 0.05). grazers displayed more isotopic variability than the other groups, which For macroalgae, the mean δ13C values varied widely from −20.8‰ showed quite similar isotopic signatures within each feeding group (Iridaea cordata)to−13.9‰ (Adenocystis utricularis) (Fig. 2 and (paired t-test; p > 0.01).

Fig. 3. Dendrogram based on the mean δ13C and δ15N values of trophic group from the benthic community oﬀ Fildes Bay (South Shetland Island, Antarctic Peninsula).

5 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx

Fig. 4. Summary of energy pathways in the Fildes Bay benthic community, according to SIAR mixing model (simrr).

The δ13C values of all consumers were distributed over a wide proportion of their diets from pelagic food sources. range, from −21.9‰ (B. gigantea)to−13.9‰ (N. concinna). The scavenger amphipods showed the widest range of δ13C values 4. Discussion (−18.1‰ ± 3.2), followed by predatory fishes (−20.0‰ ± 2.1), and predatory invertebrates (−21.4‰ ± 1.0). In terms of δ15N signatures, The overall evidence reported here indicates that, currently, the the consumers ranged from 3.7‰ (L. elliptica) to 11.1‰ (H. antarcticus) nearshore benthic community of Fildes Bay assimilates different pri- (Table 1 and Fig. 2). The predatory fishes showed the widest range of mary food sources, revealing a system sustained by multiple pathways δ15N (10.4‰ ± 0.9), followed by the predatory invertebrates of organic matter, but with variable degrees of benthopelagic coupling (7.0‰ ± 0.8). depending on the consumers’ trophic positions. These results are con- Finally, we found that predatory fishes and the predatory sea star sistent with previous studies of shallow benthic Antarctic communities occupied the upper trophic level (TP5-TP4), whereas scavenger am- (e.g. Kaehler et al., 2000; Dunton, 2001; Norkko et al., 2007; Gillies phipods, predatory invertebrates, grazers, and suspensivores occupied et al., 2013). Although some primary food sources were not assessed in the range of TP3-TP2 (Fig. 4). this study, our results certainly reflect the main energy pathways be- lieved to sustain the macrobenthic marine communities off Antarctica, namely pelagic primary food sources, surface sediment OM, and mac- 3.2. Relative contribution of food sources to consumers roalgae primary food sources (Kaehler et al., 2000; Dunton, 2001; Norkko et al., 2007; Gillies et al., 2013). Assimilation of δ13C and δ15N derived from pelagic and benthic trophic pathways differed among consumers in Fildes Bay (Fig. 5). Overall, The SIAR output showed that, for most consumers analyzed, 4.1. Surface sediment organic matter (OM) as a food source for upper contributions of surface sediment OM and macroalgae tended to prevail trophic levels over pelagic primary food sources (Fig. 4). The predatory fishes, N. coriiceps and N. rossii, derived a high proportion of their diets (> 50%) The dominant energy channel in the upper trophic level was the from surface sediment OM (Fig. 5a and b). Conversely, H. antarcticus surface sediment OM and, to a lesser degree, macroalgae and pelagic derived over 80% of its isotopic signature from the intertidal macro- primary food sources. Similar patterns have been observed in other algae, A. utricularis (Fig. 5c). The predatory invertebrate, O. validus, regions off Antarctica (Norkko et al., 2007), suggesting that sediment showed a similar proportion of food sources, although surface sediment OM is an important source of detritus to higher trophic levels, con- OM was slightly larger (Fig. 5d). Similarly, surface sediment OM played tributing strongly to bottom-up controls and dampening seasonality. a major role in the diets of P. corrugatus (Fig. 5e) and B. gigantea The mineralization and utilization of C by microbial communities de- (Fig. 5g). pletes the δ13C values of surface sediment OM (Hollander and Smith, Scavenger amphipods showed a marked variation in the proportion 2001). Similarly, it is also highly likely that bacterial degradation in- of primary food sources making up their diets. Whereas G. antarctica creases δ15N values (Macko and Estep, 1984). Surface sediments from (Fig. 5i) and P. miersii (Fig. 5h) mainly assimilated pelagic sources, Ch. the Antarctic Peninsula are characterized by high nutrient contents and femoratus mainly assimilated intertidal macroalgae (Fig. 5f). a large presence of diatoms (e.g. Chaetoceros), Bacteroidetes, and At lower trophic levels, grazers and herbivore amphipods re- Cercozoan (Learman et al., 2016), which can, in turn, benefit upper ceived > 60% of their diet from macroalgal material. On the contrary, consumers via trophic mediation. For instance, Norderhaug et al. as expected, the suspensivores (L. elliptica and K. bicolor) derived a high (2003) experimentally showed low C:N ratios in decomposed

6 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx

Fig. 5. Dietary composition of the species according to the SIAR mixing model (simmr). Boxplots represent 95% credible interval of primary food sources assimilation for each consumer. The centerline in the box is the median of all solutions, and the box is drawn around the 25th and 75th quartiles, thereby representing 50% of the solutions. The δ13C and δ15N values of consumers were corrected for the discrimination factor so that they could be compared to the source values. macroalgae due to decreased phytotoxin concentrations. Similarly, confirming the high diversity of feeding modes in this taxon (Dauby Norkko et al. (2004) reported that fragmentation and bacterial de- et al., 2001; Amsler et al., 2014). Specifically, the broad range of δ13C gradation facilitates the entrance of accumulated Phyllophora antarctica values recorded here reflects the selection of different primary food drifts into the food web in McMurdo Sound, but with a slow degrada- sources based on scavenger amphipod foraging locations (Amsler et al., tion rate due to the cold temperatures (Brouwer, 1996). These results 2014; Aumack et al., 2017). Paracerodocus miersii and G. antarctica are demonstrating that high nutrient concentrations and microbial ac- displayed depleted δ13C values, indicating the importance of pelagic tivity in sediments constitute a relevant component of benthic food prey items, but macroalgae also comprised a significant portion of the webs in these low-temperature ecosystems. G. Antarctica diet. Actually, digestive tract analyses revealed that dia- Stable isotope ratios of some predatory invertebrates (P. corrugatus, toms contributed 50% of the G. antarctica diet (Aumack et al., 2017). B. gigantea) and predatory fishes (N. coriiceps, N. rossii) were clearly Consistent with our results, previous isotopic measurements also re- coupled with the surface sediment OM. The overlaps observed among vealed less enriched δ13C values for this species, associated with the surface sediment OM, pelagic primary food sources, and macroalgae consumption of the subtidal macroalgae, P. decipiens (Dunton, 2001; suggest a mixing of detrital and fresh organic matter derived from Aumack et al., 2017). Interestingly, our results showed a strong link phytoplankton and macroalgae material. Thus, it is likely that these between C. femoratus, a scavenger amphipod, and A. utricularis,an predatory invertebrates and fishes are consuming prey that feed on a abundant macroalgae in the Fildes Bay intertidal zone (Aghmich et al., mix of fresh and detrital material accumulating in the surface sediment. 2016), underscoring the opportunistic feeding behavior of this Ant- Fishes are usually more mobile than their prey, which allows them arctic amphipod (Bregazzi, 1972). Furthermore, this result agrees with to integrate different energy pathways originating in different foods controlled-diet experiments, in which A. utricularis was more palatable sources. Accordingly, the high contribution of surface sediment OM to C. femoratus than G. skottsbergii and P. decipiens (Nuñez-Pons et al., (> 50%) in the N. rossii diet may be related to the wider home range 2012). observed for this species (Burchett, 1983; Barrera-Oro and Winter, 2008). On the other hand, for N. coriiceps, a cryptic species (Casaux et al., 1990), the integration of different sources of OM could be asso- 4.3. The role of macroalgal and pelagic primary food sources in lower ciated with an efficient detritivore pathway. The macroalga, A. utricu- trophic levels laris, contributes heavily to the H. antarcticus diet, suggesting a lower integration of different energy sources likely related to the species' At the lower trophic levels, the dominant energy pathways were specialist feeding behavior and narrow home range: under rocks on the macroalgae and pelagic food sources. Benthic suspensivore organisms δ13 seafloor of the Antarctic Peninsula (Casaux, 1998). were clearly separated from the other consumers on the C axis, showing a stable isotope ratio similar to those of pelagic food sources and, to a lesser extent, surface sediment OM. This is consistent with 4.2. The polyphagous nature of scavenging amphipods previous records (e.g. Barnes and Clarke, 1995; Ahn, 1997; Gili et al., 2001). Stable isotope ratios varied widely for scavenger amphipods, Regional variability in food available for benthic suspensivores is

7 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx closely tied to oceanographic processes (Barnes and Clarke, 1995). For 4.5. Conclusion instance, Ahn (1997) found that benthic diatoms are the dominant food supply for L. elliptica in summer, whilst smaller-sized plankton fractions In summary, our work shows that the primary food sources analyzed and resuspended sediment dominate in winter. The high δ13C and δ15N can flow through multiple pathways across the benthic food web of values of pelagic primary food sources found here were among the Fildes Bay, with upper trophic consumers receiving most energy from highest recorded for the Antarctic Peninsula (e.g. Wada et al., 1987), surface sediment OM, and lower trophic consumers supported mainly most likely a consequence of the massive diatom blooms (chlor- by macroalgae and pelagic primary food sources. In addition, this is the − ophyll > 30 mg m 3) and high primary production rates (e.g. first attempt to apply quantitative Bayesian methods to the study of the − − 6.28 gC d 1 m 2) recorded within Fildes Bay during January 2017 nearshore benthic community in Fildes Bay, and it certainly advances (Höfer et al., in this volume). Furthermore, we found significant dif- knowledge of the mechanisms that determine the resistance of local ferences between the δ13C values of suspended and sinking POM, which communities to environmental disturbances. could be explained by different successional stages of the phytoplankton bloom (Ostrom et al., 1997), with benthic suspensivores Acknowledgements heavily assimilating the 13C-depleted sinking POM. On the other hand, contributions of surface sediment OM in suspensivore diets suggest The authors acknowledge financial support from the grants INACH persistent resuspension of organic material by winds or iceberg RG 20-16, Fondecyt 1161129, and FONDAP IDEAL 15150003. Thanks scouring in Fildes Bay (Brandini and Rebello, 1994). to Paulina Brüning, Mateo Cáceres, Hans Bartsch, and Bastian Garrido As in other studies in Antarctic waters (Corbisier et al., 2004; for field assistance: both diving and laboratory work. Logistical support Norkko et al., 2007; Gillies et al., 2013; Aumack et al., 2017), 13C en- provided by the Instituto Antártico Chileno (INACh) and the staff of richment was higher in macroalgae than in the other analyzed sources, Base Julio Escudero Station, King George Island, is greatly appreciated. although some overlapping occurred with suspended POM. This overlap The studies in the Antarctic were carried out under permission granted could be explained by the incorporation of the 13C depleted from epi- by INACh in accordance with the Protocol on Environmental Protection phyte/endophyte species (e.g. green algae filament, phaeophytes, and to the Antarctic Treaty. The study did not involve Antarctic Specially diatoms) growing in macroalgal tissues (e.g. Aumack et al., 2017). Protected Areas (ASPAs), nor sampling of protected or endangered Stable isotope ratios of gastropod grazers and herbivore amphipods species. were closely aligned with the enriched δ13C values of macroalgal primary food sources, confirming that most of the carbon utilized by these Appendix A. Supplementary material organisms is derived from benthic production. Aumack et al. (2017) reported patterns of quantitative interactions in macroalgae and Ant- Supplementary data to this article can be found online at https:// arctic amphipods at different sites near Anvers Island (Antarctic Pe- doi.org/10.1016/j.pocean.2018.10.016. ninsula) using isotopic measurements and gut contents. This study showed that macroalgal material (both filamentous and multiseriate) References comprise a significant portion of the amphipod assemblage’s diet, but also that benthic diatoms are a significant dietary element for most Aghmich, A., Taboada, S., Toll, L., Ballesteros, M., 2016. First assessment of the rocky amphipods. Nevertheless, the accidental ingestion of frustules probably intertidal communities of Fildes Bay, King George Island (South Shetland Islands, Antarctica). Polar Biol. 39, 189–198. inflates these percentages (Dauby et al., 2001). In any case, the overall Ahn, I.Y., 1997. Feeding ecology of the Antarctic lamellibranch Laternula elliptica evidence indicates a prevalence of macroalgae over surface sediment (Laternulidae) in Marian Cove and vicinity, King George Island, during one austral summer. In: Battaglia, B., Valencia, J., Walton, D.W.H. (Eds.), Antarctic and pelagic sources as primary food sources for gastropod grazers and Communities: Species, Structure and Survival. Cambridge University Press, herbivore amphipods. Nonetheless, on a seasonal scale, the melting/ Cambridge, pp. 142–151. freezing regime may considerably influence biomass production and Amsler, C.D., McClintock, J.B., Baker, B.J., 2014. Chemical mediation of mutualistic interactions between macroalgae and mesograzers structure unique coastal commu- species composition in macroalgal assemblages (Clark et al., 2013), nities along the Western Antarctic Peninsula. J. Phycol. 50, 1–10. leading to qualitative and quantitative differences in individual inter- Amsler, C.D., Rowley, R.J., Laur, D.R., Quetin, L.B., Ross, R.M., 1995. Vertical distribu- actions of benthic food webs. Hence, further studies are necessary to tion of Antarctic peninsular macroalgae. Cover, biomass and species composition. Phycologia 34, 424–430. elucidate shifts in the trophic connections of these heterogeneous as- Aumack, C.F., Lowe, A.T., Amsler, C.D., Amsler, M.O., McClintock, J.B., Baker, B.J., 2017. semblages of the Antarctic Peninsula. Gut content, fatty acid, and stable isotope analyses reveal dietary sources of macroalgal-associated amphipods along the western Antarctic Peninsula. Polar Biol. 40, 1371–1384. Barnes, D.K.A., Clarke, A., 1995. Seasonality of feeding-activity in Antarctic suspension – 4.4. A cautionary note and future directions feeders. Polar Biol. 15, 335 340. Barrera-Oro, E.R., Winter, D.J., 2008. Age composition and feeding ecology of early ju- venile Notothenia rossii (Pisces, Nototheniidae) at Potter Cove, South Shetland Islands, Our assessment of the assimilation of primary food sources by Antarctica. Antarct. Sci. 20, 339–341. Blankley, W.O., 1982. Feeding ecology of three inshore fish species at Marion Island consumers is based on the assumption that the analysis of a relatively (Southern Ocean). South Afr. J. Zool. 17, 164–170. small number of food sources can be extrapolated to build general- Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. izations of trophic pathways. Thus, a number of confounding factors Can. J. Biochem. Physiol. 37, 911–917. Bone, D.G., 1972. Aspects of the biology of the Antarctic amphipod Bovallia gigantea should be considered. Firstly, this approach can lead to biased con- Pfeffer at Signy Island, South Orkney Islands. Br. Antarctic Survey Bull. 27, 105–122. clusions because of a low sampling effort of other potential primary Brandini, F.P., Rebello, J., 1994. Wind field effect on hydrography and chlorophyll dy- food sources (e.g. the microphytobenthos, benthic diatoms, allochtho- namics in the coastal pelagial of Admiralty Bay, King George Island, Antarctica. Antarct. Sci. 6, 433–442. nous material, and filamentous algae). Secondly, spatial variability in Bregazzi, P.K., 1972. Life cycles and seasonal movements of Cheirimedon femoratus advection processes, or increased deposition, can affect the assimilation (Pfeffer) and Tryphosella kergueleni (Miers) (Crustacea: Amphipoda). Br. Antarctic Survey Bull. 30, 1–34. of food sources by consumers, especially in hydrographically complex Brockington, S., 2001. The seasonal energetics of the Antarctic bivalve Laternula elliptica areas (Tu et al., 2015). Finally, turnover times and the fractionation (King and Broderip) at Rothera Point, Adelaide Island. Polar Biol. 24, 523–530. factor may affect variations of isotopic composition among consumers Brouwer, P.E.M., 1996. Decomposition in situ of the sublittoral Antarctic macroalga Desmarestia anceps Montagne. Polar Biol. 16, 129–137. (Tieszen et al., 1983; Hobson et al., 1995; Post, 2002). Consequently, Bunn, S.E., Loneragan, N.R., Kempster, M.A., 1995. Effects of acidon stable-isotope ratios additional work combining isotopic composition, grazing experiments, of C ans N in Penaeid shrimp and seagrass. Implications for food-web studies using – and fatty acid analyses is needed to tease out the complex trophic links multiple stable isotopes. Limnol. Oceanogr. 40, 622 625. Burchett, M.S., 1983. Food, feeding and behavior of Notothenia rossii nearshore at South mobilizing primary food sources to consumers (Layman et al., 2012). Georgia. Br. Antarctic Survey Bull. 45–51. Casaux, R., 1998. The contrasting diet of Harpagifer antarcticus (Notothenioidei,

8 L. Zenteno et al. Progress in Oceanography xxx (xxxx) xxx–xxx

Harpagiferidae) at two localities of the South Shetland Islands, Antarctica. Polar Biol. 5500 km of antarctic surface sediment implicates organic matter as a driver of 19, 283–285. benthic community structure. Front. Microbiol. 7, 11. Casaux, R.J., Mazzotta, A.S., Barreraoro, E.R., 1990. Seasonal aspects of the biology and Lorrain, A., Savoye, N., Chauvaud, L., Paulet, Y.M., Naulet, N., 2003. Decarbonation and diet of nearshore Nototheniid fish at Potter Cove, South Shetland Islands, Antarctica. preservation method for the analysis of organic C and N contents and stable isotope Polar Biol. 11, 63–72. ratios of low carbonated suspended particulate material. Anal. Chim. Acta 491, Cherel, Y., Hobson, K.A., 2007. Geographical variation in carbon stable isotope signatures 125–133. of marine predators: a tool to investigate their foraging areas in the Southern Ocean. Macko, S.A., Estep, M.L.F., 1984. Microbial alteration of stable nitrogen and carbon Mar. Ecol. Prog. Ser. 329, 281–287. isotopic compositions of organic matter. Org. Geochem. 6, 787–790. Clark, J.S., Poore, A.G.B., Ralph, P.J., Doblin, M.A., 2013. Potential for adaptation in Martens, E., 1885. Vorläufige Mittheilungen über die Molluskenfauna von Süd Georgien. response to thermal stress in an intertidal macroalga. J. Phycol. 49, 630–639. Sitzungsberichte der Gesellschaft Naturforschender Freunde zu Berlin 89–94. Clarke, A., Prothero-Thomas, E., Beaumont, J.C., Chapman, A.L., Brey, T., 2004. Growth McCutchan, J.H., Lewis, W.M., Kendall, C., McGrath, C.C., 2003. Variation in trophic shift in the limpet Nacella concinna from contrasting sites in Antarctica. Polar Biol. 28, for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102, 378–390. 62–71. Miller, R.J., Page, H.M., 2012. Kelp as a trophic resource for marine suspension feeders: A Corbisier, T.N., Petti, M.A.V., Skowronski, R.S.P., Brito, T.A.S., 2004. Trophic relation- review of isotope-based evidence. Mar. Biol. 159, 1391–1402. ships in the nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C Minagawa, M., Wada, E., 1984. Stepwise enrichment of δ15N along food-chains. further stable isotope analysis. Polar Biol. 27, 75–82. evidence and the relation between δ N-15 and animal age. Geochim. Cosmochim. Daniels, R.A., 1982. Feeding ecology of some fishes of the Antarctic Peninsula. Fish. Bull. Acta 48, 1135–1140. 80, 575–588. Moreno-Pino, M., De la Iglesia, R., Valdivia, N., Henriquez-Castilo, C., Galan, A., Diez, B., Dauby, P., Scailteur, Y., De Broyer, C., 2001. Trophic diversity within the eastern Weddell Trefault, N., 2016. Variation in coastal Antarctic microbial community composition Sea amphipod community. Hydrobiologia 443, 69–86. at sub-mesoscale: Spatial distance or environmental filtering? FEMS Microbiol. Dearborn, J.H., 1977. Food and feeding characteristics of antarctic asteroids and Ecol. 92. ophiuroids. In: Llano, G.A. (Ed.), Adaptations Within Antarctic Ecosystems. Gulf Newsome, S.D., Koch, P.L., Etnier, M.A., Aurioles-Gambao, D., 2006. Using carbon and Publishing Company, Houston, pp. 293–326. nitrogen isotope values to investigate maternal strategies in northeast Pacific otariids. Deniro, M.J., Epstein, S., 1978. Influence of diet on distribution of carbon isotopes in Mar. Mammal Sci. 22, 556–572. animals. Geochim. Cosmochim. Acta 42, 495–506. Norderhaug, K.M., Fredriksen, S., Nygaard, K., 2003. Trophic importance of Laminaria Deniro, M.J., Epstein, S., 1981. Influence of diet on the distribution of nitrogen isotopes in hyperborea to kelp forest consumers and the importance of bacterial degradation to animals. Geochim. Cosmochim. Acta 45, 341–351. food quality. Mar. Ecol. Prog. Ser. 255, 135–144. Dunton, K.H., 2001. δ15N and δ13C measurements of Antarctic peninsula fauna: trophic Norkko, A., Thrush, S.F., Cummings, V.J., Funnell, G.A., Schwarz, A.M., Andrew, N.L., relationships and assimilation of benthic seaweeds. Am. Zool. 41, 99–112. Hawes, I., 2004. Ecological role of Phyllophora antarctica drift accumulations in Fischer, G., Wiencke, C., 1992. Stable carbon isotope composition, depth distribution and coastal soft sediment communities of McMurdo Sound, Antarctica. Polar Biol. 27, fate of macroalgae from the Antarctic Peninsula region. Polar Biol. 12, 341–348. 482–494. Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R., 1994. DNA primers for am- Norkko, A., Thrush, S.F., Cummings, V.J., Gibbs, M.M., Andrew, N.L., Norkko, J., plification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan Schwarz, A.M., 2007. Trophic structure of coastal antarctic food webs associated with invertebrates. Mol. Mar. Biol. Biotechnol. 3, 294–299. changes in sea ice and food supply. Ecology 88, 2810–2820. Fry, B., Sherr, E.B., 1984. δ13C Measurements as indicators of carbon flow in marine and Nuñez-Pons, L., Rodriguez-Arias, M., Gomez-Garreta, A., Ribera-Siguan, A., Avila, C., freshwater ecosystems. Contrib. Mar. Sci. 27, 13–47. 2012. Feeding deterrency in Antarctic marine organisms: Bioassays with the omni- Gibson, R., 1983. Antarctic nemerteans: The Anatomy, Distribution and Biology of vore amphipod Cheirimedon femoratus. Mar. Ecol. Prog. Ser. 462, 163–174. Parborlasia corrugatus (McIntosh, 1876) (Heteronemertea, Lineidae). Biol. Antarctic Nyssen, F., Brey, T., Dauby, P., Graeve, M., 2005. Trophic position of Antarctic amphi- Seas XIV. pods. Enhanced analysis by a 2-dimensional biomarker assay. Mar. Ecol. Prog. Ser. Gili, J.M., Coma, R., Orejas, C., Lopez-Gonzalez, P.J., Zabala, M., 2001. Are Antarctic 300, 135–145. suspension-feeding communities different from those elsewhere in the world? Polar Obermüller, B.E., Truebano, M., Peck, L.S., Eastman, J.T., Morley, S.A., 2013. Reduced Biol. 24, 473–485. seasonality in elemental CHN composition of Antarctic marine benthic predators and Gillies, C.L., Stark, J.S., Johnstone, G.J., Smith, S.D.A., 2013. Establishing a food web scavengers. J. Exp. Mar. Biol. Ecol. 446, 328–333. model for coastal Antarctic benthic communities: A case study from the Vestfold Ortiz, M., Berrios, F., Gonzalez, J., Rodriguez-Zaragoza, F., Gomez, I., 2016. Macroscopic Hills. Mar. Ecol. Prog. Ser. 478, 27–41. network properties and short-term dynamic simulations in coastal ecological systems Gutt, J., Schickan, T., 1998. Epibiotic relationships in the Antarctic benthos. Antarct. Sci. at Fildes Bay (King George Island, Antarctica). Ecol. Complexity 28, 145–157. 10, 398–405. Ostrom, N.E., Macko, S.A., Deibel, D., Thompson, R.J., 1997. Seasonal variation in the Hebert, P.D.N., Gregory, T.R., 2005. The promise of DNA barcoding for taxonomy. Syst. stable carbon and nitrogen isotope biogeochemistry of a coastal cold ocean en- Biol. 54, 852–859. vironment. Geochim. Cosmochim. Acta 61, 2929–2942. Hobson, K.A., Ambrose, W.G., Renaud, P.E., 1995. Sources of primary production, Parnell, A.C., 2016. simmr: A stable isotope mixing model. R package version 0.3. benthic-pelagic coupling, and trophic relationships within the Northeast Water Parnell, A.C., Inger, R., Bearhop, S., Jackson, A.L., 2010. Source partitioning using stable Polynya: Insights from δ13C and δ15N analysis. Mar. Ecol. Prog. Ser. 128, 1–10. isotopes: Coping with too much variation. Plos One 5. Höfer, J., Giesecke, R., Hopwood, M. J., Carrera, V., Alarcón, E., González, H.E., (in this Phillips, D.L., Inger, R., Bearhop, S., Jackson, A.L., Moore, J.W., Parnell, A.C., Semmens, volume). The role of water column stability and wind mixing in the production/ B.X., Ward, E.J., 2014. Best practices for use of stable isotope mixing models in food- export dynamics of two bays in the Western Antarctic Peninsula. Prog. Oceanogr. web studies. Can. J. Zool. 92, 823–835. Hollander, D.J., Smith, M.A., 2001. Microbially mediated carbon cycling as a control on Post, D.M., 2002. Using stable isotopes to estimate trophic position: Models, methods, and the δ13C of sedimentary carbon in eutrophic Lake Mendota (USA): New models for assumptions. Ecology 83, 703–718. interpreting isotopic excursions in the sedimentary record. Geochim. Cosmochim. Reichardt, W.T., 1987. Burial of Antarctic macroalgal debris in bioturbated deep-sea Acta 65, 4321–4337. sediments. Deep-Sea Res Part A – Oceanogr. Res. Papers 34, 1761–1770. Hong, G.H., Kim, D.Y., Chung, H., Pae, S., 1991. Coastal and inshore water interaction, Seefeldt, M.A., Campana, G.L., Deregibus, D., Quartino, M.L., Abele, D., Tollrian, R., Held, mixing and primary productivity in the Bransfield Strait, Antarctica during austral C., 2017. Different feeding strategies in Antarctic scavenging amphipods and their summer, 1989/1990. Polar Res. 2, 43–59. implications for colonisation success in times of retreating glaciers. Front. Zool. Iken, K., 1999. Feeding ecology of the Antarctic herbivorous gastropod Laevilacunaria 14, 15. antarctica Martens. J. Exp. Mar. Biol. Ecol. 236, 133–148. Segovia-Rivera, V., Valdivia, N., 2016. Independent effects of grazing and tide pool ha- Iken, K., Bluhm, B., Dunton, K., 2010. Benthic food-web structure under differing water bitats on the early colonisation of an intertidal community on western Antarctic mass properties in the southern Chukchi Sea. Deep-Sea Res. Part Ii-Top. Stud. Peninsula. Rev. Chil. Hist. Nat. 89. Oceanogr. 57, 71–85. Smith, C.R., Mincks, S., DeMaster, D.J., 2006. A synthesis of bentho-pelagic coupling on Jacob, U., Brey, T., Fetzer, I., Kaehler, S., Mintenbeck, K., Dunton, K., Beyer, K., Struck, the Antarctic shelf: Food banks, ecosystem inertia and global climate change. Deep- U., Pakhomov, E.A., Arntz, W.E., 2006. Towards the trophic structure of the Bouvet Sea Res. Part Ii-Top. Stud. Oceanogr. 53, 875–894. Island marine ecosystem. Polar Biol. 29, 106–113. Tieszen, L.L., Boutton, T.W., Tesdahl, K.G., Slade, N.A., 1983. Fractionation and turnover Kaehler, S., Pakhomov, E.A., McQuaid, C.D., 2000. Trophic structure of the marine food of stable carbon isotopes in animal tissues: implications for δ13C analysis of diet. web at the Prince Edward Islands (Southern Ocean) determined by δ13C and δ15N Oecologia 57, 32–37. analysis. Mar. Ecol. Prog. Ser. 208, 13–20. Tu, K.L., Blanchard, A.L., Iken, K., Horstmann-Dehn, L., 2015. Small-scale spatial varia- Khim, B.K., Park, B.K., Yoon, H.I., 1997. Oxygen isotopic compositions of seawater in the bility in benthic food webs in the northeastern Chukchi Sea. Mar. Ecol. Prog. Ser. 528, Maxwell Bay of King George Island, West Antarctica. Geosci. J. 1, 115–121. 19–37. Khim, B.K., Yoon, H.I., 2003. Postglacial marine environmental changes in Maxwell Bay, Valdivia, N., Diaz, M.J., Holtheuer, J., Garrido, I., Huovinen, P., Gomez, I., 2014. Up, King George Island, West Antarctica. Polar Res. 22, 341–353. down, and all around: Scale dependent spatial variation in rocky-shore communities Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: Molecular Evolutionary Genetics of Fildes Peninsula, King George Island, Antarctica. Plos One 9. Analysis Version 7.0 for Bigger Datasets. Mol. Biol. Evol. 33, 1870–1874. Wada, E., Mizutani, H., Minagawa, M., 1991. The use of stable isotopes for food web Lawton, J.H., 1994. What do species do in ecosystems? Oikos 71, 367–374. analysis. Crit. Rev. Food Sci. Nutr. 30, 361–371. Layman, C.A., Araujo, M.S., Boucek, R., Hammerschlag-Peyer, C.M., Harrison, E., Jud, Wada, E., Terazaki, M., Kabaya, Y., Nemoto, T., 1987. 15N and 13C abundances in the Z.R., Matich, P., Rosenblatt, A.E., Vaudo, J.J., Yeager, L.A., Post, D.M., Bearhop, S., Antartic Ocean with emphasis on the biogeochemical structure of the food web. Deep 2012. Applying stable isotopes to examine food-web structure: An overview of ana- Sea Res. Part A Oceanogr. Res. Pap. 34, 829–841. lytical tools. Biol. Rev. 87, 545–562. Zamzow, J.P., Amsler, C.D., McClintock, J.B., Baker, B.J., 2010. Habitat choice and Learman, D.R., Henson, M.W., Thrash, J.C., Temperton, B., Brannock, P.M., Santos, S.R., predator avoidance by Antarctic amphipods: the roles of algal chemistry and mor- Mahon, A.R., Halanych, K.M., 2016. Biogeochemical and microbial variation across phology. Mar. Ecol. Prog. Ser. 400, 155–163.