Contents lists available at ScienceDirect

Continental Shelf Research

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

Abundance and size patterns of echinoderms in coastal soft-bottoms atDeception Island (South Shetland Islands, Antarctica)

Carlos Angulo-Precklera,⁎, Fernando Tuyab, Conxita Avilaa

a Department of Evolutionary Biology, Ecology, and Environmental Sciences, University of Barcelona & Biodiversity Research Institute (IrBIO), Av.Diagonal 643, 08028 Barcelona, Catalonia, Spainb IU-ECOAQUA, Campus Tafira, Universidad de Las Palmas de Gran Canaria, Canary Islands, 35017 Tafira, Las Palmas, Spain

A R T I C L E I N F O

Keywords:Benthic invertebratesBrittle starSea urchinSea starSedimentationDensity

A B S T R A C T

Deception Island is an active volcano in Antarctic waters under high sedimentation regimes, which may affectthe abundance and structure of soft-bottom assemblages. During the summer of 2012–2013, a survey of theshallow water soft-bottom assemblages of Deception Island was carried out to examine patterns of abundanceand size structure of the three dominant echinoderms (Ophionotus victoriae, Sterechinus neumayeri andOdontaster validus) at 8 locations encompassing a gradient in proximity from the open ocean, including twodepths (5 vs. 15 m) per location. Abundance patterns of the three species varied with depth; organisms weretypically more abundant at 15 relative to 5 m depth. Our results partially supported the hypothesis thatechinoderms from locations adjacent to the open ocean present larger abundances. Body sizes variedsignificantly among locations and depths for the three species and some places presented a density-sizepattern. High sedimentation rates, combined with low ice-related disturbance, may be the reason behind thelarge abundances of echinoderms found in this waters.

1. Introduction

The distribution of benthic organisms on the Antarctic continentalshelf is highly variable across scales of spatial variation (Arntz et al.,1994; Barnes and Conlan, 2007; Gray, 2001; Gutt, 2000; Teixidó et al.,2004). This patchiness is probably related to the effects of icedisturbance at current timescales, through the actions of ice scouringand anchor ice formation (Brown et al., 2004; Dayton et al., 1969; Guttand Piepenburg, 2003; Smale, 2007), as well as at evolutionarytimescales, following the cyclical advance and retreat of the Antarcticice sheet (Clarke et al., 1992; Clarke et al., 2004). These mechanisms ofdisturbance, in combination with other factors such as sedimentation,food availability and currents flow, have shaped a highly heterogeneousseascape. Coastal benthic communities of Antarctica are characterizedby a high abundance and diversity of marine invertebrates (Dell, 1972;Arntz et al., 1994, 1997), and the dominance of suspension feedingorganisms (Jażdżeski et al., 1986; Ramos, 1999). The most abundantinvertebrates include soft corals, tunicates, and, especially, spongesand echinoderms (Dearborn, 1972; Dayton et al., 1974; White, 1984;Cattaneo-Vietti et al., 1999; Clarke and Johnston, 2003). Particularly,Antarctic echinoderms are important contributors to the overallbenthic biomass and production, playing a significant role in the global

marine carbon cycle (Lebrato et al., 2010). Furthermore, echinodermseasily add up to 45% of the overall abundance of the benthos (Moyaet al., 2003), even reaching 98% of the total epifaunal community at 8–12 m depth in Whaler´s Bay, Deception Island (Barnes et al., 2008).

Deception Island (DI), one of the South Shetland islands locatedwest of the Antarctic Peninsula, is part of an environment with lowfluctuations in temperature and salinity (Arntz et al., 1994). Comparedto nearby islands, DI is clearly depauperated in terms of fauna (Lovelland Trego, 2003). In fact, DI is a volcanically active island that has noterupted since 1970 (Elderfield, 1972); many taxa are still poorlyrepresented, and the colonizers are mainly those with planktotrophiclarvae (Barnes et al., 2008). The walls of the volcano are covered bypyroclastic ashes and lapilli-tuff, which are snow-covered from Junethrough November. The transportation of sediment (lapilli-tuff) bywinds, as well as the melting water forming into Port Foster, results ina relatively flat area of sand and silt across the caldera. The sedimentsof DI are similar to those of other South Shetland islands (Sáiz-Salinaset al., 1997), yet the benthic community may be different from thoseelsewhere. In 1964–65, the two most abundant epibenthic organismswere: Ophionotus victoriae and Sterechinus neumayeri; during theeruption of December 1967, however, their populations suffered from amass mortality (Gallardo and Castillo, 1968). Between 1967 and 1981,

http://dx.doi.org/10.1016/j.csr.2016.12.010Received 12 June 2016; Received in revised form 18 December 2016; Accepted 19 December 2016

⁎ Corresponding author.E-mail address: carlospreckler@hotmail.com (C. Angulo-Preckler).

Continental Shelf Research 137 (2017) 131–141

Available online 04 January 20170278-4343/ © 2017 Elsevier Ltd. All rights reserved.

MARK

the epibenthic megafaunal community recovered to its pre-eruptionconditions, with the return of Ophionotus victoriae, Sterechinusneumayeri and Odontaster validus (Retamal et al., 1982).Subsequent studies confirmed that epibenthic assemblages remainedstable, with the continued dominance of O. victoriae, and remarkablelow abundances of sessile organisms associated to soft-bottoms(Cranmer et al., 2003; Lovell and Trego, 2003; Barnes et al., 2008).According to the literature, suspension-feeders are very scarce in DI,where ascidians characterize the filter-feeding community (Arnaudet al., 1998).

Three species of echinoderms, generally occurring together, reachhigh values of abundance at DI: the ophiuroid Ophionotus victoriaeBell 1902 (Cranmer et al., 2003), the echinoid Sterechinus neumayeriand the asteroid Odontaster validus (Koehler, 1912). O. victoriae is anopportunistic generalist with high diet plasticity, feeding on 13different phyla including krill (Dearborn, 1977; Fratt and Dearborn,1984; Pawson, 1994) but mainly detritivore. It is extremely commonalong the Antarctic Peninsula. This species inhabits a variety ofsubstrates at depths ranging from 5 to 1266 m (Madsen, 1967). S.neumayeri is the most abundant sea urchin in shallow Antarctic watersand occurs on gravel and rocks down to 500 m water depth (Brey andGutt, 1991). They feed mainly on benthic diatoms as well on red algaeand seal faeces (Dearborn, 1965; Pearse and Giese, 1966; McClintock,1994). The sea star O. validus is abundant and widely distributed inmost of the shallow benthic environments surrounding the Antarcticcontinent (Dearborn, 1965; McClintock et al., 1988), being mostcommon between 15 and 200 m depth (Dearborn, 1977). O. validushas an opportunistic behavior and a variable diet; it may act as asuspension feeder, as an herbivore, as a scavenger, and as an activepredator (Dearborn, 1977; Dayton, 1974; Pearse, 1964; Peckham,1964).

Although the benthic community of DI has been studied on severaloccasions, it is not known if the abundance and distribution ofepibenthic megafauna vary spatially. The aim of this study was toexamine patterns of abundance, distribution and size structure of threedominant benthic echinoderms (Ophionotus victoriae, Sterechinusneumayeri and Odontaster validus) at shallow coastal waters. Thesampling was designed to cover two depths (5 and 15 m) at each of 8locations along the bay, which encompasses a gradient in proximityfrom the open ocean. We hypothesized that there would be differencesin the abundance, distribution and size structure of each speciesbetween both depths and among locations, due to varying environ-mental conditions with varying proximity from the entrance of the bay.

2. Material and methods

2.1. Study area

During January 2013, the abundances and size structure of thethree dominant soft-bottom echinoderms (Ophionotus victoriae,Sterechinus neumayeri and Odontaster validus) were determined,through SCUBA diving, at eight locations along Port Foster, insidethe caldera of DI, South Shetland Island, Antarctica (Fig. 1, Table 1).This area is typically covered by ice, for at least seven months everyyear. The water temperature at 15 m depth ranges from 0° to 2 °C, andthe primary production is highly seasonal (Smith et al., 2003). Thewater column is stratified in the warmer months (Lenn et al., 2003),followed by mixing during the austral winter months (June-October)(Glatts et al., 2003). The locations, in order of increasing distance fromthe open sea, were assigned to three groups: (i) outer group; Whaler'sBay South (WB-S), Whaler's Bay North (WB-N), (ii) intermediategroup; Bidones Point East (BID-E), Bidones Point West (BID-W),and Spanish Antarctic Base (BAE; in front of the Gabriel de CastillaAntarctic Base), and (iii) inner group; Fumarole Bay (FUM), TelephoneBay (TEL; area recently opened to Port Foster), and Pendulum Cove(PEN; an area with abnormal water temperature due to an hydro-

thermal vent) (Somoza et al., 2004). At each location, two depths (5and 15 m) were surveyed. Port Foster is very steep all along the baywith a sheltered platform down to ca. 4–6 m depth; below thisplatform, the slope increases dramatically reaching rapidly depths ofca. 100 m.

2.2. Sampling design and biotic measurements

Three replicated transects were carried out at each location.Transects were 5–10 m apart and ran along a depth gradient from15 to 5 m depth. Along each transect line, we counted the number ofspecimens of the three species by recording, every 0.5 m, the organismslaying under the transect line (line-intercept transect, LIT). In addition,at each of two depth strata (5 and 15 m) per location, six replicatedquadrats (0.5 m×0.5 m) were randomly deployed, and specimensinside counted. When possible, 60 organisms per species and depthwere collected, all of which were measured (disk diameter for the brittlestar, body diameter for the sea urchin and body length for the sea star)and weighted (wet weight), before being returned to the sea.

2.3. Abiotic measurements

Three replicated 10 cm diameter (0.008 m2) cores were taken ateach depth and location, which penetrated approximately 10 cm intothe sediment. Sub-samples were analyzed for grain size, the degree ofsorting and the content of organic matter. Organic matter content wascalculated after drying each sediment sample for 72 h at 60 °C andsubsequently calcinating the sample with a muffle furnace at 550 °C for10 h. The percentage of organic matter was then calculated using thedifference in weight before and after calcination. To assess the grainsize and the degree of sorting, sediments were initially treated with 1 Mhydrochloric acid to remove carbonates before combustion. Sedimentgrain size was obtained by calculating the mean and standard deviationfrom cumulative plots along a phi scale, analyzed on a Beckman-Coulter LS 13 320 Laser Diffraction Particle Size Analyzer. The degreeof sorting was expressed as: φ=–log2 D/D0 (Krumbein, 1934), whereD is the diameter of the particle and D0 is a reference diameter, equalto 1 mm (to make the equation dimensionally consistent).Unfortunately, sediment samples from Whalers Bay South (5 m depth)were lost, and data could not be determined for this location.

2.4. Statistical analyses

To test for differences in the abundances (ind m−2) of eachechinoderm from deployed quadrats at each depth strata and location,a two-way analysis of variance (ANOVA) including `depth´ as a fixedfactor and `location´ as a random factor, was performed (Zar, 1984)for each species. In case of significant 'depthxlocation' interactions,pair wise tests between `depth´ levels were performed separately foreach location. Abundance data were fourth root-transformed prior toeach analysis to avoid heterogeneous variances. We omitted PendulumCove from the analysis, due to the absence of any organism. To workout the influence of environmental drivers on patterns of echinodermabundances, a generalized linear model (GLM) was performed for eachspecies. Environmental (quantitative) drivers were included, as well asdepth and the distance from the entrance of channel, to predictabundances. To determine the degree to which echinoderm popula-tions were clumped, uniformly or randomly distributed, the coefficientof variation (standard deviation/mean) of each species were calculated.Finally, data from the three transects per location were pooled for each0.5 m interval; a simple linear regression was then adjusted to predictechinoderm abundances according to depth for each location.Differences in the frequency of sizes of each specie between the twodepths were statistically tested through a chi-square test for eachlocation.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

132

3. Results

3.1. Patterns of abundance: data from quadrats

Maximum abundances reached 412 ind m−2 (TEL-15) forOphionotus victoriae, 324 ind m−2 (BID-W-15) for Sterechinusneumayeri and 24 ind m−2 (BID-E-15) for Odontaster validus(Fig. 2). We did not find any specimen of O. validus in 6 out of the 8locations at 5 m depth; only WB-N-15 and WB-S-15 showed thepresence of sea-stars at this depth stratum. No specimens of O.victoriae were observed at BAE-5, BID-W-5, PEN-5 and TEL-5, whileno S. neumayeri were found at BID-W-5, FUM-5 and PEN-5.

Pendulum Cove was the only location without any specimen of thethree species. Patterns of abundance of the three echinoderms incon-sistently differed between depth strata (`De x lo´, p < 0.0001). Pairwisetests, however, showed that the three species generally reached largerabundances at 15 m than at 5 m (Fig. 2). In turn, this inter-locationvariation did not prevent overall differences in the abundances of thethree species between the two depth strata (`De´, p < 0.0001). Further,coefficient of variation indicates different structure between speciesand locations (Table 2). For the whole study, O. victoriae (0.77 ± 0.64),S. neumayeri (0.52 ± 0.45), and O. validus (1.44 ± 0.76) showingsignificant differences between them (p < 0.0087) and no significantdifferences between depths. S. neumayeri were the less aggregated

Fig. 1. Sampling area (A) at the Antarctic continent, including (B) South Shetland Islands, and (C) Deception Island. Black stars correspond to sampling locations:WB-N, Whalers BayNorth; WB-S, Whalers Bay South; BID-E, Bidones Point East; BID-W, Bidones Point West; BAE, Antarctic Spanish Base; FUM, Fumaroles Bay; TEL, Telephone Bay; PEN,Pendulum Cove. GdC, Gabriel de Castilla Spanish Antarctic Base.

Table 1Sediment characteristics of each sampling location, including mean grain size (n=3), ᶲ: degree of sorting, and percent organic matter content (% O.M., n=3). All values are given as mean± standard deviation.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

133

specie evenly distributed in all location except BAE-5, whereas O.victoriae showed a similar uniformly distribution pattern, though three5 m locations displayed a patchy distribution (WBN, WBS, FUM).Moreover, O. validus exhibited a clumped distribution in all locationbut BID-W-15 and BID-E-15.

3.2. Patterns of abundance: data from transects

A positive, significant, relationship was majorly observed betweenabundances of O. victoriae and depth (only two locations showed nosignificant differences, BAE and FUM) (Fig. 3). For S. neumayeri, apositive significant relationship was found at BAE, BID-W, FUM, andTEL; a negative relation at BID-E and WB-N, and no relation at WB-S.Finally, depth did not influence abundances of O. validus, probably dueto its lower abundance respect to the other two studied species.

3.3. Abiotic factors

The mean organic matter content of the sediment did not sig-nificantly differ (p=0.067) between 5 m (2.89 ± 0.86%) and 15 m depth(4.55 ± 1.09%), including a high variability between locations (from1.12% to 9.44%, with a maximum of 19.45% in one of the replicates ofBID-W-5; Table 1). Grain size and the degree of sorting significantlydiffered between depths (p < 0.001); sediment grain size was larger at 5relative to 15 m depth, while the degree of sorting showed a reversalpattern, both of them correlated with the depth.

3.4. Environmental drivers of abundance patterns

Abundance patterns were differently affected by proximity from theopen ocean. O. victoriae significantly decreased its abundance withincreasing distance, while S. neumayeri and O. validus showed thelarger abundances at the intermediate locations (Fig. 4). Depth was thevariable most contributing to spatial variation of the three species,including significantly higher abundances at 15 m depth (Table 3). Thecontent in organic matter only influenced the brittle star, while grainsize contributed to explain variation for both the brittle star and the seaurchin.

3.5. Size frequency patterns

For the whole study, all three echinoderms showed significantdifferences in mean sizes between depth strata (Fig. 5, Table 4). O.victoriae showed larger individuals at 15 m, whereas S. neumayeri andO. validus showed the opposite trend, with larger individuals at 5 m.The largest individuals of O. victoriae were found in PEN-15 and FUM-15, while the smaller specimens were found in TEL-15, BID-E (5 and15 m), and BID-W-15 (Supplementary material A). S. neumayerishowed significant differences in sizes between depth strata at alllocations (except at FUM). These differences were more pronounced inBAE, with the smallest specimens at BAE-15 (Supplementary materialII). The rest of the locations displayed similar sizes (about 30–35 cm).Meanwhile, O. validus only showed significant differences at BID-Wwith the smaller specimens at 15 m depth (Supplementary material B).Furthermore, overall size distribution structure displayed differentpopulation size structure (Supplementary material C1), but site by site

Fig. 2. Abundances of (A) Ophionotus victoriae, (B) Sterechinus neumayeri and (C)Odontaster validus from quadrats deployed at 5 and 15 m at each location. Error barsdepict standard errors. *; significant differences between depths (p < 0.05).

Table 2Coefficient of variation (abundance standard deviation/mean abundance) of each speciesby locations. Sample abbreviations as listed in Table 1.

Locations Depth (m) Ophionotusvictoriae

Stherechinusneumayeri

Odontastervalidus

WBN 15 0.25 0.42 0.925 1.05 0.48 2.45

WBS 15 0.48 0.53 2.455 1.22 0.25 1.55

BID-W 15 0.66 0.1 0.355 – – –

BID-E 15 0.48 0.22 0.795 0.57 0.5 –

BAE 15 0.28 0.28 1.315 – 1.82 –

FUM 15 0.69 0.8 –

5 2.45 – –

TEL 15 0.32 0.28 1.675 – 0.53 –

Total mean 0.77 0.52 1.44

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

134

Fig. 3. Abundances of (A) Ophionotus victoriae, (B) Sterechinus neumayeri and (C) Odontaster validus from transects deployed at each location (data pooled from the 3 replicatedtransects). Lineal regression equations are included when significant differences were obtained.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

135

Fig. 3. (continued)

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

136

Fig. 3. (continued)

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

137

comparisons indicated that only S. neumayeri presented a clearlydifferent population size structure (Supplementary material C2).

4. Discussion

Port Foster is a semi-enclosed bay fed by glaciers, many of themstratified, with layers of volcanic ash from past volcanic eruptions, andwith exchange of seawater from the Bransfield Strait through a narrowopening at Neptune´s Bellows. Volcanic activity is believed to be thekey environmental factor that controls epibenthic and infaunal popula-tions (Gallardo and Castillo, 1968, 1970). This physical disturbance, inconjunction with deposition of secondary ash caused by winds,continued to alter the composition of the substrate (Gallardo et al.,1975), and has contributed to a higher sedimentation particulate flux,compared to the nearby Bransfield Strait (Baldwin and Smith, 2003).High ash sedimentation rates were suggested to explain the scarcefilter-feeding megafaunal communities at Port Foster (Lovell andTrego, 2003). DI has showed particulate matter fluxes of similarmagnitude to other Antarctic shallow waters (1026–3699 mg m−2 d−1),with important episodic fluxes related to the austral summer (Baldwinand Smith et al., 2003). Moreover, particulate matter fluxes continueduring winter, correlating with local wind speed; lithogenic material,

and possibly wind-blow or ice-rafted material, increase the winterfluxes in Port Foster, being up to five orders of magnitude larger than inthe surrounding waters of Bransfield Strait (Baldwin and Smith, 2003).Thus, this could be the most important factor structuring assemblagesin soft-bottoms at DI, and one of the reasons that explains soft-bottomsessile megafauna was not found in the study down to 15 m depth. DIcommunities fit into the pattern of strong dominance by echinoderms,particularly for Ophionotus victoriae and Sterechinus neumayeri,which were also dominant in the samples of the ExpeditionAntarctique Française 1908–1910 (Koehler, 1912). This suggests, inagreement with Gallardo et al. (1977), that probably this benthiccommunity has reached the highest level of organization that thevolcanic activity of the island allows.

The dominance of O. victoriae and S. neumayeri agrees withprevious observations (Sáiz-Salinas et al., 1997; Arnaud et al., 1998).Gallardo et al. (1977) reported O. validus as one of the dominantspecies too. Remarkably, Cranmer et al. (2003) did not find anyspecimen of O. validus at the center of Port Foster, while densities ofO. victoriae were correlated with sinking mass flux. This ophiuroidappeared to dominate the epibenthic assemblages, comprising 89–96%of the total individuals observed (11.76 ind m2) (Cranmer et al., 2003).We found a similar pattern, but at larger abundances than previously

Fig. 4. Densities of (A) Ophionotus victoriae, (B) Sterechinus neumayeri and (C) Odontaster validus related to distance from the open sea. Polynomial regression equations areincluded when significant differences were obtained.

Table 3Results of Generalized Lineal Models with a Poisson distribution for each species. Significant codes; '***' 0.001 '**' 0.01 '*' 0.05.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

138

described for the three studied echinoderms, including a highercontribution of S. neumayeri than previously reported. This sea-urchinwas twice more abundant that the brittle-star at 5 m depth, withsimilar values at 15 m depth, being Sterechinus neumayeri the mostabundant echinoderm in the coastal waters of DI. Probably, due to thefreshwater input at BID-W-5 from an ice-melting creek, the frozenseabed did not allow for a normal development of soft-bottomcommunities. Actually, no echinoderms were recorded in this site,which are replaced by the limpet Nacella polaris (Nacella concinna inprevious studies; unpublished observations from the authors). In fact,the abundance of Nacella polaris at BID-W-5 was 230.7 ± 26.3 ind m2

(unpublished data from the authors), a higher value compared to the65 ind m2 found by Filcek (1993) at the southern shores of AdmiraltyBay, King George Island. In our study, N. polaris was rarely found inthe rest of the sampled locations. Surprisingly, BID-W-15 seemed tobe unaffected by the nearly frozen sea floor at 5 m depth, since highdensities of S. neumayeri, O. victoriae, and O. validus were foundthere, despite the short distance between these two locations.

Our results do not support the hypothesis that populations ofinvertebrates from locations adjacent to the open sea would showhigher abundances; in particular, for S. neumayeri and O. validus. Thishypothesis was based in the “gradient from the channel”, proposed byBarnes et al. (2008), but it has been ruled out by our results. Thisgradient was exclusively evident for O. victoriae and was probablyrelated to abiotic factors. Further, the organic matter was onlysignificant for O. victoriae, highlighting the relationship of this specieswith sediment parameters, probably associated with food availability.Correlations between macrobenthic abundance and the sedimentparameters examined, suggests that depth (or depth-related factorssuch as food availability) play a pivotal role in structuring themacrobenthic communities of echinoderms. The abiotic factors (sort-ing and grain size) were strongly connected with depth. Shallow

benthic megafaunal communities seem to be not directly affected bythe organic input from primary production. Unlike shallow benthicmacrofauna, megafaunal communities are more stable overtime andless affected by the physical constraints. These constrains are normallyassociated with ice disturbances, such as anchor ice and ice scours,absent here, due to the special conditions of the island. It seems thatsedimentation has become the more outstanding physical factor,promoting these communities of echinoderms (Grenz et al., 2000),avoiding proliferation of filter-feeding communities at DI. The absenceof correlation between abundances of the two echinoderms (S. neu-mayeri and O. validus) and the abiotic factors indicates that thesespecies are basically non-selective species with regard to sediment type.

Pendulum Cove was characterized by the absence of the threespecies, probably due to abnormal water conditions provoked byfumaroles emissions (Smellie and López-Martínez 2000). At the otherlocations, echinoderms presented different distributions, characteristicof motile megafauna (Starmans et al., 1999), increasing with depth, butunrelated with distance to the open sea or barely influenced by theabiotic factors measured here (organic matter, grain size, and sorting).According to Retamal et al. (1982) and Barnes et al. (2008), DI hasbeen considerably recolonized since the last eruption, but many taxaare still very poorly represented, shaping a community of abundantopportunistic species, both at infaunal and megafaunal levels (Lovelland Trego, 2003; Angulo-Preckler et al., submitted). These commu-nities, even with the continuous disturbances produced by the highsedimentation rates, seem to have been stable during these years.Probably, due to the absence of ice-related disturbance, S. neumayeriappeared to be less aggregated and more evenly distributed than O.victoriae while O. validus presented a clumped distribution, character-ized by congregations of small clusters and low abundances, bothprobably related to feeding behavior. Sáiz-Salinas et al. (1997, 1998)suggested that the general distribution patterns of epibenthic commu-

Fig. 5. Box plots denoting overall differences in size (disk diameter, body diameter and body length in cm) for (A) Ophionotus victoriae, (B) Sterechinus neumayeri and (C) Odontastervalidus at 5 and 15 m. The horizontal line inside the boxes represents the mean size; the outer edges of the boxes are the 25th and 75th percentiles, while the vertical lines indicate 5thand 95th percentiles. N; Total number of organisms measured. *; significant differences (p < 0.05).

Table 4Differences in the mean size of each species between 5 and 15 m at each site. *; significant differences. Sample abbreviations as listed in Table 1.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

139

nities in the South Shetland islands are linked to food availability.Suspension-feeding assemblages are present down to 100 m depth,while deposit-feeders, such as brittle stars, dominate down to 400 m.Different patterns have been observed inside DI, where opportunisticdeposit-feeders completely dominate the shallow soft-bottom commu-nities. Even though O. victoriae and S. neumayeri are circumpolarspecies with a wide distribution, their high abundances and obviousdominance have never been reported previously in any other Antarcticor sub-Antarctic zones (Brey et al., 1995; Manjón-Cabeza and Ramos,2003). Instead, similar abundances of O. validus relative to thosedescribed by McClintock et al. (1988) at McMurdo Sound wererecorded here. The stable population structure of O. validus is, atleast, related to the extremely low growth rates observed, and indicatesvery low levels of recruitment, migration, and mortality (see Ebert,1973, 1983).

Individual size in echinoderms reflect fitness related to foodavailability and competition (Lawrence, 1985). Limited food availabil-ity has been implicated as primary basis of low growth rates in manyantarctic marine invertebrates (Clarke, 1980; Brey and Clarke, 1993).We found maximum sizes corresponding to extremely slow growthorganisms, reaching a maximum of 1) 28.5 cm size for O. victoriae,where similar values were obtained by Dahm and Brey (1998) atMcMurdo Sound, aging these around 20–22 years old, 2) 52 cmdiameter for S. neumayeri, where Brey (1991) established an age of40 years to 70 cm diameter specimens, and 3) 64 cm for O. validus,where, according to McClintock et al. (1988), this sea-star needs atleast 9 years to reach that size. Body sizes varied significantly amonglocations and depths for the three species, especially considering thatno organisms were present at 5 m depth at several locations. Theseechinoderms, in shallow waters, would be expected to feed opportu-nistically on mobile prey, as sessile prey are absent in soft-bottoms ofDI. The high densities of macrofauna (Angulo-Preckler et al., sub-mitted) could be maintaining the high densities of echinoderms foundhere. In general, the largest or smallest (i.e. juveniles) specimensseemed to be concentrated at some locations. O. victoriae showedlarger organisms at 15 m depth, and a high heterogeneity withinlocations (Supplementary material C1 and C2). A large contrast in sizeswas found between PEN-15, which hosted the largest organisms andTEL-15, where the smallest organisms were found, probably due toimportant recruitment processes. In addition, the absence of compe-titors at Pendulum Cove (no species were recorded), which may haveresulted from hydrothermal heating, could favor the growth of theomnivorous O. victoriae, allowing them to reach the largest sizes of thewhole study. A similar density-size pattern was observed for S.neumayeri, especially clear at the Antarctic Spanish Base, being thelargest individuals at BAE-5, and the smallest at TEL-15.Interestingly, no sea urchins smaller than 20 mm diameter wereobserved at 5 m, suggesting that despite the absence of potentialpredators at 5 m depth, juvenile recruitment does not seem to occurat this depth. Several studies have shown a size-gradient in thedensities of echinoderms inhabiting the lower intertidal and subtidalzones (Larson, 1968; Freeman, 2003). These within-species differencesin their abundances have been attributed to stress caused by watermovement (Tuya et al., 2007) and reinforced here, due to the steepslope. Finally, O. validus showed similar depth and spatial size-patterns between depths. S. neumayeri and O. validus were larger at5 m depth. Larger body size of shallow-water individuals of O. validusand S. neumayeri may reflect higher levels of food availability in theshallowest depth strata. This could be related with organic enrichmentcaused by mass mortalities of algae and pelagic invertebrates (e.g.Euphasia crystallorophias), which often appear dead in the beaches,due to the geothermally warm waters typical from this volcano(Brierley 1999; authors's personal observations). Our data indicatethat the epibenthic megafaunal community of Port Foster has recov-ered after the last volcanic eruption, with the return of key echinodermspecies, such as O. victoriae, S. neumayeri, and O. validus. Actually, S.

neumayeri is the dominant species at shallow waters, and co-domi-nated with O. victoriae at 15 m depth.

In the near future, the warming trends for the Northern AntarcticPeninsula (Clarke et al., 2007), could increase the sedimentation ratesdue to glacial retreatment. Thus, disturbances produced by highsedimentation rates could potentially become more relevant than ice-disturbances, dramatically enhancing sediment-feeders communitiesover the suspension-feeding communities, reducing the biodiversity inthis waters.

Ethical statement:

The research reported here, has been conducted in an ethical andresponsible manner and comply with all relevant legislation.

1. We have no potential conflict of interest.2. All procedures performed in this study involving animals were in

accordance with the ethical standards of the institution or practice atwhich the study was conducted.

3. Informed consent: Informed consent was obtained from all indivi-dual participants included in the study.

Contributors

C.A-P; Design the experiments. C. A-P and C.A; Field work. C. A-Pand F.T; Statistic analysis. C. A-P, F.T. and C.A; Write the manuscript.

Acknowledgments

Thanks are due to F.J. Cristobo, L. Núñez-Pons, J. Moles, A. Riesgo,and M. Bas for their help with the fieldwork. Thanks are also due to thecrew of BAE Gabriel de Castilla for their logistic support during theAntarctic dives. This work was developed within the frame of theACTIQUIM-II research project (CTM2010-17415) and C. Angulo-Preckler was funded by the FPU Program (AP2009-2081) from theSpanish Government. The echinoderm pictures were used with thepermission of Dr. Figuerola, and for them we are thankful.

Appendix A. Supporting information

Supplementary data associated with this article can be found in theonline version at http://dx.doi.org/10.1016/j.csr.2016.12.010.

References

Angulo-Preckler, C., Leiva, C., Avila, C., Taboada, S., 2016. Macroinvertebratecommunities from the shallow soft-bottoms of Deception Island (Southern Ocean): aparadise for opportunists. Marine Environmental Research (submitted).

Arnaud, P.M., Lopez, C.M., Olaso, I., Ramil, F., Ramos-Esplá, A.A., Ramos, A., 1998.Semi-quantitative study of macrobenthic fauna in the region of the South ShetlandIslands and the Antarctic Peninsula. Polar Biology 19, 160–166.

Arntz, W.E., Gutt, J., Klages, M., 1997. Antarctic marine biodiversity: anoverviewAntarctic communities: species, structure and survival. CambridgeUniversity Press, 3–14.

Arntz, W.E., Brey, T., Gallardo, V.A., 1994. Antarctic Zoobenthos. Oceanography andMarine Biology 32, 241–304.

Baldwin, R., Smith, K., 2003. Temporal dynamics of particulate matter fluxes andsediment community response in Port Foster, Deception Island, Antarctica. DeepSea Research Part II Topical Studies in Oceanography 50, 1707–1725.

Barnes, D.K., Linse, K., Enderlein, P., Smale, D., Fraser, K.P.P., Brown, M., 2008. Marinerichness and gradients at Deception Island, Antarctica. Antarctic Science 20,271–279.

Barnes, D.K., Conlan, K.E., 2007. Disturbance, colonization and development ofAntarctic benthic communities. Philosophical Transactions of the Royal Society ofLondon B: Biological Science 362, 11–38.

Brey, T., Pearse, J., Basch, L., McClintock, J., Slattery, M., 1995. Growth and productionof Sterechinus neumayeri (Echinoidea: Echinodermata) in McMurdo sound,Antarctica. Marine Biology 124, 279–292.

Brey, T., Clarke, A., 1993. Population dynamics of marine benthic invertebrates inAntarctic and subantarctic environments: are there unique adaptations? AntarcticScience 5, 253–266.

Brey, T., 1991. Population dynamics of Sterechinus antarcticus (Echinodermata:

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

140

Echinoidea) on the Weddell Sea shelf and slope, Antarctica. Antarctic Science 3,251–256.

Brey, T., Gutt, J., 1991. The genus Sterechinus (Echinodermata: Echinoidea) on theWeddell Sea shelf and slope (Antarctica): distribution, abundance and biomass.Polar Biology 11, 227–232.

Brierley, A., 1999. A comparison of Antarctic euphausiids sampled by net and fromgeothermally heated waters: insights into sampling bias. Polar Biology 22, 109–114.

Brown, K.M., Fraser, K.P.P., Barnes, D.K., Peck, L.S., 2004. Links between the structureof an Antarctic shallow-water community and ice-scour frequency. Oecologia 141,121–129.

Cattaneo-Vietti, R., Chiantore, M., Misic, C., Povero, P., Fabiano, M., 1999. The role ofpelagic-benthic coupling in structuring littoral benthic communities at Terra NovaBay (Ross Sea) and in the Straits of Magellan. Scientia Marina 63, 113–121.

Clarke, A., Griffiths, H.J., Linse, K., Barnes, D.K. a, Crame, J.A., 2007. How well do weknow the Antarctic marine fauna? A preliminary study of macroecological andbiogeographical patterns in Southern Ocean gastropod and bivalve molluscs.Diversity and Distributions 13, 620–632.

Clarke, A., Aronson, R., Crame, J., Gili, J., Blake, J., 2004. Evolution and diversity of thebenthic fauna of the Southern Ocean continental shelf. Antarctic Science 16,559–568.

Clarke, A., Johnston, N., 2003. Antarctic marine benthic diversity. Oceanography andMarine Biology 41, 47–114.

Clarke, A., Crame, J., Stromberg, J.O., Barker, P.F., 1992. The Southern Ocean benthicfauna and climate change: a historical perspective. Philosophical Transactions of theRoyal Society B: Biological Sciences 338, 299–309.

Clarke, A., 1980. A reappraisal of the concept of metabolic cold adaptation in polarmarine invertebrates. Biology Journal of the Linnean Society 14, 77–92.

Cranmer, T.L., Ruhl, H., Baldwin, R.J., Kaufmann, R.S., 2003. Spatial and temporalvariation in the abundance, distribution and population structure of epibenthicmegafauna in Port Foster, Deception Island. Deep Sea Research Part II TopicalStudies in Oceanography 50, 1821–1842.

Dahm, C., Brey, T., 1998. Determination of growth and age of slow growing brittle stars(Echinodermata: Ophiuroidea) from natural growth bands. Journal Marine BiologyAssociation of the United Kingdom 78, 941–951.

Dayton, P., Robilliard, G., Paine, R., Dayton, L., 1974. Biological accommodation in thebenthic community at McMurdo Sound, Antarctica. Ecological Monograph 44,105–128.

Dayton, P.K., Robilliard, G.A., Devries, A.L., 1969. Anchor ice formation in McMurdoSound, Antarctica, and its biological effects. Science 163, 273–274.

Dearborn, J., 1977. Foods and feeding characteristics of Antarctic asteroids andophiuroids. Adaptations within Antarctic Ecosystems, 293–326.

Dearborn, J., 1972. Ecological and Taxonomic studies of echinoderms, molluscs, andfishes from Antarctic Peninsula. Antarctic Journal of the Unite State 7, 80–82.

Dearborn, J., 1965. Ecological and faunistic investigations of the marine benthos atMcMurdo Sound, AntarcticaDepartment of Biological Sciences. Stanford University.

Dell, R.K., 1972. Antarctic benthos. Advances in Marine Biology 10, 1–216.Ebert, T., 1983. Recruitment in echinoderms. Echinoderm studies 1, 169–203.Ebert, T., 1973. Estimating growth and mortality rates from size data. Oecologia 11,

281–298.Elderfield, H., 1972. Effects of volcanism on water chemistry, Deception Island,

Antarctica. Marine Geology 13, M1–M6.Filcek, K., 1993. Patellidae. The Maritime Antarctic ecosystem of Admiralty Bay. Polish

Academy of Science.Fratt, D., Dearborn, J., 1984. Feeding biology of the Antarctic brittle star Ophionotus

victoriae (Echinodermata: Ophiuroidea). Polar Biology 3, 127–139.Freeman, S.M., 2003. Size-dependent distribution, abundance and diurnal rhythmicity

patterns in the short-spined sea urchin Anthocidaris crassispina. Estuarine, Coastaland Shelf Science 55, 703–713.

Gallardo, V., Castillo, J.G., Retamal, M.A., Moyano, H.I., Hermosilla, J.G., 1977.Quantitative studies on the soft-bottom macrobenthic animal communities ofshallow Antarctic baysAdaptations within Antarctic ecosystems. SmithsonianInstitution, Washington, DC, 631387.

Gallardo, V., Castillo, J., Retamal, M.A., Hermosilla, J., Trucco, R., 1975. BenthicCommunity Studies in South Shetland-Islands. Antarctic Journal of the UnitedStates 10, 135–135.

Gallardo, V., Castillo, J., 1970. Quantitative benthic survey of the infauna of Chile Bay(Greenwich Island)Antarctic Ecology 1. Academic Press, 242–243.

Gallardo, V., Castillo, J., 1968. Mass Mortality in the Benthic Infauna of Port FosterResulting from the Eruptions in Deception Island (South Shetland Is.).

Glatts, R., Uhlman, A., Smith, K., Baldwin, R., 2003. Long time-series monitoring of theecosystem at Deception Island, Antarctica: description of instrumentation. Deep SeaResearch Part II Topical Studies in Oceanography 50, 1631–1648.

Gray, J., 2001. Marine diversity: the paradigms in patterns of species richness examined.Scientia Marina 65, 41–56.

Grenz, C., Cloern, J.E., Hager, S.W., Cole, B.E., 2000. Dynamics of nutrients cycling andrelated benthic nutrient and oxigent fluxes during spring phytoplankton bloom inSouth San Francisco Bay (USA). Marine Ecology Progress Series 197, 67–80.

Gutt, J., Piepenburg, D., 2003. Scale-dependent impact on diversity of Antarctic benthos

caused by grounding of icebergs. Marine Ecology Progress Series 253, 77–83.Gutt, J., 2000. Some ‘driving forces’ structuring communities of the sublittoral Antarctic

macrobenthos. Antarctic Science 12, 297–313.Jażdżeski, K., Jurasz, W., Kittel, W., Presler, E., 1986. Abundance and biomass estimates

of the benthic fauna in Admiralty Bay, King George Island, South Shetland Islands.Polar Biology 6, 5–16.

Koehler, R., 1912. Echinodermes (Asteries, Ophiures et Echinides). DeuxièmeExpèdition Antarctique Francaise, 1908–1910.

Krumbein, W.C., 1934. Size Frequency Distributions of Sediments. Journal ofSedimentary Research, 4.

Larson, B.A., 1968. Scuba-studies on vertical distribution of Swedish rocky-bottomechinoderms. A methodological study. Ophelia 5, 137–156.

Lawrence, J., 1985. The energetic echinoderm. Balkema, Rotterdam, 47–67.Lebrato, M., Iglesias-Rodriguez, D., Feely, R., Greeley, D., Joones, D., Suarez-Bosche, N.,

Lampitt, R., Cartes, J., Green, D., Alker, B., 2010. Global contribution ofechinoderms to the marine carbon cycle: CaCO3 budget and benthic compartments.Ecological Monograph 80, 441–467.

Lenn, Y.D., Chereskin, T.K., Glatts, R.C., 2003. Seasonal to tidal variability in currents,stratification and acoustic backscatter in an Antarctic ecosystem at Deception Island.Deep Sea Research Part II Topical Studies in Oceanography 50, 1665–1683.

Lovell, L.L., Trego, K.D., 2003. The epibenthic megafaunal and benthic infaunalinvertebrates of Port Foster, Deception Island (South Shetland Islands, Antarctica).Deep Sea Research Part II Topical Studies in Oceanography 50, 1799–1819.

Madsen, F.J., 1967. Ophiuroidea. British and Australian NZ Antarctic ResearchExpedition 1929-1931. Rep Adelaide (Ser B Zool Bot) 9, 123–145.

Manjón-Cabeza, M.E., Ramos, A., 2003. Ophiuroid community structure of the SouthShetland Islands and Antarctic Peninsula region. Polar Biology 26, 691–699.

McClintock, J., Pearse, J., Bosch, I., 1988. Population structure and energetics of theshallow-water Antarctic sea star Odontaster validus in contrasting habitats. MarineBiology 99, 235–246.

McClintock, J., 1994. Trophic biology of Antarctic shallow-water echinoderms. MarineEcology Progress Series 111, 191–202.

Moya, F., Ramos, A., Manjón-Cabeza, M.E., 2003. Distribution and ecology ofOphionotus victoriae Bell, 1902 (Ophiuroidea, Echinodermata) in the SouthShetland Islands area (Antarctica). Boletin del Instituto Español de Oceanografia 19,49–55.

Pawson, D., 1994. Antarctic echinoderms: history, distribution, ecology, 1968-1993Echinoderms through Time. Balkema, Rotterdam, 99–110.

Pearse, J., Giese, A., 1966. Food, reproduction and organic constitution of the commonAntarctic echinoid Sterechinus neumayeri (Meissner). The Biological Bulletin 130,387–401.

Pearse, J., 1964. Reproductive periodicities in several contrasting populations ofOdontaster validus Koehler, a common Antarctic asteroid. American GeophysicalUnion, 39–85.

Peckham, V., 1964. Year-round SCUBA diving in the Antarctic. Polar Record 12,143–146.

Ramos, A., 1999. The megazoobenthos of the Scotia Arc islands. Scientia Marina 63,171–182.

Retamal, M., Quintana, R., Neira, F., 1982. Qualitative and quantitative analysis ofbenthic communities in Foster Bay, Deception Island (35th Chilean AntarcticExpedition, Jan. 1981). Serie Cientifica Instituto Antártico Chileno 29, 5–15.

Sáiz-Salinas, J., Ramos, A., Munilla, T., Rauschert, M., 1998. Changes in the biomass anddominant feeding mode of benthic assemblages with depth off Livingston Island(Antarctica). Polar Biology 19, 424–428.

Sáiz-Salinas, J.I., Ramos, A., García, F.J., Troncoso, J.S., San Martin, G., Sanz, C.,Palacin, C., 1997. Quantitative analysis of macrobenthic soft-bottom assemblages inSouth Shetland waters (Antarctica). Polar Biology 17, 393–400.

Smale, D., 2007. Ice disturbance intensity structures benthic communities in nearshoreAntarctic waters. Marine Ecology Progress Series 349, 89–102.

Smellie, J., López-Martínez, J., 2000. Geological map of Deception Island. Serie BASGEOMAP, Sheet.

Smith, K., Baldwin, R., Kaufmann, R.S., Sturz, A., 2003. Ecosystem studies at DeceptionIsland, Antarctica: an overviewDeep Sea Research Part II Topical Studies inOceanography 50, 1595–1609.

Somoza, L., Martínez-Frías, J., Smellie, J., Rey, J., Maestro, A., 2004. Evidence forhydrothermal venting and sediment volcanism discharged after recent short-livedvolcanic eruptions at Deception Island, Bransfield Strait, Antarctica. Marine Geology203, 119–140.

Starmans, A., Gutt, J., Arntz, W.E., 1999. Mega-epibenthic communities in Arctic andAntarctic shelf areas. Marine Biology 135, 269–280.

Teixidó, N., Garrabou, J., Gutt, J., Arntz, W.E., 2004. Recovery in Antarctic benthos aftericeberg disturbance: trends in benthic composition, abundance and growth forms.Marine Ecology Progress Series 278, 1–16.

Tuya, F., Cisneros-Aguirre, J., Ortega-Borges, L., Haroun, R.J., 2007. Bathymetricsegregation of sea urchins on reefs of the Canarian Archipelago: role of flow-inducedforces. Estuarine, Coastal and Shelf Science 73, 481–488.

White, M.G., 1984. Marine benthos.Zar, J., 1984. Biostatistical analysis 2nd. Prentice Hall, USA.

C. Angulo-Preckler et al. Continental Shelf Research 137 (2017) 131–141

141