Oceanographic and topographic conditions structure benthic meiofauna communities in the Weddell Sea, Bransfield Strait and Drake Passage (Antarctic)

Accepted Manuscript Oceanographic and topographic conditions structure benthic meiofauna communities in the Weddell Sea, Bransfield Strait and Drake Passage (Antarctic) Gritta Veit-Köhler, Stephan Durst, Jan Schuckenbrock, Freija Hauquier, Laura Durán Suja, Boris Dorschel, Ann Vanreusel, Pedro Martínez Arbizu PII: DOI: Reference:

S0079-6611(16)30204-X https://doi.org/10.1016/j.pocean.2018.03.005 PROOCE 1936

To appear in:

Progress in Oceanography

Received Date: Revised Date: Accepted Date:

5 October 2016 5 March 2018 6 March 2018

Please cite this article as: Veit-Köhler, G., Durst, S., Schuckenbrock, J., Hauquier, F., Durán Suja, L., Dorschel, B., Vanreusel, A., Martínez Arbizu, P., Oceanographic and topographic conditions structure benthic meiofauna communities in the Weddell Sea, Bransfield Strait and Drake Passage (Antarctic), Progress in Oceanography (2018), doi: https://doi.org/10.1016/j.pocean.2018.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Oceanographic and topographic conditions structure benthic meiofauna communities in the Weddell Sea, Bransfield Strait and Drake Passage (Antarctic)

Gritta Veit-Köhler1*, Stephan Durst1,2, Jan Schuckenbrock1,2, Freija Hauquier3, Laura Durán Suja3,4, Boris Dorschel5, Ann Vanreusel3, Pedro Martínez Arbizu1

1

Senckenberg am Meer, German Centre for Marine Biodiversity Research DZMB,

Südstrand 44, D-26382 Wilhelmshaven, Germany 2

Carl von Ossietzky University Oldenburg, Department of Biology and Environmental

Science, Biodiversity and Evolution, D-26111 Oldenburg, Germany 3

Marine Biology Research Group, Ghent University, Krijgslaan 281/Sterre S8, B-

9000 Ghent, Belgium 4

School of Life Sciences, Heriot-Watt University, Edinburgh, EH14 4AS, United

Kingdom 5

Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung, Van-

Ronzelen-Str. 2, D-27568 Bremerhaven, Germany

* Corresponding author: [email protected] Tel. 0049 (0) 4421 9475 102

1

Fax 0049 (0) 4421 9475 111

Data

are

available

from

the

PANGAEA

online

repository

at

https://doi.org/10.1594/PANGAEA.882582 (Veit-Köhler et al. 2017).

Abstract The marine environment of the tip of the Antarctic Peninsula is characterised by three oceanographically distinct regions for which we linked continental-slope meiofaunal patterns and environmental drivers on a large scale (100–300 km among ecoregions). Samples for meiofauna communities and sediment analyses were collected with a multicorer, water-column data were derived from water samples and CTD recordings. Meiofauna communities including individuals from 19 higher taxa were compared to a set of 16 environmental variables. We detected significant differences between the communities of Weddell Sea and those of Bransfield Strait and Drake Passage. The amount of phytopigments in the sediment, their freshness and the silt and clay content were driving factors for this separation. The highest meiofauna abundances were found at slopes in the Weddell Sea. Food banks may facilitate high standing stocks. There, the highest ever recorded copepod percentages for the Antarctic were related to the highest phytopigment contents while nematodes were extremely abundant even in deeper sediment layers at stations with fresh organic material. For Bransfield Strait and Drake Passage a sampling scheme of slopes and adjacent troughs was applied. The two regions were divided into three geographical “areas” with the two “habitat” types investigated for each area. 2

Multivariate non-parametric permutational analysis of variance (PERMANOVA) showed that in Bransfield Strait slope and trough meiofauna communities differed significantly in all geographical areas while in Drake Passage this was only the case in the East. These differences were explained best by the regionally and topographically distinct characteristics of 7 out of 11 water-column and sedimentbound factors related to sediment grain size, food quantity and quality, water temperature and salinity. Environmental drivers of the benthic habitat are dependent on large-scale oceanographic conditions and are thus sensitive to changes in water mass characteristics, sea-ice cover and the related primary production.

Keywords meiobenthos; Nematoda; Copepoda; environmental settings; water masses; current systems

Regional terms Southern Ocean, Antarctic Peninsula, South Shetland Islands

1. Introduction

The Antarctic Peninsula is among the most rapidly warming regions world wide (e.g., King 1994). The impact of global change already caused considerable loss of ice3

shelves and glaciers (Cook et al. 2005) as well as declining sea-ice cover along the western Antarctic Peninsula (King 2014). The marine environment around the Antarctic Peninsula is thus a key region for the study of biological processes and the response of biota to environmental change (Ingels et al. 2012; Kaiser et al. 2013 and references therein). With its low temperatures and pronounced seasonal changes in primary production and food availability the marine ecosystem of the Antarctic is unique. However, the complex interactions between seasonal environmental changes and marine taxa are still poorly known (Griffiths 2010). Additionally, the reaction of biota to future environmental conditions and the consequences for ecosystem functioning are far from being predictable (Kennicutt II et al. 2014, Gutt et al. 2016). 1.1 Key characteristics of the study region Along the Antarctic Peninsula the three investigated regions Weddell Sea, Bransfield Strait, and Drake Passage can be distinguished by their different environmental settings. The questions addressed in this study are whether, and how, these different environmental settings influence benthic communities. Currents and temperature. The Antarctic Circumpolar Current transports relatively warm and nutrient-rich Circumpolar Deep Water to the continental shelf in Drake Passage west of the South Shetland Islands (Hofmann et al. 1996). These water masses enter Bransfield Strait through a deep trough between Brabant and Smith Islands and nutrients and biota are transported north-east along the eastern side of the South Shetland Islands (Zhou et al. 2002; Zhou et al. 2006). Weddell Sea Water enters Bransfield Strait mainly north-east of Joinville Island and flows in southwesterly direction along the Antarctic Peninsula (Huneke et al. 2016; Zhou et al.

4

2002). Cold Weddell Sea Water predominates at the eastern side of the Antarctic Peninsula (Clarke et al. 2009). Primary production and sea-ice cover. During the expedition (January to March 2013) a phytoplankton bloom developed in Drake Passage and the northwestern part of Bransfield Strait (Hauquier et al. 2015). However, long-term chlorophyll a data show that overall water-column productivity is highest in Weddell Sea (Dorschel et al. 2016). In general, sea-ice concentration and duration of ice cover are gradually increasing from Drake Passage to Weddell Sea. An overall slight increase in sea-ice cover was observed when averages of the years 2008–2012 were compared to averages of 2003–2012. In 2012, sea-ice concentration augmented especially south of Joinville Island and in Bransfield Strait as compared to the long-term averages (Dorschel et al. 2016). Sedimentary processes. Seasonal blooming events lead to rapid sinking of primary production to the sea floor due to aggregates and fecal pellets (Gooday 1993; Ragueneau et al. 2000). A wide array of sedimentary processes is current-driven and organic matter can be transported laterally and deposited in areas that do not correspond to the area of production (Yoon et al. 1992; Palanques et al. 2002; Isla et al. 2004). The same is true for terrigenous sediments from glacial and melt-water runoff that are deposited in sea-floor depressions and troughs along the Antarctic Peninsula (Prieto et al. 1999). In our study area troughs are mainly found in Bransfield Strait and Drake Passage. Troughs, canyons, and slopes. The Antarctic shelf as a whole is deeper than the shelves of other continents due to isostatic depression by the huge weight of the continental ice sheets. Erosion formed its submarine landscape with slopes, troughs and canyons that hold unique local sediment characteristics (Prieto et al. 1999; Isla et 5

al. 2004). Canyons are found world-wide along continental shelves. They are Vshaped in profile, often start within the shelf far away from the shelf break and disembogue into abyssal plains (Harris and Whiteway 2011). Contrary to canyons, troughs are of glacial origin, incised into the continental shelf, generally U-shaped in profile and with a raised sill at their seaward terminus (Shepard 1981). Despite their differing origin and topography, troughs may share some important characteristics of canyons: canyons may be influenced by strong currents and as they are located along the productive continental shelves they may act as conduits for organic material and sediment (Puig et al. 2000; Danovaro et al. 2010). Meiofauna (and microfauna) play an important role for the remineralisation of organic matter at the sea floor (e.g., Renaud et al. 2007). The small metazoan sedimentdwelling animals (by definition for marine benthic meiofauna between 32 µm and 1 mm; Soltwedel 2000) represent an ecological group which contrary to most macrofauna taxa does not have pelagic larval stages. The habitat determines the inventory of species of a certain body size and whether their lifestyle is burrowing (endobenthic), interstitial (among sand grains) or epibenthic (Remane 1933). By far the most abundant meiofauna taxon is the nematodes, followed by harpacticoid copepods and a number of other taxa such as kinorhynchs, ostracods, gastrotrichs, tardigrades, and acari. Antarctic meiofauna communities were studied in shallow waters by e.g., Vanhove et al. (2000), de Skowronski and Corbisier (2002), and Pasotti et al. (2014). Several meiofauna studies concentrated on intermediate depths down to 2000 m (e.g., Herman and Dahms 1992; Vanhove et al. 1995; Fabiano and Danovaro 1999; Rose et al. 2014) or mainly abyssal areas (e.g., Gutzmann et al. 2004; Lins et al. 2014; Veit-Köhler et al. 2011). No comprehensive comparative study

6

on the meiofauna communities of the three environmentally differing regions along the Antarctic Peninsula has been carried out to date. Meiofauna communities are shaped to a great part in response to food availability. Continental slopes are characterised by high organic input as organic material and sediment are transported down-slope from the shelf to the neighbouring abyssal plains (Walsh 1991; Danovaro et al. 2009b). Slopes and canyons may hold very distinct conditions for benthic organisms (Garcia et al. 2007). Meiofauna from marine canyons and adjacent slopes have been studied e.g., along the western Iberian, Celtic, and Mediterranean margins (Garcia et al. 2007; Grémare et al. 2009; Danovaro et al. 2009a; Ingels et al. 2011; Romano et al. 2013) or from the western Greenland Sea and the Gulf of Guinea (Soltwedel et al. 2005; Van Gaever et al. 2009). Information on meiofauna communities from channels and troughs is scarcer (e.g., Mississippi Trough and adjacent canyon features: Baguley et al. 2006; Gollum Channel, Celtic Margin: Ingels et al. 2011). In many parts of the world macro- and megabenthic canyon communities have been investigated (Northwest Atlantic: Rowe 1971; off southeastern New Zealand: Probert et al. 1979; off California: Vetter and Dayton 1998; Northeast Atlantic: Duineveld et al. 2001; off southeastern Australia: Schlacher et al. 2007). However, for the Antarctic comparative macro- and megafauna data exist only for three trough/slope systems in Bransfield Strait (Ambroso et al. 2016; Gutt et al. 2016; Kersken et al. 2016; Michel et al. 2016; Segelken-Voigt et al. 2016) and three fjord/open shelf systems further south along the West Antarctic Peninsula (Grange and Smith 2013). Meiofaunal studies on canyons or troughs and the adjacent slopes are new for the Antarctic. In our study area canyons are mainly found in Drake Passage from the

7

shelf break down to the deep sea (Fig. 1). We focused on upper slopes and (in Drake Passage and Bransfield Strait) neighbouring troughs. The novelty of our study is the comprehensive linkage of patterns in meiofauna communities and environmental drivers on continental slopes of the three oceanographically very distinct regions Weddell Sea, Bransfield Strait, and Drake Passage (Fig.1). Thus a large geographical coverage is accomplished along the Antarctic Peninsula. A detailed comparison of meiofauna of slopes and troughs in Bransfield Strait and Drake Passage and the structuring role of topographic conditions complete the study. During the RV Polarstern expedition PS 81 ANT XXIX-3 (22.01–18.03.2013) the following hypotheses were tested: (1) Meiofauna communities of continental slopes in Weddell Sea, Bransfield Strait and Drake Passage differ according to the oceanographic regimes in these regions. We expect high standing stocks of meiofauna as a consequence of high primary production and differences in community composition in relation to sediment and food characteristics. (2) Meiofauna communities of topographically distinct structures (slopes and troughs) are different. We expect a complex pattern related to environmental factors that change with region (Bransfield Strait and Drake Passage) and habitat type.

2. Material and methods

8

2.1 Study sites and sampling During the RV Polarstern expedition PS 81 ANT-XXIX/3 (22.01–18.03.2013) samples for the study of meiofauna communities and environmental characteristics were collected at 16 stations along the continental slopes of three regions in the vicinity of the Antarctic Peninsula (Hauquier and Veit-Köhler 2013, Fig. 1). In the regions Weddell Sea, Bransfield Strait and Drake Passage several target areas from 220 to 758 m water depth were investigated. Designations for locations are combined from abbreviations for regions (BS = Bransfield Strait, DP = Drake Passage, WS = Weddell Sea), areas (W = West, C = Central, E = East, ET = Erebus and Terror Gulf, JE = Joinville Island East, JN = Joinville Island North) and habitats (B = Bank, D = Deep, S = Slope, T = Trough), e.g., BS-W-S = Bransfield Strait West Slope. Station information and abbreviations are listed in Table 1. Sediment samples were taken with the multicorer (MUC) equipped with 12 plexiglas core liners (inner diameter 57 mm, surface area 25.5 cm2). At each location (with the exception of station BS-C-S, see Table 1) three MUC deployments were carried out in order to collect true replicates. From each deployment sediment samples were collected for different analyses. Samples for meiofauna community analysis were sliced in 1 cm-layers down to 5 cm depth and stored in a 4% formaldehyde-seawater solution (borax-buffered), 2 cm of overlying water were unified with the first centimetre. Other cores were subsampled for environmental parameters with cut-off 10 ml syringes that were pushed into the sediment: 1 syringe for grain size and 1 for total organic carbon (TOC) and total nitrogen (TN) determination (stored at -20°C), 1 syringe for pigment analyses (stored at -80°C). Water column characteristics were measured with a CTD (Schröder et al. 2013a, b, c) and from water samples collected with Niskin bottles mounted on the CTD rosette 9

(chloroplastic pigments). At each CTD station water-column parameters were measured for water depths representing the chlorophyll a maximum (Chlamax) and near bottom layers (bottom water; see Table 1 for CTD station and depth information). Prior to filtration water samples were poured over a 100-µm sieve in order to remove larger particles. Water was then filtered over glass fibre filters GF/C at approximately 250 mbar to avoid rupture of cells. The colouring of the filters determined the amount of sea water used (3–5 l). The GF/C filters were stored at 80°C.

2.2 Environmental parameters: sediment and water column An overview of measured environmental variables, calculated ratios, the used acronyms and a short guide to the interpretation of the results are given in Table 2. Prior to all sediment analyses the respective subsamples collected with syringes were sliced in 1-cm slices (0–1 cm, 1–2 cm, 2–3 cm, 3–4 cm, 4–5 cm). Sediment grain sizes were determined with a Malvern Mastersizer 2000 (particle size range 0.02–2000 µm) and expressed as relative percentages of the different size fractions according to Wentworth (1922). For better handling, fractions were summed to restrict their number to three: Silt&Clay (<63 µm), Sand (63–500 µm) and Coarse Sand (sand fraction >500 µm). Total organic carbon (TOC) and total nitrogen (TN) were determined from freeze-dried samples by elemental analysis [Flash 2000 organic elemental analyser (protocol available through Interscience B.V., Breda, The Netherlands)]. Prior to combustion samples were acidified with 2% HCl in order to remove inorganic carbon. The molar C/N ratio as an indicator of the degradation state of organic matter was calculated as TOC/TN*14/12. Pigments were extracted 10

with 10 ml acetone (90%) from lyophilised sediment, separated by HPLC (High Performance Liquid Chromatography) and determined with a fluorescence detector. Chlorophyll a (Chla) and its degradation products the phaeopigments (Phaeo) were expressed in µg g-1 and summed as CPE (Chloroplastic Pigment Equivalents). The ratio of Chla/Phaeo was used as an indicator for the freshness of primary products. Chloroplastic pigments in the water column (CmaxChla at the chlorophyll maximum, bottomChla

near the sea bottom) were extracted with 10 ml acetone (90%) from GF/C

filters, determined using a fluorimeter and expressed in µg l-1 (equivalent to mg m-3).

2.3 Meiofauna treatment Fixed sediment samples of stations WS-JE-D [presented as W-120 in Hauquier et al. (2015)], WS-ET-D (W-163), DP-C-T (DP-243), and DP-E-T (DP-250) were treated as described by Hauquier et al. (2015). The sediment slices of the remaining samples were washed with filtered tap water through a 32-µm mesh sieve. Meiofauna and organic material were extracted from the sediment by centrifugation with a colloidal silica polymer (Levasil, H.C. Stark, 200/40%) as the flotation medium and kaolin to prevent the sediment particles from contaminating the sample during decantation (McIntyre and Warwick 1984). The centrifugation was repeated 3 times at 4000 rpm for 5 min. After each centrifugation step, the floating matter was decanted over a 32µm mesh sieve and rinsed with tap water. This supernatant normally contains all organic material and animals present in the sample. Samples were stained with rose bengal. Before sorting, the samples were decanted over a 1-mm mesh sieve to separate meiofauna from macrofauna. Meiofauna organisms were determined to higher taxon level with the aid of Leica Mz 12.5 and Mz 125 stereo microscopes according to the pictorial key of Higgins and Thiel (1988). Apart from copepods, 11

which were subsequently transferred into glycerine, members of all other meiofauna groups were counted, remained in the samples and were stored in 70% ethanol. The counts of unidentified larvae and other organisms were unified as “Others”. Furthermore, Turbellaria (most of the time amorphous and unidentifiable when fixed in formalin) and Tantulocarida (only identified at the DZMB) were added to the category “Others”.

2.4 Statistical analyses Statistical analyses were carried out using SigmaPlot 11 and the PRIMER v6 software with the PERMANOVA+ add-on package (Clarke and Gorley 2006; Anderson et al. 2008). Figures were edited using Macromedia FreeHand MXa 11.0.2. For the comparison of environmental data among the three regions, oceanographic and biological water-column measurements (Chla, salinity Sal, water temperature T of the chlorophyll maximum layer and bottom water) and sedimentary factors (Silt&Clay, Sand, Coarse Sand, TOC, TN, C/Nmolar, Chla, Phaeo, CPE, Chla/Phaeo) were used. In order to address hypothesis (1) sediment data were averaged from 0 to 5 cm depth. Environmental characteristics were described by means of Principal Component analysis (PCA). Draftsman plots were used to visualise relationships among variables. When strong correlations among ecologically similar variables were observed one variable was left out. Skewed variables were transformed (sqr or 4throot) and all variables were normalised before PCA was carried out. Differences in meiofauna communities among the three regions (WS, BS, DP) [hypothesis (1)] were tested with one-way analyses of similarity (ANOSIM with 9999 permutations) by using (1a) the total of all sediment layers per core per taxon (taxa 12

as variables, factor “region”) and (1b) individual numbers of all taxa summed per depth layer (0–1, 1–2, 2–3, 3–4, 4–5 cm as variables, factor “region”). Bray-Curtis similarity was applied as resemblance measure. For the community test (1a) abundance data were square-root transformed. Non-metric multidimensional scaling (MDS) was used to visualise similarities among community cores. To detect whether there were differences among meiofauna communities of slopes and troughs [hypothesis (2)] multivariate non-parametric permutational ANOVA (PERMANOVA with 9999 permutations) was applied. A hierarchical PERMANOVA was performed with three factors (“region”, fixed = BS, DP; “area” nested in “region”, random = W, C, E; “habitat” nested in “area” nested in “region”, random = S, T). Station BS-JN-S was omitted from the analysis as no corresponding trough station was sampled in this area. Pseudo-F ratio and P(perm) value were calculated using type III sums of squares and permuting residuals under a reduced model. Where only a restricted number of unique permutations <100 was possible, p-values were obtained from Monte Carlo samplings (Anderson and Robinson 2003). Post-hoc pairwise tests for significant factors were carried out when distance-based PERMDISP tests (distances to centroid, P-values from permutation of least-squares residuals) confirmed homogeneity of sample dispersions in the multivariate space. This test is essential as significant results for a given factor from PERMANOVA can have several reasons: groups may differ in their location in the plot, in their dispersion or in a combination of the two (Anderson et al. 2008). The group of taxa and unidentified organisms unified as „Others” was excluded from statistical analyses as they represented a taxonomically inhomogeneous group. Multivariate environmental patterns were linked to community patterns among the three regions [hypothesis (1)] with the help of a BEST analysis (Spearman Rank 13

correlation). Based on the Bray-Curtis similarity of the meiofauna communities (see above) and the Euclidean distance for the transformed and normalised environmental data (see above) this analysis tests by permutation for the variable combination that best explains the community patterns. Distance-based linear models (DISTLM) and the visualisation by distance-based redundancy analysis (dbRDA graph) were used to detect the structuring role of environmental variables in trough and slope communities of BS and DP [hypothesis (2)]. The DISTLM routine (selection criterion: adj. R2; selection procedure: forward) analysed and modeled the relationship between the square-root transformed multivariate community data described as Bray-Curtis resemblance matrix and 11 environmental predictor variables (selection procedure for environmental variables see PCA). P-values for testing the null hypothesis of no relationship were obtained using permutation techniques. The dbRDA routine performed an ordination of the fitted values from the given model in the multi-dimensional space (Anderson et al. 2008).

3. Results

3.1 Environmental characterisation: sediment and water column The silt and clay content was higher in Weddell Sea (WS) and Drake Passage (DP) (80.3 and 88.1%, Table 3) than in Bransfield Strait (BS) (64.7%). WS differed clearly from BS and DP with 27 to 125 times higher levels of CPE and a 6 to 18 times higher 14

Chla/Phaeo ratio. These results indicate a larger and fresher stock of deposited primary production in the sediment of the WS. Contrary to the Chla content of the sediment, Chla in the water column was highest in the chlorophyll maximum layers of DP and lowest in WS. Increasing sea water temperatures were observed from WS over BS to DP at the chlorophyll maximum (difference WS to DP 2.8°C) and in near bottom water (difference WS to DP 2.4°C). Salinity at the chlorophyll maximum slightly decreased from WS to DP while the opposite was true for the bottom water. Within the regions, sediment C/Nmolar ratio and TN% played a structuring role and can be used to distinguish among sites with fresh and sites with more degraded organic material (Fig. 2). The PCA plot for all available replicates within the stations of the three regions was based on data of 11 normalised environmental variables (Fig. 2): Silt&Clay, Coarse Sand* (sqr transformed), TN, C/Nmolar, CPE* (4throot transformed), Chla/Phaeo* (sqr transformed), CmaxChla,

bottomChla,

CmaxSal,

bottomSal,

CmaxT. The water column

variable bottomT and the sediment variables Sand, TOC, Chla, and Phaeo were left out of the analysis due to their correlations with other ecologically closely related variables (based on nearly linear relationships visualised in Draftsman plots, e.g. the sum of Chla and Phaeo is CPE, their relation is Chla/Phaeo; the amount of Sand is reversely related to the Silt&Clay content; TOC is represented by the closely related TN and the C/Nmolar relationship; compare Table 2). The first three PC axes described 79% of the detected variation. Along PC1 the three regions were mainly separated by CmaxChla (coefficient -0.404) and CmaxT (-0.429). Within the regions Silt&Clay (0.552) and Coarse Sand* (0.585) (PC2), as well as TN (-0.594) and bottomChla (0.532) (PC3) were detected to be structuring factors.

15

3.2 Meiofauna communities of the three regions and their relation to environmental factors [hypothesis (1)] In the sediment samples collected for meiofauna community analyses altogether 542,011 individuals were recorded. They belonged to 19 higher taxa and copepod nauplii, 462 specimens could not be determined. Copepod nauplii were counted separately due to ecological differences in size and diet compared to adults and copepodids (Decho and Fleeger 1988). Nematodes dominated the samples at all stations, followed by copepod nauplii and copepods. All other taxa were found in low numbers (<1% of total abundance), here listed in descending order: annelids, kinorhynchs, ostracods, bivalves, priapulids, gastrotrichs, tardigrades, tanaids, acari, coelenterates, isopods, rotifers, loriciferans, amphipods, cumaceans, cladocerans, and ophiurids. A one-way ANOSIM of the pooled data of all sediment layers per core [hypothesis test (1a), 0–5 cm, taxa as variables, factor “region”] revealed significant differences among the regions (Global R = 0.44, P = 0.01%, sample distribution visualised as MDS; Fig. 3). Pairwise tests showed that BS and DP differed from WS (DP vs WS: R = 0.894, P = 0.01%; BS vs WS: R = 0.64, P = 0.01%; DP vs BS: R = 0.157, P = 0.06%). Within WS the samples from the deeper stations WS-ET-D and WS-JE-D (with the exception of one core) were separated from the samples of the shallower station WSET-B.

Bubble

plots

(Fig.

3;

data

available

from

https://doi.org/10.1594/PANGAEA.882582, Veit-Köhler et al. 2017) show that nematodes were found in highest numbers in WS (mean values at WS-ET-D and WS-JE-D 5348.7 ind. 10 cm-2, WS-ET-B 6847.6 ind. 10 cm-2), followed by BS (mean 3836.3 ind. 10 cm-2) and DP (mean 3310.5 ind. 10 cm-2). The two BS-C-S cores had 16

the lowest numbers of nematodes (max. 1512.5 ind. 10 cm-2). Copepods and their nauplii (nauplii not shown; Fig. 3) were extremely abundant in the deep Weddell Sea stations (copepods: mean 611.4 ind. 10 cm-2; nauplii mean 704.6 ind. 10 cm-2) but not in the shallower WS-ET-B where copepods were only represented by a mean of 110.4 ind. 10 cm-2. Here, kinorhynchs (mean 172.8 ind. 10 cm-2) and ostracods (mean 108.5 ind. 10 cm-2) had highest abundances with ostracods being as abundant in some of the deeper WS cores. Tardigrades showed a distribution pattern opposed to most of the other taxa: only 2 of the 135 recorded individuals were found in WS. Meiofauna communities from the three regions differed clearly regarding their vertical distribution within the sediment (Fig. 4, note differing scales among regions for CPE and copepods). In the Weddell Sea very high CPE values were observed even in deeper sediment layers. Here, nematodes showed high abundances down to 5 cm sediment depth. Steeper declines in numbers with depth were observed for nematodes in BS and DP. Most of the numerous copepods in the WS were found in the first centimetre, but even down to 2 cm sediment depth more copepods were found in WS than in BS and DP. When total individual numbers per sediment layer [hypothesis test (1b), sum of all taxa per cm as variables] were compared, clear differences among WS and the two other regions were found (one-way ANOSIM, 1-cm steps, untransformed data, factor “region”; Global R = 0.369, P = 0.01%; DP vs WS: R = 0.792, P = 0.01%; BS vs WS: R = 0.378, P = 0.02%; DP vs BS: R = 0.187, P = 0.1%). With a BEST analysis we linked environmental data to the similarity results of the meiofauna communities. The differing community patterns among the three regions were best explained by three sediment variables: Silt&Clay, CPE*, and Chla/Phaeo* (correlation ρ= 0.709, P = 0.1%). 17

3.3 Meiofauna communities of slopes and troughs in BS and DP and their relation to environmental factors [hypothesis (2)] Meiofauna communities differed between the two habitats slope (S) and trough (T, Fig. 5). The graphical finding was supported by the results of a hierarchical PERMANOVA with three factors (region, area, habitat) [Table 4, “habitat” (nested in “area” nested in “region”) P(perm) = 0.0001]. Within each region the factor “area” (W, C, E nested in “region”) did not play a significant role. For the fixed factor “region” Monte Carlo samplings had to be taken into account due to a restricted number of possible permutations. Differences detected between BS and DP did not prove to be significant [P(MC) = 0.1124]. A PERMDISP test allowed for pairwise comparisons of the significant factor “habitat” within the different areas (Table 5). However, with the number of replicates at each station being n = 3 (except for station BS-C-S) PERMDISP results should be interpreted with care (Anderson et al. 2008). Pairwise tests found significant differences between slope and trough meiofauna communities in all three areas of Bransfield Strait (West, Central and East). In Drake Passage, slope and trough communities of the easternmost area were separated (Table 5). DISTLM analyses revealed 7 out of 11 environmental variables that explained best the distribution of the community data cloud (Table 6): Silt&Clay, C/Nmolar, CPE*, Chla/Phaeo*, CmaxChla, CmaxT, and CmaxSal (adj. R2 = 0.43). The routine included CmaxChla in the model although its P value was not significant (P values of variables in marginal tests ≤ 0.05), while the significant predictor variable Coarse Sand* was not included. The visualisation by dbRDA showed that the variation explained by the 18

first two axes together was 76.3% for the fitted model and 41.7% for the total variation (Fig. 6). The 3D projection (not shown) indicates that the communities of the two regions BS and DP were separated by the variables Silt&Clay, C/Nmolar and CmaxT (all higher in DP than in BS). Within DP, the slopes of the Central and East areas were separated from the rest by their higher C/N values. In BS, the trough and slope communities were more scattered. The eastern area BS-E was influenced by higher CmaxChla values than the other areas. CPE* and Chla/Phaeo* played a role in the separation of trough and slope of BS-W and especially BS-C, where environmental variables and communities of the two habitats of showed the greatest differences. The two Bransfield Strait Central Slope (BS-C-S) cores were characterised by the lowest nematode numbers detected during the cruise (1443.5 and 1512.5 ind. 10 cm-2) and lower than average copepod numbers, hence their separation in the dbRDA plot. Taxa that were responsible for the differences in communities between the two regions BS and DP were mainly kinorhynchs and gastrotrichs (higher individual numbers in Drake Passage), as well as tardigrades (more abundant in Bransfield Strait, Fig. 6).

4. Discussion

The three investigated regions are characterised by very distinct environmental settings. Water temperatures at the chlorophyll maximum and the sea bottom are determined by currents and topography: cold Weddell Sea Water (WSW) 19

predominates in the WS (Clarke et al. 2009). Our WS stations (Fig. 1) were situated east of Joinville Island and north-east of James Ross Island at the entrance to the Antarctic Sound (CmaxT -1.48 to -1.81° C, Schröder et al. 2013b). WSW enters BS and affects water temperatures along the north-western side of the Antarctic Peninsula (Huneke et al. 2016; Zhou et al. 2002). In BS, CmaxT increased in southwesterly direction (-1.42 to -0.43° C, Schröder et al. 2013b) from the station north of Joinville Island along the stations of our slope/trough sampling scheme (Fig. 1). DP – under the influence of Circumpolar Deep Water (CDW) driven by the Antarctic Circumpolar Current (ACC; Hofmann et al. 1996) – showed water temperatures ranging from 1.02 to 1.26° C with no clear geographic tendency along the slopetrough scheme (Schröder et al. 2013b). Sea-surface temperature and nutrient availability (besides a stable stratified upper water column) are the main drivers for phytoplankton bloom development. Our three regions are situated within the Chaetoceros spp. diatom province. These diatoms dominate the waters close to the Antarctic Peninsula (Sachs et al. 2009). They are known for a high carbon export and indicate nutrient- and iron-enriched surface waters (Abelmann et al. 2006; Sachs et al. 2009). Temporarily sea-ice covered regions such as the WS are characterised by a highly seasonal primary production and strong pulses of organic matter to the seafloor. Important factors distinguishing WS from the BS/DP system are the organic matter input derived from ice algae and sea-ice meltwater as a water-column stabilising factor that facilitates phytoplankton blooms (Bathmann et al. 1991; Park et al. 1999). Park et al. (1999) observed that the shelf waters in the north-western WS may show a mean primary productivity as high as the marginal ice zone (shelf waters: 2.6 g C m-2 day-1; MIZ: 2.1 g C m-2 day-1). In the Southern Ocean, microalgae aggregate and/or are concentrated in fecal pellets 20

by zooplanktonic grazers such as krill or copepods and sink rapidly to the seafloor [Gooday (1993) reports sinking velocities of 150 m per day; Ragueneau et al. 2000]. Benthic communities are additionally provided with organic matter translocated by currents and deposited far away from the area of production (Dunbar et al. 1989; Yoon et al. 1992; Palanques et al. 2002; Isla et al. 2004).

4.1 Meiofauna communities reflect environmental drivers along the Antarctic Peninsula Primary

production

fuels

the

pelagic-benthic

coupling

but

water-column

measurements can only show a snapshot of the environmental situation while the sediment itself acts as a reservoir of deposited material. Sediment factors determine the composition of meiofauna communities: food availability, sediment grain size or oxygen content play a major role in abundance and spatial distribution of meiofauna taxa (Coull 1970; Soltwedel 2000; Neira et al. 2001; Veit-Köhler et al. 2009; Kuhnert et al. 2010). The ecological drivers behind the significant differences among the meiofauna communities of the Weddell Sea on the one hand and those of Bransfield Strait and Drake Passage on the other were sediment factors mainly resulting from water-column processes (Fig. 2; BEST analysis): the amount of chloroplastic equivalents (CPE), the ratio of chlorophyll a to phaeopigments (Chla/Phaeo, a measure for the freshness of deposited primary production in the sediment), and the silt and clay content (Silt&Clay). For Antarctic meiofauna communities, the important structuring role of food was shown by several shallow-water and slope studies (Table 7). The high abundances of meiofauna in the WS (Hauquier et al. 2015) came along with the highest numbers of 21

copepods ever reported for comparable depths in the Antarctic: 741 and 916 copepods per 10 cm2 (0–5 cm depth) at the deep WS-ET-D and WS-JE-D stations, respectively. To date no comparable densities and percentages of benthic copepods (up to 12.2% of total meiofauna) were reported from similar depths in the vicinity of the Antarctic continent (Table 7 and references therein). The extraordinarily high abundances of copepods in WS were related to the highest amounts of CPE (Fig. 4, Table 7, BEST analysis) observed during our expedition. Copepod numbers were one of the reasons for the community differences among the deep WS stations and the regions north and west of the Antarctic Peninsula (Fig. 3). In BS copepods reached a maximum of 4.6% of the total meiofauna, in DP 4.1% (Table 7 presents cores with highest copepod numbers not percentages). High individual numbers of copepod nauplii resulted from high adult stocks in WS. In general, benthic copepods seem to prefer coarse sediments while nematodes are more abundant at fine-grained sites. Many copepod species make use of the interstitial spaces among the sand grains (Noodt 1957; Remane 1964). Coull (1970) observed even more copepods (967 ind. 10 cm-2) than nematodes in coarse sands of the Bermuda Platform. From the Ross Sea Fabiano and Danovaro (1999) reported that harpacticoid copepods and their nauplii prefer coarse biogenic-calcareous sediments. However, our WS sites were characterised by silty sediments (mean grain size <63 µm) that provide no interstitial habitat. As the copepods were mainly present in the first centimetre (Fig. 4), we assume that they stay in the organic fluff layer at the sediment surface or lead a burrowing life style (Remane 1964). Not completely in line with copepods, ostracods were abundant even in the shallow WS station (Fig. 3). Ostracods are reported to react with increasing population densities to enhanced food supply in intertidal sediments (Kuhnert et al. 2010). Yet 22

lower CPE values were detected at WS-ET-B compared to the deeper WS stations, while the Chla/Phaeo ratio was higher indicating fresher material. Contrary to copepods and ostracods, tardigrades were more abundant at stations with coarser grain sizes. Marine tardigrades are reported from all kinds of habitats but they are more common in sandy habitats (Renaud-Mornant 1988). In our study, only two individuals were found in the silty sediments of WS. In DP and BS they were more abundant in habitats with coarser sediments. The habitat preferences of kinorhynchs are not as straightforward to interpret: on average, their numbers were 16 times higher at WS-ET-B than at the deeper WS stations and in BS, and 7 times higher than in DP. Grain size cannot be a decisive factor for kinorhynchs as BS and DP contained stations with silty sediments, too. Low oxygen levels are a limiting factor for kinorhynch abundance (Neira et al. 2001; Veit-Köhler et al. 2009). However, the load of organic matter at the deeper WS stations was not high enough to cause oxygen depletion: low oxygen values would have severely reduced copepod numbers (VeitKöhler et al. 2009) but copepods were, on the contrary, very abundant at WS-ET-D and WS-JE-D. The intermediate CPE and TOC% content of WS-ET-B may represent an optimum for kinorhynchs. Their low numbers in BS and DP may originate from a lack of food, while competition for space (especially with the highly abundant copepods) may be a reason for the low numbers at the deeper WS stations. Nematodes were more abundant in WS than in BS and DP (Figs 3, 4). Generic composition showed similarities between WS and DP but the relative contribution of genera to the community was different (Hauquier et al. 2015). The authors explained the high nematode abundances in the WS with the higher food availability due to the existence of food banks.

23

4.2 High abundances of copepods in the first centimetre and nematodes in deeper sediment layers in WS – the result of food banks Food banks in the sediments have a structuring influence on the Antarctic benthos. They develop under environmental circumstances that include high (possibly pulsed) productivity, rapid sinking of fresh organic material and low temperatures at the sea floor that decelerate microbial degradation (Isla et al. 2002; Mincks et al. 2005; Smith et al. 2006; Glover et al. 2008). Contrary to the situation in the water column this labile organic material is available to benthic consumers for several months (Mincks et al. 2005) and leads to a less pronounced or even absent seasonality in the benthos (Smith et al. 2006; Mincks and Smith 2007). Our results indicate the presence of food banks in the WS stations. Sediment organic carbon concentrations alone did not reveal large differences between WS and BS (Isla 2016 and this study). However, chlorophyll a and phaeopigment content were clearly highest in WS sediments despite a high standard deviation (Table 3). At the two deeper WS stations we found maximum CPE values in the first centimetre of 33.32 µg g-1 and 23.47 µg g-1, respectively. In deeper sediment layers we detected even higher CPE values up to 78.51 µg g -1 (WS-JE-D, 3–4 cm sediment depth). At first glance, this finding seems to be contradictory to the low Chla content in the WS water column we observed during our expedition [own water-column measurements (WS CmaxChla 0.07 ± 0.01 µg l -1) and cumulative satellite data in Hauquier et al. (2015)]. Overall high CPE and Chla content, even in deeper sediment layers, points towards bioturbation (Mermillod-Blondin and Rosenberg 2006). Infauna (molluscs, polychaetes, crustaceans, echinoderms) had higher standing stocks in WS than in the other regions (H. Link pers. comm.) and many taxa, especially polychaetes, may have influenced the sediment stratification by their burrowing activities (Hauquier et 24

al. 2015). Additionally, increased vertical mixing may be driven by near-bottom currents and sedimentary resuspension processes (Dunbar et al. 1989; Isla et al. 2004). We observed increased turbidity (confirmed by brown coloring of filters) in near-bottom water in all WS stations. High food availability down to 5 cm sediment depth led to deeper vertical occurrence of nematodes in the WS sediments (Fig. 4). Most of the studies on Antarctic meiofauna showed that individual densities decreased more steeply with sediment depth (Herman and Dahms 1992; Vanhove et al. 1998) as did the communities in DP and BS (Fig. 4). Besides food availability, oxygen depletion may be a reason for vertically decreasing densities. On the one hand, certain nematodes are known to resist oxygen depletion caused by high organic load and bacterial activity in deeper sediment layers. On the other hand, bioturbation, resuspension and/or the generally higher oxygenation of the Weddell Sea Deep Water (Gordon et al. 2001) may have led to deeper oxygen penetration in the sediments of WS causing higher nematode abundances in deeper sediment layers. However, relative abundances of nematode genera changed with sediment depth and communities showed a segregation of upper sediment layers (0-1, 1-2 cm) from the deeper layers for the investigated WS and DP stations (Hauquier et al. 2015). Genera typical for fine grained deep-sea habitats and organically enriched sediments were represented by high individual numbers in the WS stations (Hauquier et al. 2015). Quality of organic matter is often reported to be an important structuring factor in macro- and meiofauna communities (e.g., Fabiano and Danovaro 1999; Grémare et al. 2002). While total meiofauna counts were similar over the three WS stations, nematodes reached their highest densities at the shallow WS-ET-B station (~222 m; max. 7957.3 ind. 10 cm-2) and copepod numbers peaked at the two deeper WS 25

stations (>436 m; max. 916 Ind. 10 cm-2; up to 795.7 Ind. 10 cm-2 in the first centimetre; Figs 3, 4). At WS-ET-B we observed the highest Chla/Phaeo values (average 19.1; 0–5 cm) indicating the freshest organic material. However, the total amount of CPE was up to 6 times lower compared to WS-ET-D and WS-JE-D. For our study we conclude that the quantity of food may be among the driving factors for copepod densities, while food quality favoured highest nematode abundances in WS.

4.3 Slope and trough communities in BS and DP – the role of topographic prerequisites and oceanographic drivers The Antarctic shelf is deeper than other shelf areas and characterised by rough coastlines and abrupt reliefs of glacial origin with slopes and troughs that show unique characteristics (Isla et al. 2004). Organic and inorganic sedimentary particles are presorted by currents and fine particles accumulate in deeper areas and depressions such as canyons or troughs (Anderson et al. 1983; Isla et al. 2004). Consequently, organic matter availability is related to sediment grain size and both are of great importance for the biomass and the biological diversity of a system (Gray 1981). Many studies from outside Antarctica report that organic content as well as macro- and meiofauna abundances are often, but not always, higher in canyons compared to adjacent slopes (Vetter and Dayton 1998; Cúrdia et al. 2004; Garcia et al. 2007; Danovaro et al. 2009a; Ingels et al. 2009, 2011; De Leo et al. 2012). Canyons are characterised by occasional sediment gravity flows that transport large amounts of sediment and organic matter to the deep ocean (Canals et al. 2006). However, depessions in the Antarctic upper continental shelf are usually not canyons but troughs of glacial origin that have a characteristic sill at their seawards terminus (Shepard 1981; Figs 7, 8). 26

Troughs in DP west of the South Shetland Islands showed the most pronounced sills (Fig. 8) compared to the troughs at the western flank of the Antarctic Peninsula (Fig. 7). In BS only the easternmost trough had a distinct sill (Fig. 7C) while in the central and western troughs sills were not as pronouced (Fig. 7A, B, see contour lines). One would assume that a sill favours a calm current situation in the landwards oriented depression. Therefore it is astonishing, that we detected very high silt&clay contents and the highest meiofauna abundances in the two “more open” troughs of BS-W-T and BS-C-T. Furthermore, two of the “closed” DP troughs had lower silt&clay contents than the adjacent slopes (Fig. 5). While DP troughs are gradually sloping down on their western side, their eastern side is flanked by a steeply rising plateau (Fig. 8). These abrupt changes in topography may cause turbulences that lead to lower than expected particle accumulation within the troughs (Garcia et al. 2007). Unfortunately, the available data did not allow for the calculation of current velocities for our expedition (Huneke et al. 2016). However, meiofauna densities are not necessarily higher in finer sediments. Garcia et al. (2007) reported the highest meiofauna densities inside Nazaré Canyon at a site with the sandiest sediment. Moreover, compared to BS mostly higher silt&clay contents were observed in DP which may be due to sediment import by glacial runoff from the South Shetland Islands. High near-bed tidal currents and/or high deposition fluxes with frequent resuspension may hinder the establishment of a stable meiofauna community (Thistle 1998; Thistle and Levin 1998; Thistle et al. 1999; Garcia et al. 2007). Macrofauna did not abound either in DP where several sessile and vagile groups were underrepresented or completely absent as compared to BS (see for Ophiuroidea: Ambroso et al. 2016; macroepibenthos: Gutt et al. 2016; Porifera: Kersken et al. 2016; Ascidiacea: Segelken-Voigt et al. 2016).

27

Meiofauna communities in DP showed a higher uniformity than in BS. Contrary to BS where all slope communities differed from those of adjacent troughs, DP communities differed only in one out of three areas (East; Fig. 5, Table 5). This higher uniformity may be, apart from sedimentological prerequisites, a result of further ecological drivers. In DP the highest CmaxChla values measured during our study were contrasting low CPE contents in the sediment. High water temperatures may have led to high turnover rates in this environment leading to low TOC and TN and high C/N in the sediments of DP. Clarke et al. (2009) state that despite the very small absolute range of seabed temperatures in the Southern Ocean, spatial variations in temperature may be an important factor for distribution of species and composition of benthic assemblages. Another possible explanation for the lack of CPE in DP sediments may be lateral advection of primary production (Yoon et al. 1992; Palanques et al. 2002; Isla et al. 2004) or remineralisation of sinking and suspended organic matter. Contrary to DP, the more open western and central troughs of BS may have acted as active seasonal tracks for nutrient and sediment fluxes comparable to canyons (Canals et al. 2006). The colder sea-bottom temperatures in BS may have created small-scale food banks and the resulting high TOC and CPE contents and a low C/N ratio may have positively influenced meiofauna communities. While DP stations were only under influence of a single water mass, the modified Circumpolar Deep Water (Huneke et al. 2016), BS is a more complex system: Firstly, trough systems in BS were more widely incised into the plateau and deeper as in DP (Figs 1, 7, 8) which may have led to mostly weaker local currents compared to adjacent slopes (at least in the W and C areas). Secondly, our sampling stations lay nearly centrally in BS. Here the shelf current on the western side of the Antarctic Peninsula (formed by Transitional Zonal Water with Weddell Sea influence) flows 28

diametrically to the northwards oriented Bransfield Current. Two frontal systems are the result: the superficial Peninsula Front and the deeper Bransfield Front (Huneke et al. 2016). These frontal systems may have enhanced pelagic-benthic coupling and TOC export to the deep by subduction, downwelling, isopycnal exchange or seasonal convective mixing (Siegel et al. 2016). Despite lower primary production in the water column as compared to DP (Hauquier et al. 2015) BS meiofauna may have benefited from such enhanced export characteristics. On a smaller scale the differences we observed between BS and DP showed what on a larger scale distinguished WS from BS and DP: water temperatures (and sea ice) influence the formation of food banks; consequently climate warming may reduce food availability and thus alter community structure and the related benthic ecosystem functions such as respiration, deposit feeding, remineralisation, and recruitment (Clarke et al. 2009; Smith et al. 2012). When sea-bottom temperatures rise, BS may become a rapidly changing system.

5. Conclusions

(1) Meiofauna communities in the Weddell Sea differed significantly from those of Bransfield Strait and Drake Passage. Several water-column related ecological drivers explained the community differences best: the amount of chloroplastic equivalents in the sediment, the freshness of deposited primary production and the silt and clay content. These environmental characteristics of the benthic habitat are dependent on

29

large-scale oceanographic conditions and are thus sensitive to changes in water temperature, salinity, sea-ice cover and the related primary production. (2) Meiofauna abundances in the Weddell Sea were the highest ever reported from comparable depths in the Antarctic. The deep penetration of many nematodes into the sediment may be due to food banks that develop from strong and often pulsed organic fluxes to the seafloor and persist under cold sea-bottom conditions. In the Southern Ocean, pulsed sedimentation events are linked to the existence of sea ice (release of ice algae) and the formation of a stable water-column stratification after sea-ice melting that facilitates phytoplankton blooms. Under changing climate conditions food banks may be reduced or lost and benthic meiofauna communities and abundances may change drastically. (3) The highest copepod abundances and proportions were reported from the deeper Weddell Sea stations. Copepods were related to very high phytopigment contents in the sediments. But contrary to nematodes they were mainly found in the upper sediment centimetre. While copepod numbers were related to the quantity of food, the highest nematode abundances were found at the sites with the best food quality. Different meiofauna groups prefer different habitats and only small changes (quantity or quality of food) may alter a complete community. (4) Bransfield Strait, being a transitional zone between Weddell Sea and Drake Passage, may be most vulnerable to future changes. Only a minor difference in water temperature (and all related consequences) between Weddell Sea and Drake Passage has a great influence on the benthic fauna (compare the abound benthic fauna in Weddell Sea with the impoverished situation in Drake Passage).

30

(5) Bransfield Strait slope and trough meiofauna communities were significantly different. In Drake Passage this was only the case in one out of three areas. The role of less abundant taxa for this separation, namely kinorhynchs, gastrotrichs, and tardigrades, should not be underestimated. The responsible environmental drivers were manifold and their relationships complex (7 out of 11 water-column and sediment-bound factors such as water temperature, primary production, quality of organic matter, pigment amount and quality, grain size). While in Bransfield Strait topography and sediment grain size (plus pigment content due to local food banks) were in line with meiofauna abundances, in Drake Passage this usually observed pattern was overpowered by other environmental influences. Water masses, currents and water temperature play a crucial role in these neighbouring but very different systems.

Acknowledgements We especially thank Annika Hellmann and Marco Bruhn (Senckenberg am Meer, DZMB) for sorting a considerable part of the meiofauna samples. Melanie Busch helped with sorting. Sediment analyses were carried out by the Marine Biology Research Group, Ghent University: Dirk Van Gansbeke is thanked for the pigment analysis, Bart Beuselinck for granulometry and C:N measurements, and Niels Viaene for TOM analysis. We also thank the captain and crew of FS Polarstern and the chief scientist Prof. Dr. Julian Gutt (Alfred-Wegener-Institute) for providing the possibility of sample collection on board. The members of the OFOS Team Prof. Dr. Dieter Piepenburg (University of Kiel) and Alexandra Segelken-Voigt (University of 31

Oldenburg) are thanked for information by video survey on board. Dr. Michael Schröder (Alfred-Wegener-Institute) and his team are thanked for CTD data and water column samplings. FH acknowledges a scholarship from the Research Foundation of Flanders (no.FWO11/ASP/256; http://www.fwo.be/en/). The presented work contributes to project no.BR/132/A1/vERSO of the Belgian Science Policy (BELSPO/BRAIN; www.belspo.be/belspo/index_en.stm). This study was supported by the Alfred-Wegener-Institute, Helmholtz-Centre for Polar and Marine Research (Grant AWI_PS81_03).

References

Abelmann

A,

Gersonde

R,

Cortese

G,

Kuhn

G,

Smetacek

V

(2006)

Extensive phytoplankton blooms in the Atlantic Sector of the glacial Southern Ocean. Paleoceanography 21:PA1013. doi:10.1029/2005PA001199 Ambroso S, Böhmer A, López-González P, Teixidó N (2016) Ophiuroid biodiversity patterns along the Antarctic Peninsula. Polar Biol 39:881–895. doi:10.1007/s00300016-1911-4 Anderson JB, Brake C, Domack EW, Myers N, Singer J (1983) Sedimentary dynamics of the Antarctic continental shelf. In: Oliver RL, James PR, Jago JB (eds) Antarctic Earth Science. Cambridge University Press, Cambridge pp 387–389 Anderson M, Gorley R, Clarke K (2008) PERMANOVA+ for PRIMER: Guide to software and statistical methods. PRIMER-E, Plymouth, UK 32

Anderson M, Robinson J (2003) Generalized discriminant analysis based on distances. Australian & New Zealand Journal of Statistics 45:301–318. Arndt JE, Schenke HW, Jakobsson M, Nitsche FO, Buys G, Goleby B, Rebesco M, Bohoyo F, Hong J, Black J, Greku R, Udintsev G, Barrios F, Reynoso-Peralta W, Taisei M, Wigley R (2013) The International Bathymetric Chart of the Southern Ocean (IBCSO) Version 1.0—A new bathymetric compilation covering circumAntarctic waters. Geophys Res Lett 40:1–7. doi:10.1002/grl.50413 Baguley JG, Montagna PA, Hydea LJ, Kalke RD, Rowe GT (2006) Metazoan meiofauna abundance in relation to environmental variables in the northern Gulf of Mexico deep sea. Deep-Sea Res I 53:1344–1362. doi:10.1016/j.dsr.2006.05.012 Bathmann U, Fischer G, Müller PJ, Gerdes D (1991) Short-term variations in particulate matter sedimentation off Kapp Norvegia, Weddell Sea, Antarctica: relation to water mass advection, ice cover, plankton biomass and feeding activity. Polar Biol 11:185–195 Canals M, Puig P, Durrieu de Madron X, Heussner S, Palanques A, Fabres J (2006) Flushing submarine canyons. Nature 444:354–357. doi:10.1038/nature05271 Clarke A, Griffiths HJ, Barnes DKA, Meredith MP, Grant SM (2009) Spatial variation in seabed temperatures in the Southern Ocean: Implications for benthic ecology and biogeography. J Geophys Res 114:G03003. doi:10.1029/2008JG000886. Clarke KR, Gorley RN (2006) PRIMER v6: User manual/Tutorial. PRIMER-E, Plymouth, UK Cook AJ, Fox AJ, Vaughan DG, Ferrigno JG (2005) Retreating glacier fronts on the Antarctic Peninsula over the past half-century. Science 308:541–544 33

Coull BC (1970) Shallow water meiobenthos of the Bermuda Platform. Oecologica 4:325–357 Cúrdia J, Carvalho S, Ravara A, Gage JD, Rodrigues AM, Quintino V (2004) Deep macrobenthic communities from Nazaré submarine canyon (NW Portugal). Sci Mar 68(Suppl. 1):171–180 Danovaro R, Bianchelli S, Gambi C, Mea M, Zeppilli D (2009a) α-, β-, γ-, δ- and εdiversity of deep-sea nematodes in canyons and open slopes of Northeast Atlantic and Mediterranean margins. Mar Ecol Prog Ser 396:197–209 Danovaro R, Canals M, Gambi C, Heussner S, Lampadariou N, Vanreusel A (2009b) Exploring benthic biodiversity patterns and hotspots on European margin slopes. Oceanography 22(1):16–25 Danovaro R, Company JB, Corinaldesi C, D’Onghia G, Galil B, Gambi C, Gooday AJ, Lampadariou N, Luna GM, Morigi C, Olu K, Polymenakou P, Ramirez-Llodra E, Sabbatini A, Sardà F, Sibuet M, Tselepides A (2010) Deep-sea biodiversity in the Mediterranean Sea: The known, the unknown, and the unknowable. PLoS ONE 5(8): e11832. doi:10.1371/journal.pone.0011832 Decho AW, Fleeger JW (1988) Ontogenetic feeding shifts in the meiobenthic harpacticoid copepod Nitocra lacustris. Mar Biol 97:191–197 De Leo FC, Drazen JC, Vetter EW, Rowden AA, Smith CR (2012) The effects of submarine canyons and the oxygen minimum zone on deep-sea fish assemblages off Hawai'i. Deep-Sea Res I 64:54–70. doi:10.1016/j.dsr.2012.01.014 Dorschel B, Gutt J, Huhn O, Bracher A, Huntemann M, Huneke W, Gebhardt C, Schröder M, Herr H (2016) Environmental information for a marine ecosystem 34

research approach for the northern Antarctic Peninsula (RV Polarstern expedition PS81, ANT-XXIX/3). Polar Biol 39:765–787. doi:10.1007/s00300-015-1861-2 Duineveld G, Lavaleye M, Berghuis E, de Wilde P (2001) Activity and composition of the benthic fauna in the Whittard Canyon and the adjacent continental slope (NE Atlantic). Oceanologica Acta 24:69–83 Dunbar RB, Leventer AR, Stockton WL (1989) Biogenic sedimentation in McMurdo Sound, Antarctica. Mar Geol 85:155–179 Fabiano M, Danovaro R (1999) Meiofauna distribution and mesoscale variability in two sites of the Ross Sea (Antarctica) with contrasting food supply. Polar Biol 22:115–123 Garcia R, Koho KA, De Stigter HC, Epping E, Koning E, Thomsen L (2007) Distribution of meiobenthos in the Nazaré canyon and adjacent slope (western Iberian Margin) in relation to sedimentary composition. Mar Ecol Prog Ser 340:207– 220 Glover AG, Smith CR, Mincks SL, Sumida PYG, Thurber AR (2008) Macrofaunal abundance and composition on the West Antarctic Peninsula continental shelf: evidence for a sediment ‘food bank’ and similarities to deep-sea habitats. Deep-Sea Res II 55:2491–2501 Gooday

AJ

(1993)

Deep-sea

benthic

foraminiferal

species

which

exploit

phytodetritus: characteristic features and controls on distribution. Mar Micropaleontol 22:187–205 Gordon AL, Visbeck M, Huber B (2001) Export of Weddell Sea deep and bottom water. J Geophys Res 106:9005–9017 35

Grange LJ, Smith CR (2013) Megafaunal communities in rapidly warming fjords along the West Antarctic Peninsula: hotspots of abundance and beta diversity. PLoS ONE 8(11): e77917. doi:10.1371/journal.pone.0077917 Gray JS (1981) The ecology of marine sediments. Cambridge University Press, Cambridge, 185p Grémare A, Medernach L, deBovée F, Amouroux JM, Vétion G, Albert P (2002) Relationships between sedimentary organics and benthic meiofauna on the continental shelf and the upper slope of the Gulf of Lions (NW Mediterranean). Mar Ecol Prog Ser 234:85–94 Griffiths HJ (2010) Antarctic marine biodiversity – what do we know about the distribution

of

life

in

the

Southern

Ocean?

PLoS

ONE

5(8):e11683.

doi:10.1371/journal.pone.0011683 Gutt J, Alvaro MC, Barco A, Böhmer A, Bracher A, David B, De Ridder C, Dorschel B, Eléaume M, Janussen D, Kersken D, López-González PJ, Martínez-Baraldés I, Schröder M, Segelken-Voigt A, Teixidó N (2016) Macroepibenthic communities at the tip of the Antarctic Peninsula, an ecological survey at different spatial scales. Polar Biol 39:829–849. doi:10.1007/s00300-015-1797-6 Gutt J, David B, Isla E, Piepenburg D (2016) High environmental variability and steep biological gradients in the waters off the northern Antarctic Peninsula: Polarstern expedition PS81 (ANT-XXIX/3). Polar Biol 39:761–764. doi:10.1007/s00300-0161937-7 Gutzmann E, Martínez Arbizu P, Rose A, Veit-Köhler G (2004) Meiofauna communities along an abyssal depth gradient in the Drake Passage. Deep-Sea Res II 51:1617–1628. doi:10.1016/j.dsr2.2004.06.026 36

Harris PT, Whiteway T (2011) Global distribution of large submarine canyons: geomorphic differences between active and passive continental margins. Mar Geol 285:69–86 Hauquier F, Durán Suja L, Gutt J, Veit-Köhler G, Vanreusel A (2015) Different oceanographic regimes in the vicinity of the Antarctic Peninsula reflected in benthic nematode

communities.

PLoS

ONE

10(9):e0137527.

doi:10.1371/journal.pone.0137527 Hauquier F, Veit-Köhler G (2013) 3.10 Meiofauna and water masses: looking for the link. In: Gutt J (ed.) The Expedition of the Research Vessel "Polarstern" to the Antarctic in 2013 (ANT-XXIX/3). Ber Polar Meeresforsch 665:73–77 Herman RL, Dahms HU (1992) Meiofauna communities along a depth transect off Halley Bay (Weddell Sea – Antarctica). Polar Biol 12:313–320 Higgins RP, Thiel H (1988) Introduction to the study of meiofauna. Smithsonian Institution Press, London, 488p Hofmann EE, Klinck JM, Lascara CM, Smith DA (1996) Water mass distribution and circulation west of the Antarctic Peninsula and including Bransfield Strait. Antarct Res Ser 70:61–80 Huneke WGC, Huhn O, Schröder M (2016) Water masses in the Bransfield Strait and adjacent seas, austral summer 2013. Polar Biol 39:789–798. doi:10.1007/s00300016-1936-8 Ingels J, Billet DSM, Van Gaever S, Vanreusel A (2010) An insight into the feeding ecology of deep-sea canyon nematodes – results from field observations and the first

37

in-situ

13

C feeding experiment in the Nazaré Canyon. J Exp Mar Biol Ecol 396:185–

193. doi:10.1016/j.jembe.2010.10.018 Ingels J, Kiriakoulakis K, Wolff GA, Vanreusel A (2009) Nematode diversity and its relation to quantity and quality of sedimentary organic matter in the Nazaré Canyon, Western

Iberian

Margin.

Deep-Sea

Res

I

56(9):1521–1539.

doi:10.1016/j.dsr.2009.04.010 Ingels J, Tchesunov AV, Vanreusel A (2011) Meiofauna in the Gollum Channels and the Whittard Canyon, Celtic Margin – how local environmental conditions shape nematode

structure

and

function.

PLoS

ONE

6(5):e20094.

doi:10.1371/journal.pone.0020094 Ingels J, Vanreusel A, Brandt A, Catarino AI, David B, De Ridder C, Dubois P, Gooday AJ, Martin P, Pasotti F, Robert H (2012) Possible effects of global environmental changes on Antarctic benthos: a synthesis across five major taxa. Ecol Evol 2(2):453–485. doi:10.1002/ece3.96 Isla E (2016) Organic carbon and biogenic silica in marine sediments in the vicinities of the Antarctic Peninsula: spatial patterns across a climatic gradient. Polar Biol 39:819–828. doi:10.1007/s00300-015-1833-6 Isla E, Masqué P, Palanques A, Guillén J, Puig P, Sanchez-Cabeza JA (2004) Sedimentation of biogenic constituents during the last century in western Bransfield and Gerlache Straits, Antarctica: a relation to currents, primary production, and sea floor relief. Mar Geol 209:265–277 Isla E, Masqué P, Palanques A, Sanchez-Cabeza JA, Bruach JM, Guillén J, Puig P (2002) Sediment accumulation rates and carbon burial in the bottom sediment in a high-productivity area: Gerlache Strait (Antarctica). Deep-Sea Res II 49:3275–3287 38

Kaiser S, Brandão SN, Brix S, Barnes DKA, Bowden DA, Ingels J, Leese F, Schiaparelli S, Arango CP, Badhe R, Bax N, Blazewicz-Paszkowycz M, Brandt A, Brenke N, Catarino AI, David B, De Ridder C, Dubois P, Ellingsen KE, Glover AG, Griffiths HJ, Gutt J, Halanych KM, Havermans C, Held C, Janussen D, Lörz A-N, Pearce DA, Pierrat B, Riehl T, Rose A, Sands CJ, Soler-Membrives A, Schüller M, Strugnell JM, Vanreusel A, Veit-Köhler G, Wilson NG, Yasuhara M (2013) Patterns, processes and vulnerability of the Southern Ocean benthos: a decadal leap in knowledge and understanding. Mar Biol 160:2295–2317. doi:10.1007/s00227-0132232-6 Kennicutt II MC, Chown SL, Cassano JJ, Liggett D, Peck LS, Massom R, Rintoul SR, Storey J, Vaughan DG, Wilson TJ, Allison I, Ayton J, Badhe R, Baeseman J, Barrett PJ, Bell RE, Bertler N, Bo S, Brandt A, Bromwich D, Cary SC, Clark MS, Convey P, Costa ES, Cowan D, DeConto R, Dunbar R, Elfring C, Escutia C, Francis J, Fricker HA, Fukuchi M, Gilbert N, Gutt J, Havermans C, Hik D, Hosie G, Jones C, Kim YD, Le Maho Y, Lee SH, Leppe M, Leitchenkov G, Li X, Lipenkov V, Lochte K, LópezMartínez J, Lüdecke C, Lyons W, Marenssi S, Miller H, Morozova P, Naish T, Nayak S, Ravindra R, Retamales J, Ricci CA, Rogan-Finnemore M, Ropert-Coudert Y, Samah AA, Sanson L, Scambos T, Schloss IR, Shiraishi K, Siegert MJ, Simões JC, Storey B, Sparrow MD, Wall DH, Walsh JC, Wilson G, Winther JG, Xavier JC, Yang H, Sutherland WJ (2014) A roadmap for Antarctic and Southern Ocean science for the

next

two

decades

and

beyond.

Antarct

Sci

27(1):3–18.

doi:10.1017/S0954102014000674 Kersken D, Feldmeyer B, Janussen D (2016) Sponge communities of the Antarctic Peninsula: influence of environmental variables on species composition and richness. Polar Biol 39:851–862. doi:10.1007/s00300-015-1875-9 39

King JC (1994) Recent climate variability in the vicinity of the Antarctic Peninsula, Int J Climatol 14:357–369 King J (2014) A resolution of the Antarctic paradox. Nature 505:491–492 Kuhnert J, Veit-Köhler G, Büntzow M, Volkenborn N (2010) Sediment-mediated effects of lugworms on intertidal meiofauna. J Exp Mar Biol Ecol 387(1-2):36–43. doi:10.1016/j.jembe.2010.03.001 Lins L, Guilini K, Veit-Köhler G, Hauquier F, Alves RMS, Esteves AM, Vanreusel A (2014) The link between meiofauna and surface productivity in the Southern Ocean. Deep-Sea Res II 108:60–68. doi:10.1016/j.dsr2.2014.05.003 McIntyre AD, Warwick RM (1984) Meiofauna techniques. In: Holme NA, McIntyre AD (eds) Methods for the study of marine benthos. 2nd edn. Blackwell, London, pp 217– 244 Mermillod-Blondin F, Rosenberg R (2006) Ecosystem engineering: the impact of bioturbation on biogeochemical processes in marine and freshwater benthic habitats. Aquat Sci 68(4):434–442 Loїc NM, David B, Dubois P, Lepoint G, De Ridder C (2016) Trophic plasticity of Antarctic echinoids under contrasted environmental conditions. Polar Biol 39:913– 923. doi: 10.1007/s00300-015-1873-y Mincks SL, Smith CR (2007) Recruitment patterns in Antarctic Peninsula shelf sediments: evidence of decoupling from seasonal phytodetritus pulses. Polar Biol 30:587–600

40

Mincks SL, Smith CR, DeMaster DJ (2005) Persistence of labile organic matter and microbial biomass in Antarctic shelf sediments: evidence of a sediment ‘food bank’. Mar Ecol Prog Ser 300:3–19 Neira C, Sellanes J, Levin LA, Arntz W (2001) Meiofaunal distributions on the Peru margin: relationship to oxygen and organic matter availability. Deep-Sea Res I 48:2453–2472 Noodt W (1957) Zur Ökologie der Harpacticoidea (Crust. Cop.) des Eulitorals der deutschen Meeresküste und der angrenzenden Brackgewässer. Z Morph Ökol Tiere 46:149–242 Palanques A, Isla E, Masqué P, Puig P, Sánchez-Cabeza J-A, Gili JM, Guillén J (2002) Downward particle fluxes and sediment accumulation rates in the western Bransfield Strait: Implications of lateral transport for carbon cycle studies in Antarctic marginal seas. J Mar Res 60:347–365 Park MG, Yang SR, Kang S-H, Chung KH, Shim JH (1999) Phytoplankton biomass and primary production in the marginal ice zone of the northwestern Weddell Sea during austral summer. Polar Biol 21:251–261 Pasotti F, Convey P, Vanreusel A (2014) Potter Cove, west Antarctic Peninsula, shallow water meiofauna: A seasonal snapshot. Antarct Sci 26(5):554–562 Prieto MJ, Ercilla G, Canals M, de Batist M (1999) Seismic stratigraphy of the Central Bransfield Basin (NW

Antarctic Peninsula): interpretation of deposits and

sedimentary processes in a glacio-marine environment. Mar Geol 157:47–68

41

Probert PK, Batham EJ, Wilson JB (1979) Epibenthic macrofauna off southeastern New Zealand and mid‐ shelf bryozoan dominance. New Zealand Journal of Marine and Freshwater Research 13(3):379–392. doi: 10.1080/00288330.1979.9515814 Puig P, Palanques A, Guillén J, García-Ladona E (2000) Deep slope currents and suspended particle fluxes in and around the Foix submarine canyon (NW Mediterranean). Deep-Sea Res I 47:343–366 Ragueneau O, Tréguer P, Leynaert A, Anderson RF, Brzezinski MA, DeMaster DJ, Dugdale RC, Dymond J, Fischer G, François R, Heinze C, Maier-Reimer E, MartinJézéquel V, Nelson DM, Quéguiner B (2000) A review of the Si cycle in the modern ocean: recent progress and missing gaps in the application of biogenic opal as a paleoproductivity proxy. Global Planet Change 26:317–365 Redfield AC (1958) The biological control of chemical factors in the environment. Am Sci 46:205–122 Remane A (1933) Verteilung und Organisation der benthonischen Mikrofauna der Kieler Bucht. Wiss Meeresuntersuchungen Kiel 21:161–221 Remane

A

(1964)

Die

Bedeutung

der

Struktur

für

die

Besiedlung

von

Meeresbiotopen. Helgoland wiss Meer 10:343–358 Renaud PE, Morata N, Ambrose Jr WG, Bowie JJ, Chiuchiolo A (2007) Carbon cycling by seafloor communities on the eastern Beaufort Sea shelf. J Exp Mar Biol Ecol 349:248–260 Renaud-Mornant J (1988) 33. Tardigrada. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, London, pp 357–364

42

Romano C, Coenjaerts J, Flexas MM, Zúñiga D, Vanreusel A, Company JB, Martin D (2013) Spatial and temporal variability of meiobenthic density in the Blanes submarine

canyon

(NW

Mediterranean).

Prog

Oceanogr

118:144–158.

http://dx.doi.org/10.1016/j.pocean.2013.07.026 Rose A, Ingels J, Raes M, Vanreusel A, Martínez Arbizu P (2015) Long-term iceshelfcovered meiobenthic communities of the Antarctic continental shelf resemble those of the deep sea. Mar Biodiv 45:743–762. doi:10.1007/s12526-014-0284-6 Rowe GT (1971) Observations on bottom currents and epibenthic populations in Hatteras Submarine Canyon. Deep-Sea Res 18:569–581 Sachs O, Sauter EJ, Schlüter M, Rutgers van der Loeff MM, Jerosch K, Holby O, (2009) Benthic organic carbon flux and oxygen penetration reflect different plankton provinces in the Southern Ocean. Deep-Sea Res I 56:1319–1335 Schlacher TA, Schlacher-Hoenlinger MA, Williams A, Althaus F, Hooper JNA, Kloser R (2007) Richness and distribution of sponge megabenthos in continental margin canyons off southeastern Australia. Mar Ecol Prog Ser 340:73–88 Schröder M, Wisotzki A, Nakayama Y, Rucker van Caspel M, Reinlein S, Albrecht T, Mengel M (2013a) 4.1. Observation of dense shelf, deep, and bottom waters downstream of their source regions at the Larsen A, B, and C ice shelves. In: Gutt J (ed) The Expedition of the Research Vessel "Polarstern" to the Antarctic in 2013 (ANT-XXIX/3). Ber Polar Meeresforsch 665:88–100 Schröder M, Wisotzki A, van Caspel M (2013b) Physical oceanography during POLARSTERN cruise ANT-XXIX/3. Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research, Bremerhaven. doi:10.1594/PANGAEA.811907

43

Schröder M, Wisotzki A, van Caspel M (2013c) Physical oceanography measured on water bottle samples during POLARSTERN cruise ANT-XXIX/3. Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research, Bremerhaven. doi:10.1594/PANGAEA.811818 Segelken-Voigt A, Bracher A, Dorschel B, Gutt J, Huneke W, Link H, Piepenburg D (2016) Spatial distribution patterns of ascidians (Ascidiacea: Tunicata) on the continental shelves off the northern Antarctic Peninsula. Polar Biol 39:863–879. doi10.1007/s00300-016-1909-y Shepard FP (1981) Submarine canyons: multiple causes and long-time persistence. AAPG Bull 65(6):1062–1077 Siegel DA, Buesseler KO, Behrenfeld MJ, Benitez-Nelson CR, Boss E, Brzezinski MA, Burd A, Carlson CA, D’Asaro EA, Doney SC, Perry MJ, Stanley RHR, Steinberg DK (2016) Prediction of the export and fate of global ocean net primary production: the EXPORTS Science Plan. Front Mar Sci 3:22. doi:10.3389/fmars.2016.00022 Skowronski RSP de, Corbisier TN (2002) Meiofauna distribution in Martel Inlet, King George Island (Antarctica): sediment features versus food availability. Polar Biol 25: 126–134 Smith CR, DeMaster DJ, Thomas C, Sršen P, Grange L, Evrard V, DeLeo F (2012) Pelagicbenthic coupling, food banks, and climate change on the West Antarctic Peninsula

Shelf.

Oceanography

25(3):188–201.

doi:http://dx.doi.org/10.5670/oceanog.2012.94 Smith CR, Mincks S, DeMaster DJ (2006) A synthesis of bentho-pelagic coupling on the Antarctic shelf: food banks, ecosystem inertia and global climate change. DeepSea Res II 53:875–894 44

Soltwedel T (2000) Metazoan meiobenthos along continental margins: a review. Prog Oceanogr 46:59–84 Soltwedel T, Hasemann C, Quéric N-V, von Juterzenka K (2005) Gradients in activity and biomass of the small benthic biota along a channel system in the deep Western Greenland Sea. Deep Sea Res I 52(5):815–835 Thistle D (1998) Harpacticoid copepod diversity at two physically reworked sites in the deep sea. Deep-Sea Res II 45:13–24 Thistle D, Levin LA (1998) The effect of experimentally increased near-bottom flow on metazoan meiofauna at a deep-sea site, with comparison data on macrofauna. Deep-Sea Res I 45:625–638 Thistle D, Levin LA, Gooday AJ, Pfannkuche O, Lambshead PJD (1999) Physical reworking by near-bottom flow alters the metazoan meiofauna of Fieberling Guyot (northeast Pacific). Deep-Sea Res I 46:2041–2052 Van Gaever S, Galéron J, Sibuet M, Vanreusel A (2009) Deep-sea habitat heterogeneity influence on meiofaunal communities in the Gulf of Guinea. Deep-Sea Res II 56(23):2259–2269. http://dx.doi.org/10.1016/j.dsr2.2009.04.008 Vanhove S, Beghyn M, Van Gansbecke D, Brockington S, Vincx M (1998) The metazoan meiofauna in its biogeochemical environment: the case of an Antarctic coastal sediment. J Mar Biol Ass UK 78:411–434 Vanhove S, Beghyn M, Van Gansbeke D, Bullough LW, Vincx M (2000) A seasonally varying biotope at Signy Island, Antarctic: implications for meiofaunal structure. Mar Ecol Prog Ser 202:13–25

45

Vanhove S, Wittoeck J, Desmet G, Van den Berghe B, Herman RL, Bak RPM, Nieuwland G, Vosjan JH, Boldrin A, Rabitti S, Vincx M (1995) Deep-sea meiofauna communities in Antarctica: structural analysis and relation with the environment. Mar Ecol Prog Ser 127:65–76 Veit-Köhler G (2005) Influence of biotic and abiotic sediment factors on abundance and biomass of harpacticoid copepods in a shallow Antarctic bay. Sci Mar 69 (Suppl. 2)135–145. doi:10.3989/scimar.2005.69s2135 Veit-Köhler G, Durst S, Schuckenbrock J, Hellmann A, Bruhn M, Durán Suja L, Hauquier F (2017) Meiofauna counts during POLARSTERN cruise ANT-XXIX/3 in January-March 2013: Weddell Sea, Bransfield Strait, Drake Passage. PANGAEA. https://doi.org/10.1594/PANGAEA.882582 Veit-Köhler G, Gerdes D, Quiroga E, Hebbeln D, Sellanes J (2009) Metazoan meiofauna within the oxygen-minimum zone off Chile: Results of the 2001-PUCK expedition. Deep-Sea Res II 56(16):1105–1111. doi:10.1016/j.dsr2.2008.09.013 Veit-Köhler G, Guilini K, Peeken I, Sachs O, Sauter EJ, Würzberg L (2011) Antarctic deep-sea meiofauna and bacteria react to the deposition of particulate organic matter after

a

phytoplankton

bloom.

Deep-Sea

Res

II

58:1983–1995.

doi:10.1016/j.dsr2.2011.05.008 Vetter EW, Dayton PK (1998) Macrofaunal communities within and adjacent to a detritus-rich submarine canyon system. Deep-Sea Res II 45:25–54 Walsh JJ (1991) Importance of continental margins in the marine biogeochemical cycling of carbon and nitrogen. Nature 350:53–55

46

Wentworth C (1922) A scale of grade and class terms for clastic sediments. J Geol 30:377–392 Yoon HI, Han MW, Park BK, Han SJ, Oh JK (1992) Distribution, provenance, and dispersal pattern of clay minerals in surface sediments, Bransfield Strait, Antarctica. Geo-Mar Lett 12:223–227 Zhou M, Niiler PP, Hu J-H (2002) Surface currents in the Bransfield and Gerlache Straits, Antarctica. Deep-Sea Res I 49:267–280 Zhou M, Niiler PP, Zhu Y, Dorland RD (2006) The western boundary current in the Bransfield Strait, Antarctica. Deep-Sea Res I 53:1244–1252

Figure captions:

47

Fig. 1 Sampling stations for meiofauna communities and environmental characteristics in the vicinity of the Antarctic Peninsula during RV Polarstern cruise PS 81 (ANTXXIX/3, 01/22–03/18/2013). Explanation of abbreviations for Region-Area-Habitat designations given in Table 1. Background bathymetry based on the International Bathymetric Chart of the Southern Ocean IBCSO (Arndt et al. 2013). Numbers indicate islands mentioned in the text: 1 – Smith Island; 2 – South Shetland Islands; 3 – Brabant Island; 4 – James Ross Island; 5 – Joinville Island. Fig. 2 Principal component analysis (PCA) of the envrionmental factors measured at the stations in Weddell Sea, Bransfield Strait and Drake Passage during PS 81 ANTXXIX/3. Sediment factors: Silt&Clay = % grain size fraction <63 µm; Coarse Sand* = % sand fraction >500 µm; TN = % total nitrogen; C/N = molar carbon:nitrogen ratio; CPE* = sum of Chla and Phaeo in the sediment (4throot transformed); Chla/Phaeo* = ratio of chlorophyll a to phaeopigments in the sediment (square-root transformed). Water column factors: Cmax = measurement at the chlorophyll maximum,

bottom

=

measurement close to the sea bottom; Chla = chlorophyll a in the water column, T = temperature, Sal = salinity. Fig. 3 Non-metric multidimensional scaling (MDS, 2D stress = 0.09) of the Bray-Curtis similarity of square-root transformed meiofauna abundance data of single cores collected in Weddell Sea, Bransfield Strait and Drake Passage (PS 81 ANT-XXIX/3). For better visualisation data of all sediment layers (0–5 cm) were summed per taxon. 48

Cores from Weddell Sea are labeled to show the difference between the shallow (ETB = Erebus and Terror Bank) and the deep stations (ET-D = Erebus and Terror Deep; JE-D = Joinville East Deep). Bubble plots show the individual numbers of selected taxa per 10 cm2 in single cores. Fig. 4 Sediment horizons (0–5 cm sediment depth) for the regions Weddell Sea, Bransfield Strait and Drake Passage (PS 81 ANT-XXIX/3). Concentration of chloroplastic pigment equivalents (CPE) and individual numbers of nematodes and copepods (per 10 cm2) are shown. Box plots indicate median (bar within box), 25 th and 75th percentile (left and right boundary of box), 10th and 90th percentiles (whiskers) as well as outlying maximum and minimum values (dots). Note that scales differ among regions for CPE and copepods. Fig. 5 Meiofauna individual numbers per 10 cm2 (0–5 cm sediment depth, bars show mean number of nematodes, copepod nauplii, copepods, others) in relation to percentage of the sediment fraction silt&clay (grain size <63 µm, line graph with standard deviations) at trough and slope stations in Bransfield Strait and Drake Passage (PS 81 ANT-XXIX/3). Fig. 6 Distance-based redundancy analysis (dbRDA) of environmental variables and slope and trough meiofauna communites of single cores in Bransfield Strait and Drake Passage (PS 81 ANT-XXIX/3). The dbRDA routine performes an ordination of the fitted values from the given DISTLM (Table 6) in the multi-dimensional space. Abbreviations for Region-Area-Habitat designations see Table 1. Sediment factors: 49

Silt&Clay = % grain size fraction <63 µm; C/N = molar carbon:nitrogen ratio; CPE* = sum of Chla and Phaeo in the sediment (4throot transformed); Chla/Phaeo* = ratio of chlorophyll a to phaeopigments in the sediment (square-root transformed). Water column factors: Cmax = measurement at the chlorophyll maximum, Chla = chlorophyll a in the water column, T = temperature, Sal = salinity. Bubble plots show individual densities per 10 cm2 for the presented taxa (maximum number in single core given in each plot). dbRDA1 explained 56.8% of fitted and 31.1% of total variation, dbRDA2 19.5% of fitted and 10.7% of total variation.

Fig. 7 Bottom topography of the investigated systems in Bransfield Strait with sampling stations on slopes and in troughs: A. West, B. Central, C. East. Explanation of abbreviations for Region-Area-Habitat designations given in Table 1. Background bathymetry based on the International Bathymetric Chart of the Southern Ocean IBCSO (Arndt et al. 2013). Fig. 8 Bottom topography of the investigated systems in Drake Passage with sampling stations on slopes and in troughs: A. West, B. Central, C. East. Explanation of abbreviations for Region-Area-Habitat designations given in Table 1. Background bathymetry based on the International Bathymetric Chart of the Southern Ocean IBCSO (Arndt et al. 2013).

50

Fig. 1

Fig. 2 51

Fig. 3

52

Fig. 4

53

Fig. 5

54

55

Fig. 6

56

Fig. 7 A-C

57

Fig. 8 A-C 58

Table 1 Station list and sampling for meiofauna communities (multicorer MUC) and environmental characteristics of sediment (MUC) and water column (CTD Conductivity, Temperature, and Depth profiler equipped with Niskin bottles) during PS 81 ANT-XXIX/3 (01/22 – 03/18/2013; Hauquier and Veit-Köhler 2013). Only successful MUC deployments are listed. For CTD Chlorophyll a-maximum and bottom-water sampling depths are given. Abbreviations for Region-AreaHabitat designations used throughout the text: First position – Region: BS = Bransfield Strait, DP = Drake Passage, WS = Weddell Sea; second position – Area: W = West, C = Central, E = East, ET = Erebus and Terror Gulf, JE = Joinville Island East, JN = Joinville Island North; third position – Habitat: B = Bank, D = Deep, S = Slope, T = Trough.

Weddell Sea - WS

Region

Region-AreaHabitat WS-JE-D Joinville East Deep

WS-ET-B Erebus and Terror Bank

WS-ET-D Erebus and Terror Deep

BS-JN-S

Station no.

Date

Latitude

Longitude

Depth [m]

Gear

120-1

28/01/2013

63°4.62´S

54°33.11´W

20/511

CTD

120-5

28/01/2013

63°4.58´S

54°31.00´W

503.6

MUC

120-6

28/01/2013

63°4.10´S

54°30.86´W

484.8

MUC

120-7

28/01/2013

63°3.72´S

54°30.87´W

436.8

MUC

162-1

10/02/2013

64°0.27´S

56°44.28´W

20/207

CTD

162-3

10/02/2013

64°0.11´S

56°44.28´W

222.1

MUC

162-4

10/02/2013

64°0.07´S

56°44.20´W

223.4

MUC

162-5

10/02/2013

64°0.14´S

56°44.33´W

221.9

MUC

163-1

10/02/2013

63°53.07´S

56°26.19´W

50/453

CTD

163-4

11/02/2013

63°50.95´S

56°24.43´W

517.6

MUC

163-5

11/02/2013

63°51.01´S

56°23.97´W

516.6

MUC

163-6

11/02/2013

63°51.03´S

56°23.68´W

517.1

MUC

118-1

27/01/2013

62°26.47´S

56°17.26´W

19.9/420.1

CTD

59

Bransfield Strait - BS

Joinville North Slope

BS-E-S East Slope

BS-E-T East Trough

BS-C-T Central Trough

BS-C-S Central Slope

BS-W-T West Trough

118-9

27/01/2013

62°26.95´S

56°17.14´W

423.3

MUC

1

118-10

27/01/2013

62°26.90´S

56°17.19´W

427

MUC

118-11

27/01/2013

62°26.89´S

56°17.22´W

427

MUC

193-1

23/02/2013

62°43.01´S

57°34.16´W

20/562

CTD

193-4

23/02/2013

62°43.03´S

57°34.23´W

577

MUC

193-5

23/02/2013

62°43.03´S

57°34.24´W

579

MUC

193-6

23/02/2013

62°43.03´S

57°34.25´W

578

MUC

196-1

24/02/2013

62°48.01´S

57°4.97´W

20/543

CTD

196-5

1

24/02/2013

62°48.03´S

57°4.97´W

567

MUC

196-62

24/02/2013

62°48.04´S

57°5.00´W

574

MUC

196-73

24/02/2013

62°48.00´S

57°4.99´W

559

MUC

202-1

27/02/2013

62°56.00´S

58°0.47´W

50/739

CTD

202-3

27/02/2013

62°56.00´S

58°0.49´W

756

MUC

202-4

27/02/2013

62°56.01´S

58°0.52´W

756

MUC

202-5

27/02/2013

62°55.99´S

58°0.61´W

757

MUC

215-1

1/03/2013

62°53.57´S

58°14.66´W

40/518

CTD

217-22

2/03/2013

62°53.31´S

58°14.17´W

529

MUC

217-3

2/03/2013

62°53.31´S

58°14.12´W

527

MUC

218-1

2/03/2013

62°56.93´S

58°25.66´W

20/672

CTD

218-4

2/03/2013

62°56.95´S

58°25.81´W

689

MUC

218-5

2/03/2013

62°56.95´S

58°25.84´W

689

MUC

218-6

2/03/2013

62°56.93´S

58°25.81´W

689

MUC

60

Drake Passage - DP

BS-W-S West Slope

DP-W-S West Slope

DP-W-T West Trough

DP-C-S Central Slope

DP-C-T Central Trough

DP-E-S East Slope

225-1

4/03/2013

62°56.07´S

58°40.62´W

20/525

CTD

225-3

4/03/2013

62°56.04´S

58°40.73´W

545

MUC

225-4

4/03/2013

62°56.06´S

58°40.76´W

544

MUC

225-5*

4/03/2013

62°56.05´S

58°40.77´W

546

MUC

235-1

7/03/2013

62°16.30´S

61°10.27´W

20/372

CTD

235-4

7/03/2013

62°16.29´S

61°10.24´W

373

MUC

235-5

7/03/2013

62°16.31´S

61°10.24´W

363

MUC

235-6

7/03/2013

62°16.35´S

61°10.25´W

350

MUC

238-2

8/03/2013

62°20.73´S

61°20.15´W

20/454

CTD

238-4

8/03/2013

62°20.82´S

61°20.01´W

460

MUC

238-5

8/03/2013

62°20.78´S

61°20.10´W

464

MUC

238-6

8/03/2013

62°20.80´S

61°20.06´W

466.5

MUC

241-1

9/03/2013

62°6.63´S

60°36.52´W

20/396

CTD

244-5

10/03/2013

62°6.64´S

60°36.53´W

398

MUC

244-6**

10/03/2013

62°6.62´S

60°36.50´W

400

MUC

244-7

10/03/2013

62°6.65´S

60°36.54´W

396

MUC

243-1

10/03/2013

62°12.27´S

60°44.42´W

20/486

CTD

243-3

10/03/2013

62°12.32´S

60°44.47´W

497.8

MUC

243-4***

10/03/2013

62°12.31´S

60°44.48´W

497.7

MUC

243-54

10/03/2013

62°12.31´S

60°44.54´W

495.2

MUC

247-2

11/03/2013

61°56.90´S

60°7.49´W

15/396

CTD

247-4

11/03/2013

61°56.93´S

60°7.48´W

396

MUC

61

DP-E-T East Trough

247-6*

11/03/2013

61°56.93´S

60°7.44´W

396

MUC

247-7

11/03/2013

61°56.91´S

60°7.47´W

400

MUC

250-1

12/03/2013

62°2.28´S

60°12.11´W

20/479

CTD

250-3

12/03/2013

62°2.22´S

60°12.01´W

489

MUC

250-4

12/03/2013

62°2.24´S

60°12.06´W

488

MUC

250-5

12/03/2013

62°2.24´S

60°12.03´W

488

MUC

Community analyses: 1

Two MUC cores from this deployment used for community analyses.

2

Core could only be sliced down to 4 cm sediment depth.

3

Disturbed cores only usable for syringe subsamples.

4

0–1 cm layer of community core contained large stone, data replaced by counts from replicate 0–1 cm layer from the same deployment.

Environmental characteristics: *Grain size, TN% and TOC% data not available, **Chla and Phaeo data not available, ***No sediment data available Table 2 Environmental variables measured in the sediment and the water column during PS 81 ANT-XXIX/3 (01/22 – 03/18/2013): their definition and meaning for the interpretation of the results. Environmental Factor Sediment Silt&Clay%

Definition

Interpretation

grain size fraction <63 µm

*fine sediment, least exposed habitats often characterized by high organic load and low oxygen penetration, burrowing organisms 62

Sand%

grain size fraction >63 and <500 µm

*intermediate grain size

Coarse Sand%

grain size fraction >500 µm

*coarse sediment, indicator for dynamic habitats (e.g., strong near bottom currents), usually low organic content, high porosity and oxygen penetration, interstitial organisms

TN%

total nitrogen

indicator for living components as organic nitrogen compounds are remineralised quickly after death of organisms or release from cells

TOC%

total organic carbon

indicator for productivity, includes deposited organic material and living and dead benthic organisms

C/Nmolar

molar carbon : nitrogen ratio

= (C% /12) / (N% / 14), a ratio of 6.6 is reported for fresh phytoplankton (Redfield 1958), the higher the ratio the more degraded is the organic material

Chla

[µg g-1]

content of chlorophyll a

measure of contribution of primary production to organic matter on the seafloor (freshly sedimented, undegraded microalgae)

Phaeo

[µg g-1]

content of phaeopigments

measure of degrading products of chlorophyll

sum of chlorophyll a and phaeopigments

CPE = “chloroplastic equivalents”, indicator for the total amount of deposited primary production

chlorophyll a : phaeopigment ratio

indicator for the freshness of deposited organic matter derived from primary production, the higher the ratio the fresher the material

content of chlorophyll a at the chlorophyll maximum (Cmax)

Cmax is an oceanographic definition for the water depth with the highest chlorophyll concentration (measured by fluorescence); the Chla content at this depth allows for the comparison of then prevailing standing

CPE

[µg g-1]

Chla/Phaeo

Water Column CmaxChla [µg l-1]

63

stocks of microalgae among regions bottomChla

[µg l-1]

content of chlorophyll a close to/at the seafloor (bottom)

the Chla content at this depth is a measure for sinking products of primary production

CmaxT and [°C] bottomT

water temperature at the chlorophyll maximum (Cmax) or close to/at the seafloor (bottom)

water temperature is among others related to current systems and may differ among regions, water depths and in relation to topographical characteristics of the seafloor

CmaxSal and bottomSal

salinity at the chlorophyll maximum (Cmax) or close to/at the seafloor (bottom)

salinity is among others related to current systems and may differ among regions, water depths and in relation to topographical characteristics of the seafloor

*note: sediments usually are composed by fractions of different grain sizes. The relation of these fractions gives better information about sediment characteristics than the mean grain size.

Table 3 Environmental characteristics of sediment (sampled with MUC; own data) and water column (sampled with CTD Conductivity, Temperature, and Depth profiler equipped with Niskin bottles; Chla: own data; temperature and salinity data: Schröder et al. 2013b,2013c) gathered during PS 81 ANT-XXIX/3 (01/22 – 03/18/2013). Sediment factors were averaged from 0 to 5 cm depth. Mean and standard deviation of environmental factors given per region (Weddell Sea, Bransfield Strait, Drake Passage). Sediment factors: Silt&Clay% = grain size fraction <63 µm; Sand% = grain size fraction >63 and <500 µm; Coarse Sand% = grain size fraction >500 µm; TN% = total nitrogen; TOC% = total organic carbon; C/Nmolar = molar carbon:nitrogen ratio; Chla = content of chlorophyll a, Phaeo = content of phaeopigments; CPE = sum of Chla and Phaeo. Water column factors: Cmax = measurement at the chlorophyll maximum, bottom = measurement close to the sea bottom; Chla = chlorophyll a content in the water column; T = temperature; Sal = salinity.

Sediment Factors (0-5 cm) 64

Area

Weddell Sea Bransfield Strait Drake Passage

Silt&Clay% <63 µm

Sand% <500 µm

Coarse Sand% >500 µm

TN%

TOC%

C/Nmolar

Chla [µg g-1]

Phaeo [µg g-1]

CPE [µg g-1]

Chla/Phaeo

80.26 ±13.31 64.65 ±17.82 88.1 ±8

19.55 ±13.13 29.05 ±12.83 10.28 ±6.55

0.19 ±0.3 6.3 ±5.87 1.62 ±3.43

0.2 ±0.06 0.19 ±0.1 0.09 ±0.01

1.2 ±0.41 0.72 ±0.24 0.61 ±0.06

6.87 ±0.69 5.37 ±1.78 7.84 ±0.31

17.09 ±12.09 0.47 ±0.55 0.07 ±0.05

1.64 ±1.23 0.22 ±0.24 0.08 ±0.11

18.73 ±13.25 0.7 ±0.78 0.15 ±0.16

13.54 ±5.1 2.22 ±1.64 0.77 ±0.32

CmaxChla [µg l-1]

bottomChla -1

bottomT

[µg l ]

CmaxT [°C]

0.07 ±0.01 0.24 ±0.12 0.61 ±0.13

0.03 ±0.01 0.02 ±0.01 0.01 ±0.01

-1.68 ±0.15 -0.93 ±0.32 1.12 ±0.08

-1.82 ±0.04 -0.96 ±0.19 0.58 ±0.28

Water Column Factors

Area Weddell Sea Bransfield Strait Drake Passage

CmaxSal

bottomSal

[°C] 34.35 ±0.03 34.23 ±0.13 34.15 ±0.05

34.48 ±0.02 34.53 ±0.02 34.55 ±0.04

65

Table 4 PERMANOVA of meiofauna communities based on Bray-Curtis similarity of square-root transformed abundance data with the fixed factor “region” (Bransfield Strait and Drake Passage) and the random factors “area” [(West, East, Central) nested in “region”] and “habitat” [(slope and trough) nested in “area” nested in “region”]. Significant term with P(perm) ≤ 0.05 in bold-face type. For the factor “region” the results of Monte Carlo samplings P(MC) had to be taken into account (limited number of unique permutations). Term with non-significant PERMDISP P(perm) > 0.05 marked with asterisk (*). This term allowed for pairwise PERMANOVA tests (Table 5).

1

Source

Degrees of freedom

Sum of squares

Mean square

Pseudo-F

P(perm)

Unique permutations

Region1

1

288.27

288.27

2.0785

0.035

60

0.1124

Area (Region)

4

556.6

139.15

0.52061

0.8545

9850

0.8791

Habitat [Area (Region)]*

6

1607.1

267.86

7.4584

0.0001

9922

0.0001

Residuals

23

826

35,913

Total

34

3160.7

P(MC)

For factor “region” dispersions of the different groups are not homogenous in the multivariate space: significant PERMDISP P(perm) 0.0042

66

Table 5 PERMANOVA pairwise comparisons of meiofauna community similarity of different test groups within significant term “habitat” [(slope and trough) nested in “area” nested in “region”], see also Table 4. PERMDISP test confirmed homogeneity of sample dispersions in the multivariate space. Due to restricted numbers of possible permutations (n = 10) P values were obtained from Monte Carlo samplings P(MC). Significant terms with P(MC) ≤ 0.05 in bold-face type. Term

test level within

test groups

t

P(MC)

Habitat [Area (Region)]

Region: Bransfield Strait Area: West

S vs T

2.1293

0.0334

Area: Central

S vs T

4.991

0.0051

Area: East

S vs T

2.4

0.0291

Area: West

S vs T

1.3196

0.2055

Area: Central

S vs T

1.4392

0.1473

Area: East

S vs T

2.4775

0.021

Region: Drake Passage

67

Table 6 Results of distance based linear models (DISTLM) for Bransfield Strait and Drake Passage. Selection among 11 environmental predictors for the variables that explained best the distribution of the meiofauna community data cloud (Bray-Curtis resemblance matrix). P values of significant predictor variables in bold-face type (P values of variables in marginal tests ≤ 0.05). P values of predictor variables selected by the procedure for the model marked with (a). Sediment factors [asterisk (*) indicates transformed data]: Silt&Clay% = grain size fraction <63 µm; Coarse Sand% = grain size fraction >500 µm; TN% = total nitrogen; C/N molar = molar carbon:nitrogen ratio; CPE = sum of chlorophyll a and phaeopigments; Chla/Phaeo = ratio of chlorophyll a to phaeopigments. Water column factors: Cmax = measurement at the chlorophyll maximum, bottom = measurement close to the sea floor; Chla = chlorophyll a content in the water column, T = temperature, Sal = salinity. Marginal tests Variable

SS(trace)

Pseudo-F

Silt&Clay%

533.46

6.701

0.0008a

0.1688

Coarse Sand%*

494.16

6.116

0.0011

0.1563

TN%

198.28

2.209

0.0819

C/Nmolar CPE* Chla/Phaeo*

245.81 302.77 325.35

2.783 3.496 3.787

P

Prop.

0.0627

0.0382

a

0.0778

0.0155

a

0.0958

0.0195

a

0.1029

a

0.0601

CmaxChla

190.07

2.111

0.0848

bottomChla

151.96

1.667

0.1471

0.0481 a

0.0858 0.0988 0.0592

CmaxT

271.29

3.098

0.0212

CmaxSal bottomSal

312.35 187.07

3.619 2.076

0.0134a 0.0913

68

a

Best solution: adj. R2 = 0.43; RSS 1432.9; 7 variables (Silt&Clay%, C/Nmolar, CPE*, Chla/Phaeo*, CmaxChla, CmaxT, CmaxSal)

Percentage of variation explained by individual axes % explained variation

% explained variation

out of fitted model

out of total variation

Axis

Individual

Cumulative

Individual

Cumulative

1

56.83

56.83

31.07

31.07

2

19.5

76.33

10.66

41.73

3

14.07

90.4

7.69

49.42

Table 7

69

Antarctic shallow-water and slope meiofauna communities (10–1000 m) and their relation to food availability. The maximum individual numbers (total meiofauna and copepods per 10cm2, % copepods) and the corresponding food measure (Chla, CPE, or TOC) are given for each study. Region Signy Island - Factory Cove - Factory Cove King George Island - Martel Inlet - Potter Cove - Potter Cove Bransfield Strait - BS-W-T - BS-C-T Drake Passage - DP-E-S Weddell Sea - WS-JE-D - WS-ET-D Larsen Shelf area formerly ice-covered Halley Bay Ross Sea

max. total meiofauna -2 [10 cm ]

max. copepod no. [10 -2 cm ]

Food availability

Sediment

Water depth [m]

Reference

18800 13200 6731

~1500 (7.9%) 890 (6.7%) ~700 (10.4%)

26 and 49 µg g-1 Chla 89* µg ml-1 Chla (0–5 cm) -2 285 mg m CPE

very fine sand very fine sand coarse silt

10 10 15

20.6* µg g CPE -1 26.4 µg g CPE

silt

12181

216 ~220 (1.8%)

10 15

Vanhove et al. 2000 Vanhove et al. 1998 de Skowronski and Corbisier 2002 Veit-Köhler 2005 Pasotti et al. 2014

5017 6733

172 (3.4%) 140 (2.1%)

0.64 µg g-1 CPE 2.18* µg g-1 CPE

silt silt

689 757

this study

4520

153 (3.4%)

0.1 µg g-1 CPE

silt

396–400

this study

7481 8653 1792

916 (12.2%) 741 (8.6%) 84 (4.7%)

17.2 µg g-1 CPE 33.5* µg g-1 CPE 23.5 µg cm-2 CPE

silt silt silt

503 516 405–427

this study Rose et al. 2014

1505, 3107 2109

120 (8.0%), 41 (1.3%) 28 (1.3%)

0.47*, 0.29% TOC 0.63* µg g-1 CPE

silt silt-clay

492, 339 556

-1

Herman and Dahms 1992 Fabiano and Danovaro 1999

*highest phytopigment content measured at study site (for this study: highest content in region)

70

Highlights -

Meiofauna communities of Weddell Sea differ from Bransfield Strait and Drake Passage

-

Water-column related ecological drivers explained the community differences best

-

Antarctic sediment food banks are especially important for nematodes and copepods

-

Slope and trough communities differ according to topography and/or oceanography

71