East Antarctica) During the CEAMARC Surveys

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Available online at www.sciencedirect.com

Polar Science 5 (2011) 195e210 http://ees.elsevier.com/polar/

Size distribution of meso- and bathypelagic ﬁsh in the Dumont d’Urville Sea (East Antarctica) during the CEAMARC surveys

Philippe Koubbi a,b,*, Percy-Alexander Hulley c, Patrice Pruvost d, Pauline Henri a,b, Jean-Philippe Labat a,b, Victoria Wadley e, Daisuke Hirano f, Masato Moteki f

a UPMC Universite´ Paris 06, UMR 7093, Laboratoire d’Oce´anographie de Villefranche, BP28, 06234 Villefranche-sur-Mer, France b CNRS, UMR 7093, LOV, BP 28, 06234 Villefranche-sur-Mer, France c Iziko-South African Museum, P. O. Box 61, 8000 Cape Town, South Africa d MNHN, DMPA-UMR 7208, 43 rue Cuvier, 75005 Paris, France e Department of Sustainability, Environment, Water, Population and Communities, Australian Antarctic Division, 203 Channel Highway, Kingston, Tasmania 7050, Australia f Faculty of Marine Science, Tokyo University of Marine Sciences and Technology, 4-5-7 Konan, Minato, Tokyo 108-8477, Japan

Received 9 February 2011; revised 7 March 2011; accepted 8 March 2011 Available online 5 April 2011

Abstract

The pelagic fish community of the Dumont d’Urville Sea (East Antarctica) was investigated during the 2008 austral-summer using IYGPT (International Young Gadoid Pelagic Trawl) samples taken in different depth layers from the surface to 1000 m. The aim of this paper is to describe the mesopelagic fish community and its size distribution. The family Myctophidae dominated the mesopelagic ichthyofauna, while bathylagids were abundant in deeper hauls. Bathylagids, Cyclothone spp., Gymnoscopelus opisthopterus, Electrona antarctica, Protomyctophum bolini, and Krefftichthys anderssoni were the most abundant taxa in the samples and showed size stratification with depth. Community and size structuring appear to be influenced by the hydrology and by the proximity of the continental margin, as well as a relationship to the circulation of the Modified Circumpolar Deep Water. Ó 2011 Elsevier B.V. and NIPR. All rights reserved.

Keywords: Mesopelagic; Bathypelagic; Southern Ocean; Myctophids; East Antarctica

1. Introduction them perform diel migration, coming to the epipelagic zone during the night (Gjosaeter and Kawaguchi, The mesopelagic zone relates to the water masses 1980). To the south of the Antarctic Polar Front, the between the epipelagic zone (0e200 m depth) and the majority of meso- and bathypelagic species have deep bathypelagic zone at 1000 m. Mesopelagic fish circumpolar distributions that are linked to specific spend the day in the mesopelagic zone and most of water masses (Hulley, 1981; Koubbi et al., 2011). Here a few families dominate the ichthyofauna: Bathy- lagidae, Gonostomatidae, Myctophidae and Paral- ´ * Corresponding author. UPMC Universite Paris 06, UMR 7093, epididae (Kock, 1992; Kellermann, 1996). Of these, Laboratoire d’Oceánographie de Villefranche, BP28, 06234 Villefranche-sur-Mer, France. Tel.: þ33 140793760. myctophids or lanternfishes are the most diverse and E-mail address: [email protected] (P. Koubbi). abundant family with more than thirty species (Hulley,

1990 and Barrera-Oro, 2002) and a biomass (probably micronekton (Hoddell et al., 2000). The ICO2TA an underestimate) of 70e200 106 t(Lubimova et al., (Integrated Coastal Ocean Observations in Terre Ade´- 1983). These micronektonic species represent a mid- lie) surveys, commencing in 2004, sampled fish larvae trophic level between zooplankton (Pakhomov et al., from the shelf area (Koubbi et al., 2011), but with 1996; Pusch et al., 2004; Flores et al., 2008; Van de limited samples in the oceanic zone. Putte et al., 2010) and top predators (seabirds and The continental shelf of the Dumont d’Urville Sea marine mammals) (Che´rel et al., 2010). Mesopelagic from Ade´lie Land to the Mertz Glacier Tongue is fish play an important role in the ecosystem since they 120e130 km wide with a shelf break located at depths allow for rapid exchange of production from the between 500 m and 550 m. The continental slope epipelagic zone to the deeper layers during their diel extends down to about 2000 m. It is characterized by vertical migrations (Torres and Somero, 1988; Lancraft alternating banks, shoaling to 200 m (Ade´lie and et al., 1989; Piatkowski et al., 1994; Duhamel, 1998; Mertz Banks), and troughs connected to submarine Pusch et al., 2004; Moteki et al., 2009). Electrona canyons. A complex network of submarine canyons is antarctica is one of the most abundant myctophids observed with the Jussieu and Cuvier canyons reach- along the East Antarctica continental shelf (Hoddell ing the shelf break and innershelf depressions; others et al., 2000; Moteki et al., 2009; Van de Putte et al., canyons are not connected to the shelf (Caburlotto 2010; Koubbi et al., 2010), and the species has et al., 2006; De Santis et al., 2007; Beaman, 2008; a broad distribution between the Antarctic Polar Front Post et al., 2011). The continental rise may be sub- and the continental shelf (Hulley, 1990; Koubbi et al., divided into two regions, a relatively steep region 2011). from 2000 to 3000 m and with deep-sea channels; and The shelf break represents a marked change a gentler lower rise in areas exceeding 3000 m between the oceanic and neritic species assemblages. (Caburlotto et al., 2006). Shelf assemblages are dominated by notothenioids, The area has been identified as a source of Antarctic mainly Antarctic silverfish (Pleuragramma antarcti- Bottom Waters (AABW), named ALBW (Ade´lie Land cum), icefish and fish larvae (Hulley, 1992; Hoddell Bottom Water) (Gordon and Tchernia, 1972; Rintoul, et al., 2000; Van de Putte et al., 2010; Koubbi et al., 1998). Circumpolar Deep Water (CDW) is differenti- 2010, 2011). Oceanic assemblages are dominated by ated from the Modified Circumpolar Deep Water meso- and bathypelagic fish, while along the conti- (MCDW) by being warmer, more saline and with lower nental margin there is often a transitional assemblage, dissolved oxygen; both are circumpolar (Orsi et al., comprising oceanic species mixed with those noto- 1999; Bindoff et al., 2000). MCDW corresponds to thenioids associated with Antarctic krill (Euphausia waters that have upwelled onto the outer continental superba) swarms (Hoddell et al., 2000; Koubbi et al., slope at intermediate depths and includes Slope Water 2010). Most of the studies near the continental shelf (Whitworth et al., 1998; Gordon and Tchernia, 1972; margin have employed RMT 1 þ 8 nets to sample fish Bindoff et al., 2000). Major currents like the larvae, post larvae and juveniles, or bongo nets for fish Antarctic Circumpolar Current (ACC) and the East larvae (Koubbi et al., 2011). Antarctic Current, are associated with gyral systems In the present study, a large pelagic trawl was used (Nicol et al., 2000). The southern boundary of the ACC to sample macrozooplankton and micronekton, corresponds to the southernmost edge of the Upper although an RMT 1 þ 8(Moteki et al., 2011) was also Circumpolar Deep Water (Orsi et al., 1995). It is deployed along the transects. The CEAMARC surveys recognized as being an important region for biological (Collaborative East Antarctic Marine Census), which production (Tynan, 1998; Nicol et al., 2000). The form part of the Census of Antarctic Marine Life Antarctic Slope Front (Jacobs, 1991) is bathymetrically (CAML, IPY Project 53) were carried out during the locked to the edge of the continental shelf close to the Austral-summer 2008 in the Dumont d’Urville Sea, 1000 m, where it separates the less dense Antarctic from Terre Ade´lie to the Mertz Glacier Tongue. The Surface Waters (AASW) from the denser MCDW Japanese TRV Umitaka Maru was used to sample the (Bindoff et al., 2000). pelagic realm from the coast and into the oceanic zone In this article, meso- and bathypelagic fish spatial to as far north as 62S. Previously and in our study distributions are studied with reference to their diver- area, a biological/oceanographic survey (BROKE- sity, abundance and size. Previously and in order to East) had been conducted in 1995/96, but at a much delineate ecoregions, Koubbi et al. (2010) have broader scale, and mainly in the oceanic zone (Nicol considered presence-absence data for the pelagic fish et al., 2000). Here, an RMT was used to study the community investigations. P. Koubbi et al. / Polar Science 5 (2011) 195e210 197

2. Materials and methods were taken in the oceanic zone (Table 1); 47 samples were taken during the whole survey, including the 2.1. Sampling continental shelf. Temperature and salinity were measured by CTD TRV Umitaka Maru conducted a comprehensive (Sea-Bird Electronics SBE911plus) to depths of pelagic survey of the epi-, meso- and bathypelagic 2000 m, or just above the bottom for shallower depths. zones on and north of the continental shelf (Fig. 1). The accuracy of temperature and conductivity were An International Young Gadoid Pelagic Trawl net 10 3 degrees and 3.10 4 Sm 1 (2.10 3 psu in terms of (IYGPT) was used to collect pelagic fish. Although the salinity) respectively. net was designed to catch young cod in Northern Hemisphere waters, it has also been used in the Southern 2.2. Preservation and identification Ocean to collect myctophids (Duhamel et al., 2000; Collins et al., 2008). The IYGPT net used during the Fish samples were preserved either in 5% buffered CEAMARC survey has approximate mouth dimensions formalin or in alcohol. The specimens in alcohol were of 5.5 m high 12 m wide when fishing, and with a mesh later used for barcoding (Dettai et al., 2011) or stable size of 100 mm in the front, tapering through 80 mm, isotope analysis (Che´rel et al., 2011). Fish were iden- 40 mm, 20 mme10 mm mesh size in the cod end. tified using Gon and Heemstra (1990), with some The IYGPT net was deployed at 16 stations from 29 identifications confirmed with molecular techniques January to 12 February 2008. However, meso- and (Dettai et al., 2011). In this study, all bathylagids were bathypelagic fish were collected at only 7 stations, pooled as Bathylagus sp. these being positioned offshore from the continental shelf. They are stations 8 to 10 and stations 14 to 20 2.3. Abundances inclusive. Samples were collected at standard depths (50, 200, 500 and 1000 m) at each station with an Abundances were calculated using the total duration average 20 min duration of sampling at each depth. As of the trawl at each standard depth (50, 200, 500 and IYGPT is not an opening/closing device, there may be 1000 m). The unit of effort was about 20 min at each some contamination with fish from shallower depths. standard depth. Catches are expressed as the number of Apart from station 18, all tows were made during individuals caught per hour. The Shannon Index daylight hours (Table 1). The sampling depth was (Frontier, 1983), employing Log2, was used to calcu- monitored by the use of logger. In total, 25 samples late species diversity of the meso- and bathypelagic fish. Notothenioids, caught mainly in the epipelagic zone (Koubbi et al., 2010) were excluded from the calculation. The influence of sampling depth on the abundance and diversity of mesopelagic fish was tested using KruskalleWallis non-parametric test for each of the species. This test was used because of hetero- scedasticity and non-normality of the distributions. Sampling depth was taken as the average depth strata (50, 200, 500 and 1000 m) for each station.

2.4. Species assemblages

Hierarchical classiﬁcation was undertaken to study the ﬁsh assemblages according to sampling depth and location. Abundances were log(xþ1) transformed (Clarke, 1990) in order to give less weight to dominant species and more to rare species. BrayeCurtis similarities among observations and species were computed (Lebart et al., 1982). Only the dominant Fig. 1. Positions of the IYGPT samples in the Dumont d’Urville Sea species were considered in this analysis: Bathylagus during the CEAMARC survey onboard TRV Umitaka Maru in sp., Cyclothone spp., Gymnoscopelus opisthopterus, E. JanuaryeFebruary 2008. antarctica, Protomyctophum bolini, and Krefftichthys 198 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Table 1 Depths of sampling of the IYGPT net in the oceanic zone of the Dumont d’Urville Sea during the CEAMARC survey onboard the TRV Umitaka Maru. Station Sampling Time of Duration at total time Longitude Latitude Bottom depth Day/Night depth (m) sampling sampling depth (hh:mm) (hh:mm) (m) 14 57 29/1/08 14:45 0:18 0:30 140.03 - 62.02 4226 Day 225 29/1/08 15:49 0:22 0:56 139.99 - 61.99 4266 Day 624 29/1/08 17:11 0:21 1:31 140.04 - 62.02 3387 Day 1268 29/1/08 19:08 0:32 2:44 139.99 - 61.97 4270 Day 16 58 30/1/08 17:08 0:20 0:56 140.00 - 63.01 3794 Day 203 30/1/08 17:08 0:23 1:05 140.01 - 63.03 3804 Day 501 30/1/08 18:33 0:20 1:19 139.99 - 62.99 3717 Day 1033 30/1/08 20:13 0:23 2:00 140.01 - 63.03 3387 Day 18 51 1/2/08 3:51 0:12 0:39 140.02 - 64.01 3695 Night 208 1/2/08 5:09 0:19 0:51 140.00 - 63.99 3691 Night 460 1/2/08 6:21 0:20 1:09 140.01 - 64.02 3698 Night 1027 1/2/08 7:50 0:21 1:39 139.98 - 63.98 3712 Dawn 20 51 2/2/08 21:07 0:19 0:44 139.79 - 65.11 2586 Day 195 2/2/08 19:57 0:22 0:50 139.75 - 65.09 2612 Day 525 2/2/08 18:27 0:18 1:09 139.83 - 65.13 2758 Day 1074 2/2/08 16:06 0:18 1:50 139.75 - 65.09 2605 Day 10 51 11/2/08 8:46 0:18 0:33 142.95 - 66.02 464 Day 194 11/2/08 9:43 0:22 0:53 142.98 - 66.00 450 Day 415 11/2/08 10:53 0:22 1:05 142.93 - 66.03 457 Day 9 55 11/2/08 19:24 0:20 0:34 143.25 - 65.71 2074 Day 200 11/2/08 20:28 0:21 0:46 143.25 - 65.68 2294 Day 500 11/2/08 21:30 0:21 1:07 143.28 - 65.70 2133 Day 1092 11/2/08 23:05 0:21 1:51 143.29 - 65.64 2117 Day 8 200 12/2/08 20:59 0:20 0:47 143.01 - 65.49 2650 Day 1000 12/2/08 18:50 0:20 1:49 142.91 - 65.44 2470 Day anderssoni. Dendrograms were obtained using group- Starting from the distribution of samples within the average sorting (Legendre and Legendre, 1984). space of size classes, and the projection of size classes within the space of the samples, this factorial analysis 2.5. Size frequency analysis provides the best possible summary of the kinetics of the size structure of a population over time. Standard length measurements (to the nearest For the analysis, we used proportions calculated for millimeter) were taken for a maximum of 50 randomly each of the size classes per sample of the most abundant selected adults and juveniles per sample. Non- species: E. antarctica, Bathylagus sp., Cyclothone parametric tests were used to see if there was a signif- microdon, G. opisthopterus, K. anderssoni, Gymno- icant difference in size between the sampling depths. A scopelus braueri and P. bolini. To avoid the over ManneWhitney test was used for comparing two depth representation of classes with few individuals, sizes- strata and a KruskaleWallis test when there were more classes with few individuals were pooled into a single than two depth strata to compare. group (Table 2). Samples were identified according to The variability of the size structure was analyzed the station and standard sampling depth where these using the following multivariate technique which is species were present : Station 14 (50 m, 200 m, 500 m, able to synthesize the key features. Correspondence 1000 m); Station 16 (200 m, 500 m, 1000 m), Station 18 Analysis (CA) (Benzećri, 1973)isanordination (500 m, 1000 m), Station 20 (500 m, 1000 m), Station 8 method which has been widely used in the analysis of (1000m)andStation9(200m,500m,1000m). ecological data (Gower, 1987), and to describe size structuring (Badia and Do-Chi, 1976). It aims to 3. Results describe the total inertia of a multidimensional set of data in a sample of fewer dimensions (or axes) that is 3.1. Hydrology the best summary of the information contained in the data. Among the inertia methods, CA employs At this latitude, the water column was vertically strat- contingency tables and uses a Chi-Square metric. ified with Deep Antarctic Water at the bottom. P. Koubbi et al. / Polar Science 5 (2011) 195e210 199

Table 2 by Koubbi et al. (2010). Water masses observed were Size classes based on standard length measurements (cm) used for the Antarctic Surface Water (AASW), Circumpolar Deep correspondence analysis. Lower and upper limits of classes are given. Water (CDW), Modified Circumpolar Deep Water CA code SL lower SL upper (MCDW) and Antarctic Bottom Water (AABW) (Fig. 3). limit limit Near the surface, a shallow layer of fresh water known as Electrona antarctica EA1 1.2 2.4 AASW was found due to ice melting. This AASW layer EA2 2.4 3.6 has the remnant of the previous winter mixed layer, also EA3 3.6 4.8 EA4 4.8 6.0 known as the temperature minimum layer (Tmin layer) or EA5 6.0 7.2 Winter Water. Cold and relatively fresh AABW was found EA6 7.2 8.4 offshore of the continental slope. MCDW with a neutral EA7 8.4 9.6 density range occupied most deep layers between AASW EA8 9.6 10.8 andAABW.MCDWenters theAdelie Depressionthrough Bathylagus sp B1 2.0 4.0 B2 4.0 6.0 the sill and follows the eastern side of the basin towards the B3 6.0 8.0 MGT. The SACCF (southern Antarctic Circumpolar B4 8.0 10.0 Front) and SB (Southern Boundary) were clearly observed B5 10.0 12.0 on transect one between station 16 and 18. B6 12.0 14.0 B7 14.0 16.0 B8 16.0 18.0 3.2. Diversity Cyclothone microdon C1 2.8 3.2 C2 3.2 3.6 Koubbi et al. (2010) have listed the species caught C3 3.6 4.1 with IYGPT during the surveys. Non-notothenioid C4 4.1 4.5 species which were caught offshore are given in C5 4.5 4.9 C6 4.9 5.3 Table 3. Myctophids had the highest number of species C7 5.3 5.7 (11) in the area. The other families were represented by C8 5.7 6.2 only one or two species. C9 6.2 6.6 Gymnoscopelus GO1 4.5 6.8 3.3. Influence of fishing depth opisthopterus GO2 6.8 9.0 GO3 9.0 11.3 GO4 11.3 13.5 The main species depth ranges are given in Fig. 4. GO5 13.5 15.8 Myctophids except for Gymnoscopelus species and GO6 15.8 18.0 Nannobrachium achirus showed the greatest vertical Protmyctophum bolini PB1 1.8 2.5 depth range, with K. anderssoni even being caught in the PB2 2.5 3.3 PB3 3.3 4.0 most shallow hauls at about 50 m. Cyclothone species PB4 4.0 4.8 also have a wide depth range, but these species do not PB5 4.8 5.5 occur in the epipelagic zone. The vertical distributions of PB6 6.3 7.0 the CEAMARC stations and the diel distributions for Krefftichthys anderssoni KA1 2.9 3.8 Bathylagus sp and the most common myctophid species KA2 3.8 4.6 KA3 4.6 5.5 (E. antarctica, P. bolini, K. anderssoni, G. opisthopte- KA4 5.5 6.4 rus) are given in Fig. 5. It must be noted that a compar- KA5 6.4 7.3 atively small number of stations were occupied during KA6 7.3 8.1 the cruise and that the IYGPT is a non-closing net. Gymnoscopelus braueri GB1 2.0 4.0 Specimens could have theoretically been taken between GB2 4.0 6.0 GB3 6.0 8.0 the maximum fishing depth and the surface, although the GB4 8.0 10.0 net deployment technique (see above) would tend to GB5 10.0 12.0 minimize this sampling error. K. anderssoni was the only species taken in the upper 50 m of the water during the Circumpolar Deep Water (CDW), which is less cold day; and most species were distributed at depths of (Fig. 2), occurred above this with an upper limit at 1000 m. 200 m and below. G. opisthopterus, in particular showed Surface Antarctic Water was found from 1000 m to the a much narrower range in vertical distribution, being surface, with temperatures ranging between 0.5 and captured at depths around 1000 m, with little evidence of 3.5 C. Fig. 3 shows vertical sections of potential any vertical migration. The paucity of material, together temperature and salinity along each transect as described with the fact that most stations were fished during the 200 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Fig. 2. Vertical profiles of temperature and salinity during the pelagic survey of the TRV Umitaka Maru in the Dumont d’Urville Sea during the CEAMARC survey in JanuaryeFebruary 2008 (from Koubbi et al., 2010). Transect 1 and 3 (Fig. 1) are represented here as they are the only ones over the oceanic zone. Fronts positions are indicated as white vertical dashed lines. The SACCF and the SB were located around station 17 and between stations 17 and 18, respectively. day, precludes further discussion of the diel distribution at 50, 200 and 500 m. For E. antarctica and P. bolini, results. abundances at 1000 and 500 m were different to those Table 4 gives the relevant statistics for abundant from the epipelagic zone. Finally, for K. anderssoni species, in which only the positive catch for each there were no significant differences among the depths. species is considered. E. antarctica and P. bolini The average number of species per sample is 1.8 occurred in the greatest number of samples: E. þ/ 3.5 but varies with depth. In the epipelagic zone, antarctica occurred in 12 (out of 25) samples and P. the number of species is often 0, but can reach bolini in 10. As shown in Fig. 4,anumberofspecies a maximum of 1 for 50 m and 4 for 200 m. The number were absent in the epipelagic zone. of species increased significantly in deeper layers with KruskalleWallis tests showed significant differences an average of 2.2 þ/ 2.3 species for 500 m and with of ranked abundances between standard sampling a maximum of 6 species. The highest number of depths for the abundant species. For Bathylagus sp,C. species is clearly for the 1000 m sampling depth with microdon, G. braueri and G. opisthopterus, abundances an average of 9þ/4.7 species and with a maximum of observed at 1000 m sampling depth differed from those 14 species observed. P. Koubbi et al. / Polar Science 5 (2011) 195e210 201

The KruskaleWallis test showed that the Shannon Index (H) differs significantly with depth when considering all tows carried out offshore. A Box-and- Whisker plot (Fig. 6) indicates that H is increasing significantly with the depth of sampling, and with major differences between the epipelagic zone and the mesopelagic zone not only in terms of the level of diversity, but also in terms of Interquartile range. The low value for the shallowest depth (50 m) can be explained by the by the presence of the single myctophid species, K. anderssoni. At 200 m, 6 species, mainly myctophids, dominate the samples: E. antarctica, is the most abundant species followed by P. bolini and K. anderssoni. At 500 m, E. antarctica is again the most dominant species, followed by Bathy- lagus sp and P. bolini. At 1000 m, 20 species are present with Bathylagus sp dominating in 35% of the catches followed by E. antarctica, C. microdon and G. Fig. 3. temperature and salinity diagram for offshore stations (8e42) opisthopterus. from the pelagic survey of the TRV Umitaka Maru in the Dumont Statistics for the Standard Length are given for the d’Urville Sea during the CEAMARC survey in JanuaryeFebruary 2008. Water masses observed are Antarctic Surface Water (AASW), adults and juveniles of fish species collected with the Circumpolar Deep Water (CDW), Modified Circumpolar Deep Water IYGPT (Table 5). Non-parametric tests on the sizes of (MCDW) and Antarctic Bottom Water (AABW). species occurring in more than one depth strata were achieved considering all stations. Significant differences were observed for Bathylagus sp. and E. antarctica, Table 3 whereas differences for C. microdon, G. braueri, K. Pelagic fish species caught with the IYGPT in the oceanic zone during anderssoni and P. bolini were not significant. the CEAMARC survey onboard the TRV Umitaka Maru in the Dumont d’Urville Sea in JanuaryeFebruary 2008. 3.4. Multivariate analysis on abundances Species Family Anotopterus pharao Anotopteridae Six groups were identified by the cluster analysis at Bathylagus antarcticus Bathylagidae the BrayeCurtis similarity level of 50% (Fig. 7). Bathylagus tenuis Chiasmodon niger Chiasmodontidae Group A comprised samples of the epipelagic zone Paradiplospinus gracilis Gempylidae (except for stations 8 and 9). Group B had samples Cyclothone microdon Gonostomatidae from 500 m (except for stations 9 and 10) and the Paraliparis terranovae Liparidae 200 m sample of station 9. Group C had the deeper Cynomacrurus piriei Macrouridae samples from 1000 m (except for station 9) and the Poromitra crassiceps Melamphaidae Muraenolepis microps Muraenolepididae 500 m sample of station 9. Groups D, E and F were Electrona antarctica Myctophidae separated from the other groups as they either have one Electrona carlsbergi species or a very dominant one (E. antarctica for Gymnoscopelus bolini groups D; Cyclothone spp. for group E and G. opis- Gymnoscopelus braueri thopterus for group F). Gymnoscopelus microlampas Gymnoscopelus nicholsi Species clustering (Fig. 7) clearly separates G. Gymnoscopelus opisthopterus opisthopterus (Group I) from the other species. This Krefftichthys anderssoni species was only present at 1000 m in station 9. The Lampanyctus macdonaldi remaining species can be separated into 3 different Nannobrachium achirus groups at the similarity level of 60%. Group II Protomyctophum bolini Scopelosaurus hamiltoni Notosudidae comprises Bathylagus sp., C. microdon and Cyclothone Oneirodes notius Oneirodidae spp. Two subgroups can be identified, the first one with Notolepis coatsi Paralepididae Cyclothone spp., the second one with Bathylagus sp. Benthalbella elongata Scopelarchidae Group III includes 2 myctophids species: E. antarctica Benthalbella macropinna and P. bolini. Finally, K. anderssoni was the only 202 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Fig. 4. Depth ranges of the main pelagic fish caught in the oceanic zone with IYGPT during the CEAMARC survey in the Dumont d’Urville Sea in JanuaryeFebruary 2008. species present in the Group IV which corresponded stations 8 and 9 at 1000 m. E. antarctica is present with the epipelagic zone. mainly in the positive part of axis 1, the size classes are more clearly separated along axis 2 with small sizes in 3.5. Size analysis the positive parts of axis 2 and 1 and the larger sizes toward the center of both axis. This trend seems to be The position of a depth/station in the multivariate related to a depth stratification of sizes with smaller space of the Correspondence Analysis (Fig. 8) is defined individuals more in the 500 m layer and the larger ones at using size classes for the species (Table 2): proximity 1000 m. However, large specimens were observed at reflects similar abundance while distance reflects scar- station 9 in 200 and 500 m. city. Axes 1 and 2 respectively explain 20.9% and 18.3% For this analysis, size classes 1 and 5 for K. ander- of the total inertia (39.2% for the factorial plane 1e2). ssoni were excluded because they unduly influenced the The analysis shows that stations that are more to the east analysis. The first size class was only observed at all and the south are separated in axis 1 (stations 8 and 9). depths of station 14, the most northern one and was the Axis 1 is also related to the sampling depth. Tempera- unique size class at 50 and 200 m depth. For the size class tures and bottom depth are also important and are related 5, it was only represented in station 14 (at 500 m) and to the positive values of axis 1. station 16 (at 200 m). The other size classes do not show For the cluster of observations in Fig. 9, axis 1 any particular trend. separates samples at 1000 m near the continental shelf P. bolini has its largest size classes separated from (stations 8, 9 and 20) in the negative part of the axis the smallest ones by axis 2. The remaining species from samples at 500 m (positive part) except for station Bathylagus sp., C. microdon, G. braueri have a less 9. Axis 2 separates mainly the sample at 500 m of clear representation in the axis. station 16 from the others. It should be noted that station 9 samples at 200 and 500 m are closer to the 4. Discussion deeper samples of stations 14, 16 and 18. For the cluster of variables in Fig. 10, two species 4.1. Sampling show clear size trends from the Correspondence Anal- ysis. G. opisthopterus is present only in the negative part Most of the previous studies in the Southern Ocean of axis 1, the size trend is expressed along axis 1 and 2. used RMT nets because they were more focused on The size classes become larger, when approaching Antarctic krill, E. superba (Hoddell et al., 2000; Flores P. Koubbi et al. / Polar Science 5 (2011) 195e210 203

Fig. 5. Diel distribution at oceanic stations for species from 22 CEAMARC IYGPT hauls. (A) Day-night hauls: symbols e open ¼ day; solid ¼ night; hatched ¼ crepuscular period. (B) Electrona antarctica: abundance ¼ 46.87e287.14 specimens.h-1; scale ¼ 100% default; symbols e open ¼ negative stations; solid ¼ positive stations. (C) Krefftichthys anderssoni: abundance ¼ 2.61e57.14 specimens.h-1; scale 100% default; scale and symbols as in B. (D) Protomyctophum bolini: abundance ¼ 3.00e37.14 specimens.h-1; scale 100% default; scale and symbols as in B. (E) Gymnoscopelus opisthopterus: abundance ¼ 48.58e273.00 specimens.h-1; scale 100% default; scale and symbols as in B. (F) Bathylagus sp.: abundance ¼ 5.71e480.00 specimens.h-1; scale 100% default; scale and symbols as in B. SMT ¼ ship’s mean time; Fdepth max ¼ maximum fishing depth in meters. et al., 2008; Van de Putte et al., 2010). Collins et al. Lu¨tzow-Holm Bay (Moteki et al., 2009), sampling was (2008) caught mesopelagic fish with an RMT25. done from the surface to 2000 m by RMT 1 þ 8, with However, most of these studies, except the one of Collins successful catches of E. antarctica (juvenile-subadults), et al. (2008) were focused on the epipelagic zone C. microdon and B. antarcticus. Besides the RMT, very (0e200 m) because they targeted E. superba. Catches of few studies used large pelagic trawls, as summarized by mesopelagic fish were then dependent on the time of the Koubbi et al. (2011). In this study, we used an IYGPT as day because of diel vertical migrations. Flores et al. a more appropriate gear to study larger fish. This gear (2008) considered that Bathylagus antarcticus and G. was also used with significant results in the Kerguelen braueri were poorly collected by RMT 1 þ 8 hauls in the archipelago by Duhamel (1998), Duhamel et al. (2000), epipelagic zone. Vertically stratified samples to deeper and by Collins et al. (2008). Off King George Island, zones were done, for example, by Moteki et al. (2009) Pusch et al. (2004) used a pelagic trawl (PT- 1088) with and by Collins et al. (2008). In the survey of the mouth dimensions of 20 m wide 10e12 m high and 204 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Table 4 RMTs. Some species like Notolepis coatsi were less Statistics of abundances (number of fish per hour) of fish collected efficiently collected by the IYGPT but were usually with the IYGPT during the CEAMARC survey onboard the TRV abundant in the RMT. Umitaka Maru in the Dumont d’Urville Sea in JanuaryeFebruary 2008. Calculations were only made on positive catch for each species. 4.2. Species diversity Species nb Mean Std Min Max station Pusch et al. (2004) pointed out that the Antarctic Bathylagus sp 7 205.49 161.13 5.71 480.00 Cyclothone sp 7 25.94 25.14 3.00 70.00 micronekton community has a diversity four times Cyclothone microdon 7 55.38 48.01 3.00 120.00 lower than subtropical and tropical oceans (Lancraft Cyclothone pseudopallida 3 4.89 2.60 2.86 7.83 et al., 1989). Hoddell et al. (2000) indicated that Cynomacruru piriei 4 5.16 5.38 1.00 13.04 during BROKE-East, no more than three or four Electrona antarctica 12 102.48 79.79 8.18 297.14 species were collected per sample, but this also Gymnoscopelus braueri 6 16.45 11.52 5.71 36.52 Gymnoscopelus opisthopterus 4 113.85 106.75 48.57 273.00 included neritic species. With IYGPT, the number of Krefftichthys anderssoni 8 15.42 18.15 2.61 57.14 species and the Shannon Index increased significantly Nannocrachium achirus 3 5.02 3.37 1.88 8.57 with depth, with a maximum of 6 species for the 500 m Poromitra crassiceps 3 11.28 10.82 2.86 23.48 samples and 14 for 1000 m. IYGPT was efficient in the Protomyctophum bolini 10 24.72 23.98 3.00 81.00 meso- and bathypelagic layers for studying adult fish diversity but care needs to be paid to the contamination of deeper samples by fish from upper layers during a mesh size of 12 mm in the cod end. By contrast and for deployment and retrieval. These fish from upper layers the CEAMARC survey, the IYGPT net has approximate could influence the estimation of abundance and mouth dimensions of 5.5 m high 12 m wide with consequently the Shannon Index. However, when e 20 mm 10 mm mesh size in the cod end. Pusch et al. comparing species richness between sampling layers, (2004) found a total of 18 mesopelagic fish species in we can see that deeper zones are clearly richer. 10 families. The present survey recorded 14 families, Deeper samples are characterized by a higher diversity with 11 species of myctophid alone. During the but also by a higher density. The oceanic fish community BROKE-East cruise, which covered part of our area, was dominated by the deep-sea smelts or bathylagids (479 Hoddell et al. (2000) found 3 families and only 6 species specimens), the lanternfishes (mainly E. antarctica e 431 of myctophids in the RMT samples from the epipelagic specimens, G. opisthopterus e 151 specimens and P. zone. From RMT samples in the epi-, meso- and bolini e 82 specimens) and Gonostomatidae (276 speci- bathypelagic zones, Moteki et al. (2009) found 7 fami- mens). C. microdon is often abundant deeper than 500 m lies, including 4 species of myctophid in Lu¨tzow-Holm but this vertical distribution pattern varies with Bay. Despite the fact that pelagic trawls are efficient geographical position (Gon and Heemstra, 1990). All samplers, meso- and bathypelagic fish were rare in the these species except G. opisthopterus were present at all samples from the epipelagic zone whereas larvae, post stations and showed the highest abundance in the samples. larvae or juveniles were well represented in epipelagic For bathylagids, two species were determined by barcoding (Dettai et al., 2011; Koubbi et al., 2010), these being B. antarcticus and Bathylagus tenuis. However, in most of the above studies only B. antarcticus is indicated. G. opisthopterus was only observed in samples from 1000 m and over the continental slope and rise (Stations 18, 20, 8 and 9). This species can live at around 500 m depth in more northern locations (Lancraft et al., 1989). E. antarctica, P. bolini and G. braueri are among the most common myctophid species in the Southern Ocean (Hulley, 1981)andE. antarctica one of the most abundant species in the mesopelagic zone as observed in other areas (Lancraft et al., 1989; Duhamel, 1998; Duhamel et al., 2000; Loots et al., 2007). Fig. 6. Box-and-Whisker Plot of the Shannon Index according to sampling depth calculated for oceanic pelagic fish collected with the Some stations are noteworthy in terms of their IYGPT from the TRV Umitaka Maru in the Dumont d’Urville Sea in higher species diversity. This is the case in station 16 JanuaryeFebruary 2008. where individuals of Chiasmodon niger, Oneiroides P. Koubbi et al. / Polar Science 5 (2011) 195e210 205

Table 5 Statistics for the Standard Length (cm) of fish collected by IYGPT by the TS “Umitaka Maru” in the oceanic zone of the CEAMARC area in JanuaryeFebruary 2008. Values are given for each standard sampling depth, and for all fish for each species. A non-parametric test was undertaken to compare standard lengths between depth strata. Species 1000 m 500 m 200 m 50 m Total Non-parametric p Nb. Mean Std Nb. Mean Std Nb. Mean Std Nb. Mean Std Nb. Mean Std Min Max test -value Anotopterus pharao 1 101.5 1 101.5 Bathylagus sp 127 10.3 3.2 70 8.7 2.8 2 6.1 0.1 199 9.7 3.1 2.6 17.5 0.00 Benthalbella elongata 1 19.0 1 19.0 Benthalbella macropinna 1 10.9 1 10.9 Chiasmodon niger 1 15.5 1 15.5 Cyclothone microdon 159 5.0 0.6 9 5.2 0.6 168 5.0 0.6 3.0 6.5 0.33 Cyclothone pseudopallida 4 3.8 0.2 4 3.8 0.2 3.5 4.0 Cynomacrurus piriei 9 30.2 6.2 9 30.2 6.2 21.5 37.3 Edentoliparis terraenovae 1 7.0 1 7.0 Electrona antarctica 145 6.1 1.6 177 5.2 1.7 31 6.3 0.9 353 5.7 1.7 2.1 9.9 0.00 Electrona carlsbergi 1 9.5 1 9.5 Gymnoscopelus bolini 2 20.5 6.4 2 20.5 6.4 15.9 25.0 Gymnoscopelus braueri 31 7.9 1.9 3 8.0 2.4 2 9.2 0.1 36 8.0 1.9 3.2 11.2 0.49 Gymnoscopelus 1 9.1 1 9.1 microlampas Gymnoscopelus nicholsi 1 12.4 1 12.4 Gymnoscopelus 88 10.8 3.4 88 10.8 3.4 4.6 16.5 opisthopterus Krefftichthys anderssoni 17 4.9 0.7 21 5.4 1.4 3 4.4 2.1 3 3.1 0.2 44 5.0 1.3 3.0 7.6 0.16 Lampanyctus macdonaldi 1 11.1 1 11.1 11.1 11.1 Lampanyctus sp 6 4.8 0.5 6 4.8 0.5 4.3 5.5 Nannobrachium achirus 5 10.7 1.1 5 10.7 1.1 9.1 12.0 Oneirodes notius 1 19.4 1 19.4 Paradiplospinus gracilis 1 35.5 1 35.5 Poromitra crassiceps 11 9.6 1.5 11 9.6 1.5 6.8 11.8 Protomyctophum bolini 12 4.4 0.7 52 4.3 0.5 4 4.3 0.2 68 4.3 0.5 2.3 6.5 0.71 notius and the myctophids Electrona carlsbergi and Multivariate statistical analysis on the abundances Lampanyctus macdonaldi were collected. The occur- of the most dominant species reveals groups of samples rence of the bi-temperate species L. macdonaldi at related to their sampled depths: a northern epipelagic station 16 represents the most southerly record for this zone, dominated by K. anderssoni, is distinctly sepa- species. In the Southern Hemisphere, it is distributed rated from the mesopelagic (500 m) and bathypelagic mainly between the Polar Front and the Subtropical zone (1000 m). Bathylagids, gonostomatids and G. Front (Hulley, 1981, 1990; McGinnis, 1982), although opisthopterus only occurred in the bathypelagic layers, these limits are crossed both to the north in eastern and whereas the remaining species such as E. antarctica western boundary currents, and to the south in the and P. bolini are present throughout the whole water Atlantic sector at 42W and 10W. column. There is a clear difference in hydrology and The northern part of transect 1 (stations 14 to 18) species composition between the two transects. The has species that are only present in this part of the 200 m sample of station 9 was grouped with the 500 m sampling transects. Beside those already noted for samples of the other stations. Similarly, the 500 m stations 16, the following species were taken: Ben- sample of station 9 was grouped with the 1000 m thalbella macropinna, Poromitra crassiceps and the samples of the other stations. This shift in vertical myctophids Gymnoscopelus nicholsi, N. achirus, and distribution at station 9 coincides with areas influenced K. anderssoni. The differences in diversity can be by the deep submarine canyon system that is just to the associated with the Southern Boundary of the ACC, north of the sill and separates the offshore zone from which is located between stations 16 and 18. However the deep George V Basin. This is the area where the results need to be considered with reservation, warmer and denser MCDW upwells to the shelf area. It since some species are only represented by few spec- is also where high salinity Shelf Water arising from the imens each. One species Cynomacrurus pirei only George V Basin enters the deep layers of the Southern occurred in transect 1. Ocean (Bindoff et al., 2000). 206 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Fig. 7. Multivariate analysis on log-transformed abundances of the dominant ﬁsh taxa of the oceanic zone investigated during the CEAMARC survey held in JanuaryeFebruary 2008 onboard the TRV Umitaka Maru. Clusters were obtained by using BrayeCurtis similarities with average grouping. Koubbi et al. (2010) have shown that there is a clear geographic segregation of oceanic and neritic pelagic ﬁsh assemblages in these surveys. Only three specimens of E. antarctica were caught over the sill separating the George V Basin from the continental slope, but none were found in deep innershelf depressions. No

Fig. 8. Correspondence analysis on size classes of ﬁsh collected in the oceanic zone in the CEAMARC area in JanuaryeFebruary 2008. Black circles indicates position of samples, open triangles the size classes of species and arrows supplementary variables such as Fig. 9. Cluster of observations from the Correspondence Analysis Longitude (Long), Latitude (Lat), Sampling Depth (SD), Bottom shown in Fig. 8. The samples are indicated by the station number and Depth (BD) and Temperatures (T). the sampling depth. P. Koubbi et al. / Polar Science 5 (2011) 195e210 207

year old fish according to Greely et al. (1999). Results of the Correspondence Analysis also show the importance of sampling depth and site on size structure. The continental slope and rise (stations 8, 9 and 20) are also separated from the northern most oceanic stations (14e18) because G. opisthopterus shows a clear size trend with size clearly increasing towards the continental shelf and particularly in the Jussieu canyon system. The other main result is the proximity of the shallower samples of station 9 with the deepest ones of the northern stations indicates similar size structures and that deep oceanic fauna is upwelled near the shelf, probably following the MCDW. The IYGPT data for oceanic species are severely limited in their spatial and temporal coverage (Fig. 5). Fig. 10. Cluster of variables from the Correspondence Analysis from Species other than K. anderssoni appeared to be absent Fig. 8. The size classes are indicated with the codes from Table 2. The black square is zoomed on the right window. from the upper 200 m, both during the day and also during the short austral-summer night (Fig. 5A). The mesopelagic fish larvae, juveniles or adults were apparent non-migratory or limited migratory behavior of collected in the very deep basins (1000 m) of the conti- K. anderssoni within the upper 200 m (Fig. 5C) both in nental shelf. Geographically, some of the submarine our data, and in samples off the east coast of Kerguelen canyons like the Jussieu canyon reach the shelf break (Duhamel et al., 2005), appears to be unique among and innershelf depressions (Caburlotto et al., 2006; De myctophids and may be related to the fact that, since the Santis et al., 2007; Beaman, 2008; Post et al., 2011) species is considered to be ancestral within the tribe and can be a way for species to migrate onto the shelf Electronini (Hulley, 1998), vertical migration in myc- area. Some abiotic factors such as water masses or tophids may be a derived behavior. E. antarctica did not presence of sea-ice can explain why the colonization of show diel patterning in the IYGPT datasets (Fig. 5B), but innershelf depressions is not possible. Some of these this was better observed in the RMT samples (Moteki species are probably highly linked to the MCDW and et al., 2011). Finally, there is confirmation that adult may be limited by the change of surface water conditions specimens of certain bathypelagic myctophid species do over the shelf due to shelf waters and sea-ice during not migrate (e.g. G. opisthopterus)(Fig. 5E), although winter. It should also be noted that the abundances of younger stages may do so. This is also probably the case mesopelagic species were low in the epipelagic zone, in bathylagids (Fig. 5F), where depth segregation of and that these species were confronted with a change of ontogenic stages has been observed, with larvae and pelagic fish assemblage, dominated by notothenioids juveniles in the upper layers and adults in deeper water mainly icefish larvae and juveniles and all stages of (Moteki et al., 2009). P. antarcticum (Moteki et al., 2011). Since E. antarctica diet is composed of the copepod Metridia gerlachei and 5. Conclusion the Antarctic krill E. superba (Pusch et al., 2004), the latter species also being the prey of adults of the The mesopelagic fish community was dominated by Antarctic silverfish P. antarcticum, we can assume that myctophids, which form the predominant group among there is a trophic competition between species in the the pelagic fish community of the oceanic region of the epipelagic zone over the shelf area. Southern Ocean. Major frontal zones clearly influence mesopelagic fish distribution, as has been described for 4.3. Vertical distribution the northern part of the Southern Ocean (Koubbi et al., 2011), but little is known about the influence of the SB Significant size differences were observed for bathy- and SACCF, which still suffer from a lack of small lagids between 500 and 1000 m and for E. antarctica at spatial scale surveys. In this study, the importance of the sampling depths of 200, 500 and 1000 m. For both area between stations 16 and 18, where these fronts were species, the size increases with sampling depth. The size observed, has been demonstrated. spectrum of E. antarctica is between 2.1 and 9.9 cm The continental slope and rise regions are where which corresponds to the length classes of one to three major hydrological features are seen, such as the slope 208 P. Koubbi et al. / Polar Science 5 (2011) 195e210 front. This study showed that the continental shelf break Beaman, R.J., 2008. A Bathymetric Digital Elevation Model (DEM) is a sharp boundary for meso- and bathypelagic fish of the George V and Terre Adelie Continental Shelf and Margin. even although bottom depth would allow for some of CAASM. http://gcmd.nasa.gov/KeywordSearch/Home.do? Portal¼amd_au&MetadataType¼0 [Data Set Text Search ¼ these species to migrate from the oceanic zone into the GVdem_2008]. deep innershelf depressions such as the George V Basin. Benzećri, J.-P., 1973. L’Analyse des Donneés. Tome 1: La Tax- Pelagic neritic fish are dominated by P. antarcticum inomie. Tome 2: L’Analyse des Correspondances, 2nd. e´d.). which occurs in all the depth layers from the epipelagic 1976. Dunod, Paris. zone to at least 500 m (Moteki et al., 2011). There is Bindoff, N.L., Rosenberg, M.A., Warner, M.J., 2000. On the circulation of water masses over the Antarctic continental slope and a single transitional area on the continental slope where rise between 80 and 1508E. Deep Sea Res. 47, 2299e2326. mesopelagic fish and neritic fish can interact; this is Caburlotto, A., De Santis, L., Zanolla, C., Cemerlenghi, A., Dix, J.K., where they forage on mesozooplankton and particularly 2006. New insights into Quaternary glacial dynamic changes on on dense swarms of E. superba. the George V Land continental margin (East Antarctica). e The importance of canyon systems of the conti- Quaternary Sc. Rev. 25, 3029 3049. Che´rel, Y., Fontaine, C., Richard, P., Labat, J.P., 2010. Isotopic niches nental slope has been demonstrated in this study. The and trophic levels of myctophid fishes and their predators in the Jussieu canyon is an area where complex hydrology Southern Ocean. Limnol. Oceanogr. 55, 324e332. and topography was observed. Here ALBW arising Che´rel, Y.,Koubbi, P., Giraldo, C., Penot, F., Tavernier, E., Moteki, M., from the shelf area mixes with upwelled MCDW. Ozouf-Costaz, C., Causse, R., Chartier, A., Hosie, G., 2011. Future submesoscale studies of the canyon systems Isotopic niches of fishes in coastal, neritic and oceanic waters off Ade´lie Land, Antarctica. CEAMARC Special Issue. Polar Sci. 5 should be planned because of their ecological impor- (2), 286e297. tance. These canyons are well mapped and some have Clarke, K.R., 1990. Comparisons of dominance curves. J. Exp. Mar. been classified by CCAMLR (Commission for the Biol. Ecol. 138, 143e157. Conservation of Antarctic Marine Living Resources) as Collins, M.A., Xavier, J.C., Johnston, N.M., North, A.W., Enderlein, P., Vulnerable Marine Ecosystems (Post et al., 2010). Tarling, G.A., Waluda, C.M., Hawker, E.J., Cunningham, N.J., 2008. Patterns in the distribution of myctophid fish in the northern Scotia Sea ecosystem. Polar Biol. 31, 837e851. Acknowledgments De Santis, L., Brancolini, G., Accettella, A., Cova, A., Caburlotto, A., Donda, F., Pelos, C., Zgur, F., Presti, M., 2007. New insights We would like to acknowledge the crew and captains into submarine geomorphology and depositional processes along of the TRV Umitaka Maru and specially the Japanese the George V Land continental slope and Upper Rise (East Antarctica), in: Cooper, A.K., Raymond, C.R. (Eds.), Keystone in cadets who helped us with sorting and processing the fish a Changing World, Online Proceedings of the 10th International samples. Thanks to Eric Tavernier for his help during the Symposium on Antarctic Earth Sciences. USGS Open-File Rep. survey. Funding sources were from CAML, TUMSAT, 2007-1047, Extended Abstract 061, 5 pp. Available at http://pubs. IPEV (ICOTA: Ichtyologie Coˆtie`re en Terre Ade´lie, PI: usgs.gov/of/2007/1047/ea/of2007-1047ea061.pdf. Philippe Koubbi and Catherine Ozouf-Costaz), AAD, Dettai, A., Lautredou, A.-C., Bonillo, C., Goimbault, E., Busson, F., Causse, R., Couloux, A., Cruaud, C., Duhamel, G., Denys, G., ANR GLIDES (PI: Charles-Andre´ Bost), ZA CNRS Hautecoeur, M., Iglesias, S., Koubbi, P., Lecointre, G., Moteki, M., Antarctique (PI: Marc Lebouvier). This work is Pruvost, P., Tercerie, S., Ozouf, C., 2011. The actinopterygian a contribution to CAML and SCAR-MarBIN. We would diversity of the CEAMARC cruises: barcoding and molecular like to acknowledge Professor Ishimaru from TUMSAT, taxonomy as a multi level tool for new findings. Deep Sea Res. 58, e the scientific leader who gave us the opportunity of 250 263. Duhamel, G., 1998. The pelagic fish community of the Polar frontal participating in this survey. We would like to thank Dr zone of the Kerguelen islands. In: Di Prisco, G., Pisano, E., Graham Hosie, the CEAMARC scientific leader for Clarke, A. (Eds.), Fishes of Antarctica: a Biological Overview. managing all the surveys and finding great collabora- Springer, Berlin Heidelberg, New York, pp. 63e74. tions between scientists from different countries. Duhamel, G., Koubbi, P., Ravier, C., 2000. Day and night mesopelagic fish assemblages off the Kerguelen islands (Southern Ocean). Polar Biol. 23, 106e112. References Duhamel, G., Gasco, N., Davaine, P., 2005. Poissons desˆ ıles Ker- guelen et Crozet, guide re´gional de l’oceán Austral. Patrimoines Badia, J., Do-Chi, T., 1976. Etude de la cine´tique de la structure des Naturels 63, 419. populations de Squilla mantis (Crustacea: Stomatopoda) par Flores, H., Van de Putte, A.P., Siegel, V., Pakhomov, E.A., van l’analyse factorielle des correspondances. Mar. Biol. 36, Franeker, J.A., Meesters, H.W.G., Volckaert, F.A.M., 2008. 159e169. Distribution, abundance and ecological relevance of pelagic Barrera-Oro, E., 2002. The role of fish in the Antarctic marine food fishes in the Lazarev Sea, Southern Ocean. Mar. Ecol. Prog. Ser. web: differences between inshore and offshore waters in the 367, 271e282. southern Scotia Arc and west Antarctic Peninsula. Antarctic Sci. Frontier, S., 1983. Strate´gies d’ećhantillonnage en ećologie. Oceanol. 14, 293e309. Acta. N Sp, 153e157. P. Koubbi et al. / Polar Science 5 (2011) 195e210 209

Gjosaeter, J., Kawaguchi, K., 1980. A Review of the World Lebart, L., Morineau, A., Feńelon, J.P., 1982. Traitement des donneés Resources of Mesopelagic Fish, FAO Fish. Tech. Paper No. 193. statistiques (me´thodes et programmes), vol. 13 Dunod, Paris, 151 pp. 502e515. Gon, O., Heemstra, P.C., 1990. Fishes of the Southern Ocean. J.L.B. Legendre, L., Legendre, P., 1984. Ecologie nume´rique, 2e`me e´d., Smith Institute of Ichthyology, Grahamstown, South Africa. Tome 2: La structure des donneés «ećologiques». Masson, Paris Gordon, A.L., Tchernia, P., 1972. Waters of the continental margin of et les Presses de l’Universite´ du Que´bec. Adelie coast, Antarctica. In: Hayes, D.E. (Ed.), Antarctic Loots, C., Koubbi, P., Duhamel, G., 2007. Habitat modelling of Research Series, Antarctic Oceanolgy II: The Australian-New Electrona antarctica (Myctophidae, Pisces) in Kerguelen by Zealand Sector. American Geophysical Union, Washington, pp. generalized Additive models and geographic information system. 59e69. Polar Biol. 30 (8), 951e959. Gower, J.C., 1987. Introduction to ordination techniques. Develop- Lubimova, T.G., Shust, K.V., Troyanovskij, F.M., Semenov, A.B., ments in numerical ecology. NATO ASI Ser. G 14, 3e64. 1983. The ecology of mass species of myctophids from Atlantic Greely, T., Gartner, J.V.J., Torres, J.J., 1999. Age and growth of Antarctica. Antarktika 22, 99e106. Electrona antarctica (Pisces: Myctophidae), the dominant McGinnis, R.F., 1982. Biogeography of Lanternfishes (family Myc- mesopelagic fish of the Southern Ocean. Mar. Biol. 133, tophidae) south of 30S. Ant. Res. Ser. 35, 1e110. 145e158. Moteki, M., Horimoto, N., Nagaiwa, R., Amakasu, K., Ishimaru, T., Hoddell, R.J., Crossley, A.C., Williams, R., Hosie, G.W., 2000. The Yamaguchi, Y., 2009. Pelagic fish distribution and ontogenetic distribution of Antarctic pelagic fish and larvae (CCAMLR vertical migration in common mesopelagic species of Lu¨tzow- Division 58.4.1). Deep-Sea Res. 47, 2519e2541. Holm Bay (Indian Ocean sector, Southern Ocean) during austral Hulley, P.A., 1981. Results of the research cruises of FRV ‘‘Walter summer. Polar Biol. 32, 1461e1472. Herwig’’ to south America. LVIII. Family Myctophidae Moteki, M., Koubbi, P., Pruvost, P., Tavenier, E., Hulley, P.A., 2011. (Osteichtyes. Myctophiformes). Arch. Fischereiwiss 31 (1), Spatial distribution of pelagic fish off Ade´lie and George V Land, 1e300. East Antarctica in the austral summer 2008. Polar Sci. 5 (2), Hulley, P.A., 1990. Myctophidae. In: Gon, O., Heemstra, P.C. (Eds.), 211e224. Fishes of the Southern Ocean. J.L.B Smith Institute of Ichtyo- Nicol, S., Pauly, T., Bindoff, N.L., Strutton, P.G., 2000. “BROKE” logie, Grahamstown, pp. 146e176. a biological/oceanographic survey off the coast of East Antarctica Hulley, P.A., 1992. Upper-slope distributions of oceanic lanternfishes (80e150E) carried out in JanuaryeMarch 1996. Deep-Sea Res. (family: Myctophidae). Mar. Biol. 114 (3), 365e383. 47, 2281e2297. Hulley, P.A., 1998. Preliminary investigations on the Evolution of the Orsi, A.H., Johnson, G.C., Bullister, J.L., 1999. Circulation, mixing tribe Electronini (Myctophiformes, Myctophidae). In: Di Prisco, G., and production of Antarctic bottom water. Prog. Oceanogr 43, Pisano, E., Clarke, A. (Eds.), Fishes of Antarctica: A Biological 55e109. Overview. Springer, Berlin Heidelberg New York, pp. 75e85. Orsi, A.H., Whitworth, T., Nowlin Jr., W.D., 1995. On the meridional Jacobs, S.S., 1991. On the nature and significance of the Antarctic extent and fronts of the Antarctic Circumpolar Current. Deep-Sea slope front. Mar. Chem. 35, 9e24. Res. 42, 641e673. Kellermann, A., 1996. Midwater fish ecology. In: Ross, R.M., Pakhomov, E.A., Perissinotto, R., McQuaid, C., 1996. Prey compo- Hofmann, E.E., Quetin, L.B. (Eds.), Foundations for Ecological sition and daily rations of myctophid fishes in the Southern Research West of the Antarctic Peninsula. Antarct. Res. Ser, 70, Ocean. Mar. Ecol. Prog. Ser. 134, 1e14. pp. 231e256. Piatkowski, U., Rodhouse, P.G., White, M.G., Bone, D.G., Kock, K.H., 1992. The Composition of the Fish Fauna. Cambridge Symon, C., 1994. Nekton community of the Scotia Sea as University Press, Cambridge. sampled by RMT 25 during austral summer. Mar. Ecol. Prog. Ser. Koubbi, P., Moteki, M., Duhamel, G., Goarant, A., Hulley, P.-A., 112, 13e28. Ishimaru, T., Pruvost, P., Tavernier, E., Vacchi, M., Hosie, G., Post, A.L., O’Brien, P.E., Beaman, R.J., Riddle, M.J., De Santis, L., 2011. Ecoregionalization of myctophid fish in the Indian sector of 2010. Physical controls on deep water coral communities on the the Southern Ocean: results from generalized dissimilarity George V Land slope, East Antarctica. Antarctic Sci. 22, 371e378. models. Deep Sea Res. 58, 170e180. Post, A.L., Beaman, R.J., O’Brien, P.E., Eleáume, M., Riddle, M.J., Koubbi, P., O’Brien, C., Loots, C., Tavernier, E., Vacchi, M., 2011. Community structure and benthic habitats across the Vallet, C., Chevallier, J., Moteki, M., 2011. Spatial distribution George V Shelf, East Antarctica: trends through space and time. and interannual variations in the size frequency distribution and Deep Sea Res. 58, 105e118. abundances of Pleuragramma antarcticum larvae in the Dumont Pusch, C., Hulley, P.A., Kock, K.-H., 2004. Community structure and d’Urville Sea from 2004 to 2010. CEAMARC Special Issue. feeding ecology of mesopelagic fishes in the slope waters of King Polar Sci. 5 (2), 225e238. George Island (South Shetland Island’s, Antarctica). Deep-Sea Koubbi, P., Ozouf-Costaz, C., Goarant, A., Moteki, M., Hulley, P.A., Res. 51, 1685e1708. Causse, R., Dettai, A., Duhamel, G., Pruvost, P., Tavernier, E., Rintoul, S.R., 1998. On the origin and influence of Ade´lie Land Post, A., Beaman, R.J., Rintoul, S., Hirawake, T., Hirano, D., bottom water. In: Jacobs, S., Weiss, R. (Eds.), Ocean, Ice, and Ishimaru, T., Riddle, M., Hosie, G., 2010. Estimating the biodi- Atmosphere: Interactions at the Antarctic Continental Margin. versity of the East Antarctic shelf and oceanic zone for ecor- Antarctic Research Series 75. Am. Geophys. Union, Washington, egionalisation: example of the ichthyofauna of the CEAMARC pp. 151e171. (Collaborative East Antarctic Marine Census) CAML surveys. Torres, J.J., Somero, G.N., 1988. Vertical distribution and metabo- Polar Sc 4, 115e133. lism in Antarctic mesopelagic fishes. Comp. Biochem. Physiol. Lancraft, T.M., Torres, J.J., Hopkins, T.L., 1989. Micronekton and A90, 521e528. macrozooplankton in the open waters near Antarctic ice-edge Tynan, C.T., 1998. Ecological importance of the southern boundary zones (AMERIEZ 1983 and 1986). Polar Biol. 9, 225e233. of the Antarctic circumpolar Current. Nature 392, 708e710. 210 P. Koubbi et al. / Polar Science 5 (2011) 195e210

Van de Putte, A.P., Jackson, G.D., Pakhomov, E., Flores, H., Whitworth, T., Orsi, A.H., Kim, S.J., Nowlin Jr., W.D., 1998. Water Volckaert, F.A.M., 2010. Distribution of squid and ﬁsh in the masses and mixing near the Antarctic Slope Front. In: Jacobs, S.S., pelagic zone of the Cosmonaut Sea and Prydz Bay region Wiess, R.F. (Eds.), Ocean, Ice, and Atmosphere: Interactions at the during the BROKE-West campaign. Deep-Sea Res. 57, Antarctic Continental Margin. Antarctic Research Series. Amer- 956e967. ican Geophysical Union, Washington, pp. 1e27.