Saskia Brix1, Anne-Nina Lorz2, Anna M Jazdzewska3, Lauren Hughes4, Anne Helene S Tandberg5, Krzysztof Pabis3, Bente Stransky2, Traudl Krapp-Schickel6, Jean Claude Sorbe7, Ed Hendrycks8, Wim Vader9, Inmaculada Frutos2, Tammy Horton10, Krzysztof Jazdzewski3, Rachael Peart11, Charles Oliver Coleman12, Lene Buhl-Mortensen13, Charlotte Havermans14,15, Ramiro Tato16, Anali Jimenez Campean17. 1. Senckenberg am Meer, German Centre for Marine Biodiversity Research (DZMB), c/o Biocenter Grindel, CeNak, Zoological Museum Hamburg, Martin-Luther-King-Platz 3, 20146 Hamburg. 2. University of Hamburg, Biocenter Grindel, CeNak, Zoological Museum Hamburg, Martin-Luther-King-Platz 3, 20146 Hamburg. 3. Laboratory of Polar Biology and Oceanobiology, Department of Invertebrate Zoology and Hydrobiology, Faculty of Biology and Environmental Protection, University of Lodz, 12/16 Banacha st., 90-237 Lodz, Poland. 4. Natural History Museum, London, Cromwell Road, South Kensington, United Kingdom. 5. University Museum of Bergen, PO box 7800, NO 5020 Bergen, Norway. 6. Forschungsmuseum A. Koenig, Adenauerallee 160, 53113 Bonn. 7. Station Biologique, 2 rue Jolyet, 33120 Arcachon, France. 8. Canadian Museum of Nature, Ottawa, Canada. 9. Tromso Museum, University of Tromso, 9037 Tromso, Norway. 10. National Oceanography Centre, Southampton, United Kingdom. 11. Coasts and Oceans, National Institute of Water and Atmospheric Research, 301 Evans Bay Pd, Greta Point, Wellington, 6021, New Zealand. 12. Museum fur Naturkunde, Leibniz-Institut fur Evolutions- und Biodiversitatsforschung, Invalidenstra?e 43, 10115 Berlin. 13. Institute of Marine Research, PB 1870 Nordnes, N-5817 Bergen, Norway. 14. Marine Zoology, Bremen Marine Ecology (BreMarE), University of Bremen, PO Box 330440, 28334 Bremen, Germany. 15. Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven. 16. Estacion de Bioloxia Marina da Grana. Universidade de Santiago de Compostela, Rua da Ribeira 1-4. A Grana. CP-15590. Ferrol. Galicia (Espana). 17. Laboratorio de Bentos Marino, Instituto del Mar del Peru, Esquina Gamarra y General Valle, S/N Chucuito, Callao, Peru.
Abstract
Amphipod crustaceans were collected at all 55 stations sampled with an epibenthic sledge during two IceAGE expeditions (Icelandic marine Animals: Genetics and Ecology) in 2011 and 2013. In total, 34 amphipod families and three superfamilies were recorded in the samples. Distribution maps are presented for each taxon along with a summary of the regional taxonomy for the group. Statistical analyses based on presence/absence data revealed a pattern of family distributions that correlated with sampling depth. Clustering according to the geographic location of the stations (northernmost North Atlantic Sea and Arctic Ocean) can also be observed. IceAGE data for the Amphilochidae and Oedicerotidae were analysed on species level; in case of the Amphilochidae they were compared to the findings from a previous Icelandic benthic survey, BIOICE (Benthic Invertebrates of Icelandic waters), which also identified a high abundance of amphipod fauna.
Amphipod crustaceans were collected at all 55 stations sampled with an epibenthic sledge during two IceAGE expeditions (Icelandic marine Animals: Genetics and Ecology) in 2011 and 2013. In total, 34 amphipod families and three superfamilies were recorded in the samples. Distribution maps are presented for each taxon along with a summary of the regional taxonomy for the group. Statistical analyses based on presence/absence data revealed a pattern of family distributions that correlated with sampling depth. Clustering according to the geographic location of the stations (northernmost North Atlantic Sea and Arctic Ocean) can also be observed. IceAGE data for the Amphilochidae and Oedicerotidae were analysed on species level; in case of the Amphilochidae they were compared to the findings from a previous Icelandic benthic survey, BIOICE (Benthic Invertebrates of Icelandic waters), which also identified a high abundance of amphipod fauna.
The international IceAGE project (Icelandic marine Animals: Genetics and Ecology)
focuses on the climatic sensitive region at the northernmost part of the North Atlantic and
the Nordic Seas (Greenland, Iceland and Nordic Seas reaching to the North Sea). The study
area is characterised by a steep temperature gradient (< -0.9 °C to 14 °C) as well as
several shallow (<800 m) submarine ridges which define distinct deep marine basins and
host cold-water coral reefs along their slopes (Buhl-Mortensen et al. 2015a, b). Previous
studies of benthic invertebrates in the North Atlantic and Nordic Seas including the BIOICE
(Benthic Invertebrates of Icelandic waters: 1991-2004) and IceAGE (since 2011: see Brix et al.
2014a) projects have shown that within the abundant peracarid crustacean fauna, more
than 50% of the species are new to science (Blazewicz-Paszkowycz et al. 2014). These projects have identified both broadly and
narrowly distributed species in the region (e.g., Weisshappel 2000, 2001, Dauvin et al. 2012), where geographically restricted
species are distributed to either north or south of Iceland (Svavarsson 1997, Brix and Svavarsson
2010). With cryptic species and species complexes a reoccurring theme for peracarid
crustaceans, and particularly for amphipods (Havermans et
al. 2013), there can be a significant underestimation of regional biodiversity
(Just and Wilson 2004). Previous studies have
indicated that integrative taxonomic approaches better allow for robust and transparent
species delineation (Sites and Marshall 2004, Dayrat 2005, Leache et
al. 2009, Padial et al. 2010). In order to
best capture the diversity, distribution range and dynamic assemblage of the amphipod
crustaceans, the first two IceAGE expeditions in 2011 (M85-3) and 2013 (POS456) expanded upon the
traditional sampling and preservation methods from previous studies to incorporate molecular
approaches.Despite amphipods being the most common peracarid crustacean order within the IceAGE samples, prior to this study, amphipod crustaceans were underrepresented
in project research outputs. The lack of scientific focus on this group was largely due to
the large amounts of time and specialised expertise required to process the volume of
material. At the beginning of the project, more than 66,000 amphipod specimens had been
collected and were available for further identification (DZMB database, unpublished
data).Identifying amphipods is a complex task and owing to the "taxonomic impediment" the number
of amphipod experts worldwide is in decline (Coleman
2015). Taxonomy is the fundamental science for understanding and assessment of
biodiversity. All ecological and modelling analyses rely on accurate taxonomic information.
Considering the known threats to biodiversity, new knowledge of existing species and the
discovery of undescribed species from extant collections are urgently required. To overcome
this impediment, the IceAGE team initiated several identification workshops. These
workshops, run by senior experts, aimed to train early career researchers and to improve
taxonomic knowledge of the amphipods around Iceland.Here the results from two such workshops are presented. These results show the distribution
patterns for amphipod families identified from IceAGE samples. For two abundant families, and
, species level identification is
also presented.
Materials and methods
The IceAGE project and the expeditions were initiated and coordinated by Senckenberg
am Meer (http://www.iceage-project.org), part
of the Senckenberg Forschungsinstitut und Naturmuseum that serves to link scientists to
samples collected by German research vessels and to make this material available. All
sorting was handled according to Riehl et al. (2014)
using an undisturbed cooling chain protocol. Following the fieldwork process and rough
sorting of material to coarse identification levels, material from the IceAGE expeditions were housed in the Senckenberg "Meteor archives" (http://www.material-archiv.de/en/home.html). The IceAGE sampling protocol minimises mechanical nd physiological stress to
specimens during the on-board rough sorting process. This protocol assists in preserving the
integrity of the specimens for both morphological and molecular analyses (Riehl et al. 2014). In terms of expedition protocol, the
sampling included six depth transects (1: Norwegian Channel, 2: Iceland-Faroe Ridge, 3:
Iceland Basin, 4: Irminger Basin, 5: Denmark Strait, 6: Norwegian Sea) between 150 and 2850
m (Figure 1), where samples were collected using
epibenthic sleds.
Figure 1.
Map of all IceAGE EBS stations where amphipods have been found. Red:
stations with amphipods determined; Blue: stations where amphipods were not further
determined (873, 874, 877, 881, 882, 1043, 1045, 1069, 1136, 1144, 1148, 1191, 1209,
1212, 1157).
Map of all IceAGE EBS stations where amphipods have been found. Red:
stations with amphipods determined; Blue: stations where amphipods were not further
determined (873, 874, 877, 881, 882, 1043, 1045, 1069, 1136, 1144, 1148, 1191, 1209,
1212, 1157).During both identification workshops, sample processing concentrated on amphipods collected
with the epibenthic sledge (EBS). It should be noted that three types of gear were used
during the IceAGE expeditions: RP sled (Rothlisberg and Pearcy 1977), Brenke sled (Brenke
2005) and C-EBS (Brandt et al 2013). Within
these samples, identification was concentrated on those samples which were preserved in 96%
ethanol to enable genetic work (see Jazdzewska et al.
2018). Only samples that would provide representatives of the most transects from
the IceAGE station grid (Figure 1, red dots).
As a result of this strategic sorting approach, a total of 21,658 specimens were identified
to family level or lower.For the families and
, all identified specimens have been
registered in the permanent zoological collection at either Senckenberg (Frankfurt), the
Naturkundemuseum Berlin or the Zoological Museum Hamburg (ZMH), Centrum fur
Naturkunde (CeNak). All specimens selected as molecular vouchers (Jazdzewska et al. 2018) will be registered in the ZMH. At the time of
manuscript PageBreakpreparation, the higher classification of the
was in a state of reassessment (Lowry and Myers 2017). Our paper follows a conservative
classification to allow preparation of the material in line with the World
Database as of May 2017 (Horton et al. 2017).Due to the "expert-bias" of participants at our two workshops and the incomplete sorting at
family level, small families often received a more detailed treatment, while some larger
taxa such as the or
were dealt with quite cursorily.
Families that are known to be very abundant in Icelandic waters, including e.g., the
(Dauvin et al. 2012) were underrepresented in our samples as we focused on
ethanol-fixed samples collected by an epibenthic sledge, which does not adequately sample
the . The approach to processing this
extensive amphipod collection did not allow enumerating the species in every family. Despite
these shortcomings, our results can provide a preliminary insight into the
collected during the IceAGE expeditions.Certain findings of singletons or rarer taxa are important for particular families, i.e.,
the (Campean and Coleman 2017) which is the first record of the family in the northern
hemisphere. Here, singletons were excluded from the analyses to reduce "noise". Distribution
maps are provided for the families (or superfamilies) recovered (excluding singletons) with
a brief description indicating its significance in the region.
Data analysis
Distribution maps were created for the amphipod families, one superfamily
() and one infraorder (Corophiida)
occurring at more than two stations using the freeware QGIS, and were assembled using
Photoshop CS6. Multivariate analyses were performed on samples where more than 40% of the
individuals were identified to family level (76-100%: 14 samples, 51-75%: 5 samples,
41-50%: 14 samples). As a result of this processing methodology we readily acknowledge
possible underestimations and restrictions within the dataset. Data were presence/absence
transformed before the analysis. Hierarchical agglomerative clustering was based on
Bray-Curtis similarity formula (Bray and Curtis
1957) using a group average method. SIMPROF test with 1% significance level was
performed in order to confirm multivariate structure within the group (Clarke and Gorley 2015). Multivariate statistical
analysis was performed using the Primer 7 package. Differences in the number of families
per sample between the groups obtained in the Cluster Analysis were tested with use of
Mann-Whitney U test in the STATISTICA 6 package.Additional multivariate analyses (Bray-Curtis formula, group average grouping method,
SIMPROF test with 1% significance level) were also carried out for the two families whose
specimens were identified to species level ( - 32
samples and -25 samples). Here, abundance data
(number of individuals per station) were standardised and square root transformed prior to
analysis (Clarke and Gorley 2015). Additionally, a
set of 39 epibenthic sled (RP sled) samples collected during the BIOICE project was analysed ( to
species level, see Suppl. material 2). Similarity analyses were performed following the same statistical methods
outlined for the IceAGE
.
Results and discussion
Amphipod crustaceans were collected at all 55 stations analysed; however, identification to
the family level was only possible for 40 of them (see Figure 1). The number of individuals per station ranged from a few specimens to more than
16,000 individuals (Figure 2). The Norwegian Sea
stations were characterised by low amphipod abundances with higher numbers of individuals
found at both the shallowest and the deepest stations. Conversely, in the Norwegian Channel,
very high abundances were observed at upper bathyal stations, while the highest numbers of
in the Iceland Basin were observed at
mid-bathyal stations. In other studied areas, no clear pattern associated with depth was
noticed. are known to be an abundant group at all
depths. Generally, amphipod abundance is high in the shelf zone and at upper bathyal depths
(500-1000 m), while they are generally replaced by the at greater
depths (Brandt 1997a, Brandt et al. 2005, 2015). In this study, a
decrease in amphipod abundance at lower bathyal stations is also observed; however, at
shallower stations, the number of individuals seems to depend more on local environmental
conditions than on depth.
Figure 2.
Family distribution at all IceAGE stations ordered by transect. Within each transect (1: Norwegian
Channel, 2: Iceland-Faroe Ridge, 3: Iceland Basin, 4: Irmninger Basin, 5: Denmark
Strait, 6: Norwegian Sea), stations are ordered by depth. The upper graph (black bars)
indicates the absolute number of amphipod individuals per station. The lower graph
indicates the amphipods sorted to families per station (legend by colours shown on the
right side).
Family distribution at all IceAGE stations ordered by transect. Within each transect (1: Norwegian
Channel, 2: Iceland-Faroe Ridge, 3: Iceland Basin, 4: Irmninger Basin, 5: Denmark
Strait, 6: Norwegian Sea), stations are ordered by depth. The upper graph (black bars)
indicates the absolute number of amphipod individuals per station. The lower graph
indicates the amphipods sorted to families per station (legend by colours shown on the
right side).In numerical order, the most abundant taxa were the ,
, , and
followed by the
(Figure 2). All together, these taxa accounted for more than 50% of the
individuals studied. It is worth noting that the dominance of dulichiids resulted from their
very high abundance at just two mid-bathyal stations in the Iceland Basin (stations 1006,
1010). The overall frequency of occurrence of this family was 60% and, at most other
stations, it was represented by a low number of individuals. A large proportion of these
same stations also included representatives of the families
, ,
, and . All of
these taxa are moderately mobile, and are known to be associated with sessile organisms,
PageBreakoften being suspension-feeders (e.g., Lincoln 1979, Caine 1989, Brandt 1993, 1997b). Brix et al. (in press) have reported that at the same stations, high numbers
of the isopod family are recorded and these are also
regarded as having a sedentary lifestyle, associated with other sessile invertebrates. A
study of benthic habitats around Iceland revealed very homogenous sediments in the Iceland
Basin, dominated by sandy muds occasionally accompanied by a small proportion of gravel
(Mei?ner et al. 2014). However, the region is
known to have very productive surface waters and high total organic carbon content in the
sediments was observed, which may explain the high abundances of suspension feeding
peracarids in our study. The other families that dominated the studied material were more
evenly distributed and more frequent (frequency of occurrence often >80%). The numerical
dominance of oedicerotids, phoxocephalids, and lysianassoids in the benthic realm is a
common feature of both shallow and deep-sea ecosystems in all regions of the World (e.g.
Buhl-Jensen 1986, Brandt 1993; Buhl-Mortensen 1996, Weisshappel and Svavarsson 1998, Golovan et al. 2013).
Stebbing, 1906
Figure 3a
Figure 3.
Distribution map for a
b
c
d
e
f
g
h
in sorted IceAGE EBS samples.
In the present study, the family was recorded at two of the 40 stations from ca. 600 m
north of Faroe Islands, with a total of ten specimens. In a revision of the
and related families, Coleman and Barnard (1991) limited the family
to the species of the genus
.
Just (1978) published a taxonomic monograph on
this genus including data on biogeography and biology. The World
Database (Horton et al. 2017) today lists ten species of
, of which
(Ross, 1835) and
(Fabricius, 1780) occur
around Iceland and Just, 1978 and
Just, 1978 are known from
along the east coast of Greenland and Spitsbergen and South Greenland respectively (Just,
1978). seems to be confined to
depths less than 200 m, whereas has been recorded to 700 m
(Just 1978). The colour patterns of species of
can be vivid, ranging from yellow with red stripes
() to bright red or even
purple (). Although Just (1978) provided details on the life history of
acanthonotozomatids, further details of the biology, such as feeding preferences are not
known.Distribution map for a
b
c
d
e
f
g
h
in sorted IceAGE EBS samples.
Pirlot, 1934
Figure 3bIn total, 50 amathillopsid individuals were reported from 12 stations, most from the two
upper bathyal stations just South of Iceland. was found at stations
west of PageBreakPageBreakIceland whereas
is
reported from the eastern stations. Forty-six specimens of
were
recorded, distributed in the Iceland Basin (eight stations) and the Irminger Basin (one
station). The consists of a few little known,
but morphologically spectacular, large amphipods, which lead a pelagic or bentho-pelagic
life. Amathillopsids are found from the Arctic to the Antarctic. In the North Atlantic,
the most commonly reported species is Heller, 1875, with the
lesser cited Miers, 1881 possibly also
present. In the present study, four specimens of this genus were recorded at three deep
stations in the Norwegian Sea. The genus was reassigned to the
in the new subfamily
Cleonardopsinae (Lowry
2006). The species K.H. Barnard, 1916 shows a
cosmopolitan distribution in the deep sea. Described from South Atlantic waters (Cape
Peninsula area; Barnard 1916), it has also been
reported from eastern Greenland (Stephensen 1944),
Bay of Biscay (Elizalde et al. 1993, Dauvin and Sorbe 1995, Frutos and Sorbe 2014a) and eastern Indonesia (off the Moluccas; Pirlot 1934).
Kroyer, 1842
Figure 3cwere reported from 23 of the 40
stations studied, with a total of 492 specimens. The are a
benthic, soft sediment, generally tube-dwelling family. This group has strong grain-size
and depth constraints (Dauvin et al. 2012). The
family is known from intertidal to abyssal depths and often have antenna and pereopod
morphology adapted for different feeding strategies at depth (Barnard 1961). The four genera in the family have some delimitation with
depth with , the most
diverse genus with more than 200 species, being a generalist in both depth and habitat
requirements. This genus most frequently occurs in shallower waters, and numerous species
are recorded from the Atlantic Ocean (Dauvin et al.
2012). A number of papers have recorded and described species from this family
from northern Atlantic waters, particularly Scandinavian, Faroe Island and Icelandic
waters and consequently there is a wealth of information regarding the depth, sediment,
and distributional patterns of the group (Sars
1890-1895, Bellan-Santini and Dauvin 1988,
1997, 2008, Dauvin 1996, Dauvin et al. 2012). Thirteen of the 20 species of
previously documented from
Icelandic waters (Dauvin et al. 2012), were
recorded in the IceAGE samples. The majority of ampeliscid species collected
are in the genus . There were
six species from the samples reported as new to Icelandic waters, with three of these new
to science (Peart 2018). Relatively low abundances
were observed compared to the BIOICE study (Dauvin et al. 2012). The
genera (with more than 90
species), (with more
than 20 species) and (six species) have been
more frequently reported from deeper waters (Dauvin et al.
2012). The distribution reported here (Figure 3c) is supported by published data, which reports a wide distribution of
around Iceland with the majority of
the species occurring between 500 and 1500 m.
Boeck, 1871
Figure 3dAmphilochids were reported from 33 of the 40 stations studied, with a total of 1110
specimens. The family is cosmopolitan and interestingly
includes one species ( Bellan-Santini and Thurston,
1996) from hydrothermal vents at the Mid-Atlantic Ridge (Bellan-Santini and Thurston 1996). The general body shape of amphilochids is
small (1-5 mm mostly) and slightly stout, and observations of living specimens from the
Norwegian and Barents seas support the impression that they are not fast swimmers
(Tandberg, Vader personal observation). Amphilochids occur at all depths and temperatures,
and are quite abundant in the North Atlantic, as in other cold seas. A total of 1110
amphilochid specimens were recorded in the the study, distributed across the genera
,
,
,
,
, and
(for a discussion on
the genus
see Tandberg and Vader (2018)). These specimens
constituted 13 of the 17 species of
previously reported from the eastern North-Atlantic and Arctic (Vader and Tandberg,
personal communication about a manuscript in preparation).
were found at 33 of the 40
stations that have been processed, with three stations (two in the Norwegian Channel and
one north east of Iceland) having more than 100 specimens each. The
were found across all depth and
temperature ranges in the IceAGE station network. Previous studies of amphilochids around Iceland
indicated that the highest abundance and diversity occurs in the north of Iceland (Weisshappel and Svavarsson 1998).
Walker, 1904
Figure 3eArgissids were collected at three of the 40 sampling stations, all located south of
Iceland, at 686-2749 m, with a total of six specimens. The family
comprises the single species
(Norman, 1869), originally
described from shallow water in St. Magnus Bay, Shetland Islands, Scotland. Another
species, Bonnier, 1896, described
from bathyal muddy bottoms of the southern Bay of Biscay, is currently considered a junior
synonym of . However, Lowry and Myers (2017) note that the genus
is in need of
revision since the distribution, depth range and morphological variation attributed to
are implausible when
attributed to a single species. In the southern Bay of Biscay,
was collected with a
suprabenthic sledge on sandy and muddy sand bottoms of the continental shelf (31-179 m),
with a decreasing frequency of occurrence with depth (Sorbe 1984) and also at bathyal depths (711-1098 m) on muddy bottoms (Dauvin and Sorbe 1995, Frutos and Sorbe 2014a, Sorbe and Elizalde
2014). is known to
ocPageBreakcur from the northeastern Atlantic Ocean and Norwegian Arctic (Palerud and Vader 1991). According to Wildish and Peer (1981)
is a deposit feeder. Given
the current monotypic status of the family, this material is assigned to
extending the bathymetric
distribution of this species to 2749 m, and supporting the need for a revision of the
genus .
Lilljeborg, 1865
Figure 3foccurred in samples from the Denmark
Strait and the Faroe Channel at stations associated with strong bottom currents, in six of
the 40 stations studied, with a total of 20 specimens. Palerud and Vader (1991) reported seven species of North-Atlantic
, currently ascribed to the genera
and
(see Bousfield and Kendall 1994). Atylids are sometimes
larger than 10 mm (Hendrycks, personal communication) and show particularly strong lateral
compression along the dorsal pereonites with some species with middorsal carinae.
Characteristic for this taxon is a notch in the dorsal keel of urosomite 1. The cuticle of
atylids in most cases is thin and yellowish or unpigmented. Little is known of the ecology
of atylids, though they appear to occur more commonly in soft bottom shallow-water
habitats.
Sars, 1893
Figure 3gThe family was present at 24 of the 40 stations studied, with a total of 470 specimens
indicating its relative importance in this cold-water area. The relatively speciose family
is represented in the north east
Atlantic by 39 species from 12 genera (Vader and Tandberg, unpublished data) which
accounts for almost 50% of the known in the
world (Horton et al. 2017).
appear to favour colder waters,
although many species are known from more temperate waters, few (if any) are found in warm
waters (Bousfield and Hendrycks 1997). Several
calliopiid species are known to be associates of molluscs, crustaceans (Vader and Tandberg 2013, 2015) and sponges (Vader 1984,
Amsler et al. 2009). Non-associate calliopid
species are found on sandy or muddy seafloors and in macroalgae from littoral to bathyal
depths (d'Udekem d'Acoz 2012). One species,
(Hansen, 1888) is
associated exclusively with sea-ice habitat. Most species are carnivorous or detritivorous
(Bousfield and Hendrycks 1997). Weisshappel (2001) showed a distinct difference in the
species composition of the on
different sides of the Greenland-Iceland-Faroe ridge, with 72% of the species being
restricted to one side of the ridge.
Stebbing, 1899
Figure 3hThe family was reported from 14 of the 40
stations studied, with a total of 190 specimens. Cressids have a compact body, specialised
mandible and an extremely lengthened mandibular palp (Krapp-Schickel 2005). The family is found
mainly in the northern regions of the Atlantic and is comprised of ten species in two
genera, Boeck, 1871 and
Stephensen,
1931 (Horton et al. 2017). Three species have been
identified so far as part of the IceAGE program, the same species as were indicated by Stephensen (1931) namely: Stephensen, 1931,
Boeck, 1871 and
Stephensen, 1931.
Barnard, 1974
Figure 4a
Figure 4.
Distribution map for a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.
was reported from three of the 40
stations studied. Twenty-four cyproideid specimens were recorded in the IceAGE material, mainly on the upper slope in both Irminger and Iceland basins
(13 and ten individuals, respectively). A single specimen was also recorded from a
northern Faroe station (station 879). The are
characterised by immensely broadened coxae 3/4 with contiguous abutting margins and
overlapping coxae 1/2. The
includes 20 genera with 46 species (Horton et al.
2017). Cyproideids are found in association with marine algae, intertidal rocks
or coral debris (Barnard 1972b, Lowry and Stoddart 2003, Azman 2009). They are also known to have associations with live corals
(Myers 1985, Thomas 1999), sponges (Ortiz et al.
2000) and crinoids (Lowry and Azman 2008).
Cyproideids are most diverse in the littoral shallow marine waters of the Indo-West
Pacific (Barnard and Karaman 1991, Lowry and Azman 2008, Ariyama 2016), with just two genera recorded in the north east Atlantic:
and
.
comprises
five species, two of which ( (Scott and Scott, 1893)
and (Stebbing, 1885)) occur
in the Atlantic (Lincoln 1979, Palerud and Vader 1991).
comprises a
single deep-sea species, Sars, 1882 with a boreal
distribution (Sars 1883, 1890-1895, Stephensen 1925, 1938, Buhl-Jensen 1986, Palerud and Vader 1991).Distribution map for a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.
Leach, 1814
Figure 4bDexaminids were found at four of the 40 stations studied in shallow areas east of Faroe
Islands, Iceland Basin and Denmark Strait and at one deep station (2749 m) in the
PageBreakPageBreakIceland Basin, with a total of 14 specimens. Four species
of the family have been reported in the
North-Atlantic (Palerud and Vader 1991).
Dexaminids usually have carinate pleon segments, except for the genus
, and their
urosome segments 2 and 3 are fused. (Montagu, 1813) and
Boeck, 1861 occur in
shallow-water to 60 m (Lincoln 1979).
(Spence Bate, 1862) is
associated with various invertebrates, living in sponges and ascidians (Lincoln 1979), and has also been shown to live in
pouches in the integument of holothurians (Laetz et al.
2013). The Greenlandic (Hansen, 1888),
recently found both in Russia and in Svalbard waters, may well occur in the IceAGE samples. The short bodied (Norman, 1868), a more
southern species, is fossorial and occurs in fine sediments (Kim et al. 2011).Figure 4cwere reported from five of the 40
stations studied, with a total of 55 specimens, in the area of Iceland-Faroe Ridge, at
depths less than 1600 m. This family usually feature prominent teeth carinae (d'Udekem d'Acoz 2010, Krapp-Schickel 2011) and/or robustly elongated coxal plates (Moore 1981, Barnard
and Karaman 1991). Members of the are
bottom-dwelling, epibenthic amphipods represented by two genera in the northern Atlantic,
and
(Palerud and Vader 1991, Horton et al. 2017). Although the species are not abundant, this
readily recognisable family are frequently recorded from shallow to deep waters of
northern Atlantic and Arctic waters as well as around Iceland. Epimeriids contain members
of several feeding types, such as filter feeders and micro-predators (Dauby et al 2001).
Stebbing, 1888
Figure 4dwere reported from 27 of the 40
stations studied with representatives in all sampling areas and depth zones. A total of
775 specimens of have so far been identified from the
IceAGE samples. Eusirids are abundant members of the deep-sea fauna off
Iceland (Weisshappel 2000), known to be predators
(Enequist 1949) with good swimming capabilities
(Bousfield and Hendrycks 1995).
Smith,
1883 was the dominant genus (see Lorz et al.
2018), with three other genera represented; namely Eusirus,
Cleonardo, and . About half of the specimens
collected, 355 individuals, were from a single station at 169 m, which was also the
shallowest sampled (station 866). This easternmost station is at the edge of the North Sea
in contrast PageBreakto other stations containing eusirids, which are in the Arctic
waters of the Norwegian Greenland Seas. This shallow station was dominated by a small
species with large eyes, d'Udekem d'Acoz, Vader
& Legezynska, 2007.
Schellenberg, 1926
Figure 4eLepechinellids were reported from ten of the 40 stations studied, with a total of 103
specimens. The lepechinellids are well adapted to a demersal or epibenthic lifestyle on
soft substrates in deeper waters (Barnard 1973). The
family comprises five genera, three of
which (;
;
) have
been reported in our study area (Thurston 1980,
Palerud and Vader 1991, Johansen and Vader 2015). The genus
was
the most speciose and abundant taxon in the IceAGE samples. More than half of the lepechinellids collected, 60
individuals, were sampled from a single station at 500 m depth at the Iceland-Faroe Ridge
(station 879). Owing to the fragility of the slender spines and thin elongate pereopods
characteristic of the taxon, intact lepechinellids are difficult to obtain; however, the
majority of specimens collected were considered to be in good condition. Morphological
characters are known to vary strongly with growth and gender amongst lepechinellids (Barnard 1973, Thurston
1980). The present samples obtained a range of sizes of both genders of
Schellenberg, 1926, providing
a promising opportunity for further studies on this species.
Dana, 1852
Figure 4fLeucothoids were reported from nine of the 40 stations studied, with a total of 35
specimens. The are well represented within the
Atlantic Ocean, though mainly in warmer regions. The documented Atlantic Ocean leucothoids
have broad shared distributions with eight species also known from South and West Africa,
13 from the Caribbean and Gulf of Mexico, five from Brazil, three species from the Azores,
Biscay, and Mid-Atlantic Ridge near Santa Cruz das Flores, and seven species from United
Kingdom waters to the central and northern Atlantic. A new species of leucothoid is
described based on specimens collected during the IceAGE expedition (Krapp-Schickel
2018). Leucothoids are usually found near, with, or in sponges or tunicates, and
thus specimens are often overlooked inquilines (White
2011). During processing of IceAGE material, only juvenile and male individuals attributable to three
species were identified, namely to the complex,
Boeck, 1861 and
Krapp-Schickel, 2018 (Krapp-Schickel 2018).Figure 4gThree hundred eleven liljeborgiid specimens were collected from 20 of the 40 stations
studied and covering a range of depths and distinct hydrological features. The
are micropredators, with some
being known associates of other invertebrates, including hermit crabs (Vader 1995). Thirteen species of
have been reported from the North
Atlantic (Vader and Tandberg, unpublished data). Liljeborgiids are primarily benthic
species and can occasionally be quite abundant in benthic samples. A single station at 303
m depth in the Norwegian Trench (station 867) was characterised by an extremely high
abundance of (113 individuals).
Dana, 1849
Figure 4hA total of 2008 specimens of lysianassoids and allied taxa was reported from 38 of the 40
stations studied occurring at depths from 169 to 2743 m. This superfamily is an incredibly
large, diverse group of amphipods, which includes scavengers, predators, ectoparasites,
obligate associates, and inquilines (e.g., Lowry and
Stoddart 1983). The recent revision of Lowry and
Myers (2017) has greatly restricted the concept of the
to 130 genera in 12 families,
where formerly the superfamily was composed of 22 families, 173 genera, and 1042 species.
Lysianassoids range from a few millimetres in body length to the largest known amphipod,
the 34 cm plus (Chevreux, 1899) which
occurs in the deep North Atlantic and Pacific.
are distributed globally and are particularly abundant at depth, where they form a
specialist necrophagous guild, feeding on large and small food-falls (Horton and Thurston 2013). Many lysianassoids are
highly mobile, fast swimmers, detecting food-falls from long range through chemoreception
(e.g., Premke et al. 2003). Vader and Tandberg
(unpublished) list almost 200 species of lysianassoid and allied taxa (including the
, ,
, and
) from the eastern North-Atlantic
and Arctic. The great diversity and abundance of lysianassoid taxa identified in the
IceAGE material precluded anything more than a cursorial observation during
the workshops and a sample set of this size is certainly worthy of a more in-depth
study.Figure 5a
Figure 5.
Distribution map for a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.
In total, 254 melphidippid specimens were recorded from 16 of the 40 stations.
Melphidippids have elaborate spination and elongate slender legs and are, at least
partially, PageBreakPageBreakepifaunal. Several studies have indicated that
the normal orientation is upside down in a sling created by the elongate pereopods V to
VII (Enequist 1949). Four species of the family
are known from the Nordic Seas,
Boeck, 1871,
Stebbing, 1899,
Sars, 1894 and
Norman, 1869. The most common
species is , which has a wide depth
distribution, from 50 to 2300 m. is a more northerly species and
is rarer on the Norwegian shelf, yet is frequently found off Iceland at 68 to 688 m. The
closely aligned and
have more southerly
distributions. has not been recorded from
Iceland, is rarely recorded from Norwegian waters but is common in the Skagerrak (Miskov-Nodland et al. 1999).
is known only from Icelandic
waters where water temperatures exceed 3 °C. Distribution of individuals appeared to
indicate aggregations with numerous specimens at some stations, up to 83 individuals.
Similar high-density records are also reported for
d'Udekem d'Acoz, 2006, off
Svalbard.Distribution map for a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.Figure 5bOedicerotids were present and often the most abundant family at 35 of the 40 stations and
occurred alongside other fossorial amphipod families including the
and
. A total of 3569 specimens was
reported; nine genera and 21 species were identified from the material. Among amphipods,
the are a dominant part of the North
Atlantic benthic fauna. Oedicerotids live within the surface sediment of the seafloor and
are deposit feeders (Enequist 1949) or can be
carnivorous (Oliver and Slattery 1985).
Shallow-water species are known to migrate into the water column for reproduction around
lunar cycles and with tidal rhythms (Alldredge and King
1977, Forward 1986).Figure 5cPardaliscids occurred at 26 of the 40 stations studied, with a total of 327 specimens,
suggesting that the family is well represented in the North Atlantic. Pardaliscids contain
approximately 80 species worldwide (Horton et al.
2017). The family have
good swimming ability and are mostly found living in deep-sea habitats (Birstein and Vinogradov 1962, Karaman 1974). A few genera are considered benthic, e.g.,
; however the
majority are thought to range between demersal and epibenthic with the ability to move far
up in the water column. Assessment of gnathopod and mouthpart morphology implies that
pardaliscids are a predatory family (Hendrycks and Conlan
2003). Weisshappel and Svavarsson (1998)
reported pardaliscids from both north and south of Iceland, but the diversity was greater
in the southern sites.
Sars, 1891
Figure 5dwere reported from 39 of the 40
stations studied, with a total of 2134 specimens. The family
contains 375 species (Horton et al. 2017) recorded in all oceans from
tropical to polar zones with very high diversity in Australian waters (Barnard and Karaman 1991). Phoxocephalids are fossorial
and burrow within soft sediments (De Broyer et al.
2003) constituting an abundant part of the infaunal amphipod assemblages from the
shallow sublittoral to the deep sea (Cummings et al.
1998, Lorz and Bamber 2010, Jazdzewska 2015). They are predators (Oliver et al. 1982, Oliver and Slattery 1985). In the North Atlantic, Palerud and Vader (1991) reported 21 phoxocephalid species ascribed to
seven genera. Phoxocephalids are particularly abundant around Iceland (Brandt 1993, Brandt and
Piepenburg 1994, Weisshappel and Svavarsson
1998) and the Norwegian Sea (Buhl-Jensen
1986, Buhl-Mortensen 1996).
Buchholz, 1874
Figure 5ewere reported from 16 of the 40
stations studied, with a total of 594 specimens. The family
currently contains 241 species
worldwide (Horton et al. 2017). Pleustid species
are mostly small to medium-sized (range from 4-20 mm, but most are approx. 10 mm or less),
benthic detritivores and carnivores (Bousfield and
Hendrycks 1994). Many of the pleustid subfamilies have members which are closely
associated with other invertebrates. Globally the distribution of pleustids is mainly
Holarctic, North Atlantic and Arctic with only a small group recorded in southern
hemisphere waters. The diversity of pleustids is most likely related to the abundance of
other benthic invertebrates at the sites, as well as availability of algae for substrate.
This potential inquiline association is supported by a patchy distribution, where two of
the 15 IceAGE stations contained more than 200 individuals.Figure 5fIn total, 1552 stegocephalids were found at 29 of 40 stations studied. Four of these
stations reported more than 100 specimens, including one with 704 individuals (station
870). have been found at all depths
and temperature ranges in the IceAGE material which aligns with the findings of the BIOICE expedition (Berge and Vader
1997). The family is common in the North Atlantic
and PageBreakcontains 26 genera and more than 100 species (Horton et al. 2017). Stegocephalids are quite variable in size, with
the largest species found in the coldest waters. Most species are benthopelagic, while a
few species ( spp.) are
truly pelagic. These latter are caught only irregularly, while some of the benthopelagic
species such as and
may
occur in large numbers over deep soft bottoms (Sars
1895, Vader pers. comm). Stegocephalids feed mainly as micro-predators on large
invertebrates, quite often coelenterates, but some species are also predators, and a few
live in loose associations with other invertebrates (Vader 1984). In the lower latitudes of the North Atlantic, 24 species have been
reported (Vader, unpublished data).Figure 5gA total of 500 stenothoid specimens were recorded from 39 of the 40 stations. Stenothoids
are well known associates of molluscs, sponges, or coelenterates (Krapp-Schickel and Vader 2015). While some probably profit from the
water current created by the host, enabling them to filter-feed or graze their epiphytes,
others are known to feed directly on tissues of the host's body or entire polyps. Thus,
some species are found only by examination of the host, and not by sledge or trawl
sampling methods. The diversity of is
high in the Atlantic Ocean with many species of (more species in shallower
waters), (more species in
deeper regions), and some species found in the study
region. Thirty-five species of and ~ 20 species of
occur in the
Atlantic Ocean and Arctic. Two genera, and
were found
abundantly within the IceAGE collections. A third genus,
, is rarely
present.
Holmes, 1908
Figure 5hA total of 30 stilipedids was sampled from ten of the 40 stations studied. The family
is divided into three subfamilies
and comprises six genera (,
,
,
,
and
) (Horton et al. 2017). The
is a cosmopolitan family, and only
the genera and
occur in the
NE Atlantic. An undescribed bathyal species has been recorded in
temperate waters of the Bay of Biscay (Lagardere
1977, Sorbe and Weber 1995, Frutos and Sorbe 2014a); whereas two
species
( Boeck, 1871 and
Stephensen, 1933) have been
reported in boreal waters of Greenland-Iceland-Faroe and Norwegian seas (Stephensen 1931, 1933, 1940, Palerud and Vader 1991, Brandt
1993, Brandt and Piepenburg 1994, Brandt et al. 1996). The genus
and some
species are
apPageBreakparently pelagic (Berge 2003).
The sampling carried out with sledges in the near-bottom environment (Sorbe and Weber 1995, Frutos and Sorbe 2014b, present study) shows these species could also exhibit a
suprabenthic behaviour. was the most frequently found
species (24 specimens), sampled at five stations in the Irminger and Iceland basins and in
the Norwegian Channel; whereas occurred in deeper water in
the Norwegian Sea. A different (possibly new) species, was recorded in the
Iceland basin at the deepest station (>2700 m) of the expedition (station 967).
Dana, 1853
Figure 6a
Figure 6.
Distribution map for a
b
, in sorted IceAGE EBS samples.
were sampled at 19 of the 40 stations
studied, with a total of 676 specimens. The are a
typical deep-sea family distributed worldwide (Barnard
1972a). Synopiids can be easily recognised by the large head and rostrum shape,
feebly developed gnathopods, and very large telson. Currently, the
comprises 108 species in 18 genera
(Horton et al. 2017). Synopiids in NE Atlantic
waters are represented by the genera: ,
,
,
,
,
,
,
, and
(Sars 1890-1895, Stephensen
1931, 1938, 1944, Buhl-Jensen 1986, Palerud and Vader 1991, Brandt et al. 1996, Brandt
1997b, Bachelet et al. 2003, Frutos and Sorbe 2014a). They occur in all the areas
studied from the Icelandic shelf and slope to the deeper Norwegian, Irminger and Icelandic
basins, the Denmark Strait, and the Faroe Channel. At least 13 species belonging to the
genera ,
,
,
, and
have
been identified so far. Preliminary identifications show that
appears to
be the most speciose genus.Distribution map for a
b
, in sorted IceAGE EBS samples.
Bousfield, 1978
Figure 6bA total of 138 individuals were present at 12 of the 40 stations examined, all situated
south of Iceland (but were not found in the Norwegian Sea). This family comprises
amphipods with small body size, 2 mm to 10 mm, which are highly adapted to a fossorial
lifestyle (Bousfield 1982).
currently comprises 61 species in six
genera (Horton et al. 2017). The family has a
cosmopolitan distribution and can be found from shallow waters to abyssal depths (Sittrop et al. 2015). Almost all North Atlantic
urothoids are shallow water, sandy bottom species; only
(Spence Bate, 1857) and
Jaume & Sorbe, 2001
occur in deep North Atlantic waters (Lincoln 1979,
Jaume and Sorbe 2001). Urothoids are detritus
feeders, and their association with soft bottom habitat supports this. The family is
commonly encountered in North Atlantic samples where this habitat dominates.
Corophiida
Taking into account frequent damage of fragile belonging
to the infraorder Corophiida, individuals with uncertain family assignation, are presented
in Figure 7.
Figure 7.
Distribution map for Corophiida a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.
Distribution map for Corophiida a
b
c
d
e
f
g
h
, in sorted IceAGE EBS samples.Figure 7aAorids were collected at 12 of the 40 stations, located east, south and west of Iceland
(apparently absent at the northern stations), at depths between 168 and 2750 m, with a
total of 105 specimens. The family contains
250 known species world-wide belonging to 25 genera. According to Palerud and Vader (1991) and Myers
(1998), only 15 aorid species belonging to six genera were listed from the North
Atlantic and Norwegian Arctic (including Icelandic waters).
,
, and
are
detritus-feeders and show tube-dwelling habits, whose construction involves secretions
produced by glands located on the third and fourth pairs of pereopods (Enequist 1949).Figure 7boccurred at half of the stations
sampled (18 of 40 stations), between 168 and 2747 m, with a total of 1052 specimens. The
family is large with 91 genera and more
than 400 species. Species are often epibionts, associated with other organisms such as
algae, hydrozoans, bryozoans (Caine 1989), or even
commensals of some marine invertebrates including echinoderms (Guerra-Garcia 2001, Guerra-Garcia et
al. 2008) and decapods (Martin and Pettit
1998). As in most groups, knowledge PageBreakPageBreakof the
biology and distribution is more extensive for shallow-water species; but they are also
known to have a significant presence in deep-sea ecosystems, with numerous records from
Pacific, Atlantic, Arctic, and circum-Antarctic waters (see Guerra-Garcia 2003). In recent years, the number of species reported
from the deep-sea has increased, through the revision of collections from different
museums (Guerra-Garcia 2003, 2004, Guerra-Garcia and Garcia-Gomez
2003) and new oceanographic expeditions (Laubitz
and Sorbe 1996, Guerra-Garcia et al.
2008).Figure 7cDulichiids were moderately abundant at more than half the stations in all areas studied
and were found in very large numbers (thousands of individuals) at two stations in the
Iceland Basin. They were reported from 24 of the 40 stations studied, with a total of 6547
specimens. In our material, are
represented by the genus , and mainly by the species
(Sars, 1879). It is one of the
most abundant groups, dominating at stations in the Iceland Basin. The
currently comprises six genera
(Rauschert 1990, Horton et al. 2017) and a total of 26 known species. The genus
, is
one of the most speciose, with seven described species that are known mainly from the deep
sea (183 to 3229 m) being widely distributed in the North Atlantic, North Pacific and
Arctic Oceans as well as in the Indian Ocean (north Madagascar) (Laubitz 1977, 1979, Ledoyer 1986). The presence of glandular pereopods 3-4
and very long, slender pereopods 5- suggests that they are filter-feeders and
stem-builders. Such behaviour, including self-constructed stems has been described in
various coastal species including (Spence Bate, 1857),
McCloskey, 1970,
(Metzger, 1875),
Spence Bate, 1857 (see
McCloskey 1970, Laubitz 1977, 1979, Moore and Earll 1985, Mattson and Cedhagen 1989, Meyer-Rochow et al.
1991, Thiel 1997, 1998), and Corbari & Sorbe, 2017.Figure 7dIschyrocerids were reported from 20 of the 40 stations studied, with a total of 731
specimens. The family is a diverse group with 269
described species worldwide (Horton et al. 2017).
In the north east Atlantic the family is represented by 28 species (Palerud and Vader 1991, Brandt
1997b). are mostly suspension- and
deposit-feeders and tube-dwellers occurring mainly on the shelf, although
PageBreaksome species occur also at bathyal and abyssal depths (Lincoln 1979, Brandt
1993, 1997b). Their key feature is the
ability to construct tubes with 'amphipod silk'. Accordingly, hemi-sessile species may
occur on soft- and hard-substrata of the northern Atlantic and the Arctic where they can
be locally quite abundant (Sars 1893, Barnard and Karaman
1991, Buhl-Mortensen 1996). In the present
study, the family was recorded from all areas except the Irminger Basin. Findings from the
deeper waters around Iceland, however, were infrequent and restricted to few species
(Raitt 1938, Stephensen 1940).Figure 7eIn this study, 638 podocerids were identified from 11 of the 40 stations studied, with
clusters in the waters south-west and south-east of Iceland. This diversity is almost
certainly underrepresented, as some specimens were most likely identified as corophiids.
The have undergone major changes due to
the work of Myers and Lowry (2003). The family
presently includes eight accepted genera with ~ 100 species and subspecies, the vast
majority belonging to the genus (Horton et al. 2017). Most members of the
inhabit temperate and warm waters
and are bottom-living genera with depressed and cylindrical bodies; however, both
and
are
primarily cold-water amphipods. Podocerids are often found as epifauna on macroalgae and
large invertebrates such as sponges and ascidians. They are poor swimmers, with the main
method of locomotion being crawling and climbing, with the abdomen flexed under the body
(Laubitz 1979).Figure 7fPhotids were collected at 12 of the 40 stations studied located all around Iceland,
excluding the Irminger Basin, at depths between 118 and 2749 m, with a total of 454
specimens. Worldwide, the contain 163 known species belonging to
17 genera (Horton et al. 2017). According to Stephensen (1933, 1940) and Palerud and Vader (1991), 13
photid species belonging to four genera were listed from the north eastern Atlantic and
Norwegian Arctic (including Icelandic waters). With few exceptions,
are known to live in littoral and
sublittoral habitats reaching shallow to bathyal depths. In the North Atlantic,
(Spence Bate, 1862) and
Kroyer, 1842 construct
short tubes of clay or detritus attached to a firm substratum forming dense aggregations
along the seafloor (Enequist 1949).
Myers & Lowry, 2003
Figure 7gA total of 155 specimens of the family was
recovered from ten of the 40 stations studied, in all areas and with a very wide depth
range (118-2749 m). are comprised of 18 genera and are
distributed worldwide in both cold and warm waters. There are two genera in the subfamily
Unciolinae present in Nordic Seas,
and
. One of the
largest and most conspicuous species is (Boeck, 1861). It is common
in the cold and deep waters of the Norwegian Sea from 300 to 2000 m, and is known to
create burrows 10 cm deep and form dense populations on soft deep-sea sediments (Buhl-Mortensen et al. 2016). Another common unciolid,
Norman, 1867, is recorded
from Skagerrak to north of Lofoten (Vader et al.
1997) and is found below 400 m on the outer parts of the Norwegian shelf (Buhl-Jensen 1986). Other
species found in
Nordic Seas include (Spence Bate, 1862) a
southerly species not common in Norwegian waters,
(Kroyer, 1845) and
(Sars, 1876), which have a
northern distribution, found rarely in the Barents Sea (Vader et al. 1997).
Milne Edwards, 1830
Figure 7hHyperiids were reported from 22 of the 40 stations studied, with a total of 134
specimens. The is a diverse planktonic suborder of
amphipods comprising almost 300 species in 76 genera (Horton et al. 2017). During the IceAGE sampling, specimens of the were
mostly found in low numbers (1-5 individuals per station). Specimens occurred in
remarkably high numbers at two stations situated west of the Norwegian shelf break, in the
Norwegian Channel, at around 1000 m. Hyperiids are often parasitic or commensal on
gelatinous zooplankton (Laval 1980). The IceAGE sampling was carried out by means of an epibenthic sledge and the
presence of high numbers of hyperiids caught above the seafloor, indicates their
hyperbenthic feeding habits. These habits may be frequent and Vinogradov (1999b) has reported swarms of Arctic
feeding on
particles on the deep-sea floor. Hyperiids seem to be commonly occurring throughout the
Norwegian Channel and were found in high numbers at around 2700 m depth in the Iceland
Basin. All specimens were recovered from depths greater than 600 m.
Statistical Analysis
Within the benthic deep-sea invertebrate assemblages, amphipods are an abundant and diverse
group. Worldwide around 10,000 species are described, about 80% of PageBreakwhich are
marine (Horton et al. 2017). As the most abundant and
diverse crustacean order in the marine benthos, the determination of IceAGE amphipod specimens provided urgently needed baseline data to understand
the scale of existing collections and for future studies in the North Atlantic and Nordic
seas.
Family level data
Similarity analysis yielded two larger groups of samples, of which one (A) is further
divided into two subclusters (Figure 8). The
subcluster A1 (65% similarity) contains mostly shallower samples, 169 to 913 m (and one
deeper station from 1386 m), from Irminger Basin, Reykjanes Ridge, Iceland Basin and north
of the Faroe Islands. These samples are characterised by a high diversity with all 37
amphipod groups recorded. The constant presence (71-100% frequency) of
, ,
, ,
, ,
Corophiida, , ,
,
,
,
,
, and
, is noted here. The shallower group
is linked with a small cluster of three deep-water samples (2537-2567 m) collected in the
Iceland and Irminger basins having 62% similarity. Compared to the shallower cluster, this
group can be defined by the significant absence of two families, the
and
, which were both a constant
element in the shallower group.
Figure 8.
Dendrogram of samples for the family level data (Bray-Curtis similarity, group
average grouping method and presence/absence transformed data). Abbreviations: RR -
Reykjanes Ridge, NS - Norwegian Sea, DS - Denmark Strait, IB - Iceland Basin, IrmB -
Irminger Basin, WFI - west off Faroe Islands, EFI - east off Faroe Islands, SFI -
South of Faroe Islands, NFI - North of Faroe Islands. (Grey spotted lines indicate the
samples that cannot be significantly differentiated by SIMPROF.) Regions are named
based on the habitats defined by Meißner et al.
(2014).
Dendrogram of samples for the family level data (Bray-Curtis similarity, group
average grouping method and presence/absence transformed data). Abbreviations: RR -
Reykjanes Ridge, NS - Norwegian Sea, DS - Denmark Strait, IB - Iceland Basin, IrmB -
Irminger Basin, WFI - west off Faroe Islands, EFI - east off Faroe Islands, SFI -
South of Faroe Islands, NFI - North of Faroe Islands. (Grey spotted lines indicate the
samples that cannot be significantly differentiated by SIMPROF.) Regions are named
based on the habitats defined by Meißner et al.
(2014).The second major group, cluster B (Figure 8),
contains nine samples at 55% similarity from the Norwegian Sea and east of the Faroe
Islands. It is a group of deep-sea samples, 1058 to 2422 m, again aligning with just one
shallow sample from 600 m. Twenty-six taxa are found in cluster B, which is characterised
by the presence of eight families: ,
, ,
,
,
,
and
. Compared to cluster A1, in
cluster B, the , ,
,
, and
have much lower frequencies
(between 11 and 44%).The spatial distribution of the clusters can be associated with the hydrography of
the region. Symbols of the clusters from figure 8 are plotted on the station map.Differences in family richness are also observed between cluster A1 and B. The mean
number of families per sample is significantly higher in cluster A1 (Z = 3.951, p=0.00007;
21.4 ± 3.6, min = 17, max = 30) compared to cluster B (11.2 ± 1.8, min = 9, max = 14).The spatial distribution of the clusters (Figure 9)
can be associated with the hydrography of the region. The subcluster A1 consists of very
widely distributed samples but their common factor is shallow depths (generally less than
1000 m). A similar pattern was observed for the anthuridean isopod
(Stimpson, 1853) (Negoescu and Svavarsson 1997). This distribution can
be linked to the warm surface current (North Atlantic Current) that comes from the south,
in the Iceland Basin, divides, with one branch flowing around the Faroe islands, and the
second branch encircling Iceland along its south and west coast (Ostmann et al. 2014). The subcluster A2 groups deeper stations from
the Iceland and Irminger basins, separated by the Reykjanes Ridge. The lack of a barrier
effect has already been observed for other peracarids, including
(Svavarsson 1997, Negoescu and Svavarsson
1997, Brix et al. 2014b, Jazdzewska et al.
PageBreakPageBreak2018). In the case of subcluster A2 (see Figure 9), the Iceland Scotland Overflow Water, which is a
deep-water, cold current moving from north east into the Iceland Basin and later flowing
along the Reykjanes Ridge into the Irminger Basin seems to be responsible for shaping the
observed assemblage (Ostmann et al. 2014).
Finally, the third group recognised consists of deep (middle and deep bathyal) samples
from the Norwegian Sea and the east Faroe Islands and may be associated with cold
Norwegian Sea deep water (Ostmann et al. 2014).
The samples of cluster B have lower diversity in comparison to subcluster A1 and similar
differences between Norwegian Sea and northernmost part of North Atlantic Ocean were
observed previously for (Svavarsson 1997).
Figure 9.
The spatial distribution of the clusters can be associated with the hydrography of
the region. Symbols of the clusters from figure 8 are plotted on the station map.
Species level data ( and
)
Figure 10-14
Figure 10.
Dendrogram of samples for the
(Bray-Curtis similarity, group average grouping method standardised and square root
transformed data). RR - Reykjanes Ridge, NS - Norwegian Sea, DS - Denmark Strait, IB -
Iceland Basin, IrmB - Irminger Basin, WFI - west off Faroe Islands, EFI - east off
Faroe Islands, SFI - South of Faroe Islands, NFI - North of Faroe Islands. Grey
spotted lines indicate the samples that cannot be significantly differentiated by
SIMPROF.
Figure 14.
Distribution of selected
species showing different distribution patterns:
(green circle),
(lilac triangle),
(red star).
For the IceAGE
, similarity analysis demonstrated
two larger groups of samples both at relatively low levels of similarity (Figure 10). Amphilochid cluster E, (40% similarity), groups
deep-sea samples from different locations (Norwegian Sea, Irminger Basin, and Iceland
Basin), and is characterised by a low diversity and dominated by
Tandberg & Vader, 2018
(Tandberg and Vader 2018). Amphilochid cluster
PageBreakF (21% similarity) contains mostly shallower samples, 118 to 891 m, in
alignment with three deeper samples from 1058 to 2372 m, from various locations.
Amphilochid cluster F is dominated by Boeck, 1871 and
Spence Bate, 1862. All 13
species of were found in the samples forming
this cluster.Dendrogram of samples for the
(Bray-Curtis similarity, group average grouping method standardised and square root
transformed data). RR - Reykjanes Ridge, NS - Norwegian Sea, DS - Denmark Strait, IB -
Iceland Basin, IrmB - Irminger Basin, WFI - west off Faroe Islands, EFI - east off
Faroe Islands, SFI - South of Faroe Islands, NFI - North of Faroe Islands. Grey
spotted lines indicate the samples that cannot be significantly differentiated by
SIMPROF.In the analysis of from BIOICE (Figure 11), the similarity
analysis yielded two larger clusters at low levels of similarity (cluster H - 30%
similarity and cluster I -15% similarity), both containing samples from a similar depth
range (from 63 to 772 m) and one smaller group of samples, amphilochid cluster G,
collected in a deeper area 1048 to 1407 m and a single sample from 776 m. Amphilochid
cluster G (100% similarity), is characterised by the consistent presence of one species
Stephensen, 1925. Amphilochid
cluster H is characterised by high abundance and frequency of
and
, while amphilochid cluster
I is dominated by Sars, 1892. BIOICE and IceAGE
were analysed in two different
datasets due to different abiotic information between the two projects. Overall, the
BIOICE amphilochid cluster H corresponds with the IceAGE amphilochid cluster F, while BIOICE amphilochid cluster G is consistent with IceAGE
cluster E (Figure 11), where the depth ranges and common species are the
same. Thus, both datasets do show the same pattern.
Figure 11.
BIOICE Dendrogram of samples for the
collected during BIOICE project (Bray-Curtis similarity, group average
grouping method standardised and square root transformed data). Grey spotted lines
indicate the samples that cannot be significantly differentiated by SIMPROF.
BIOICE Dendrogram of samples for the
collected during BIOICE project (Bray-Curtis similarity, group average
grouping method standardised and square root transformed data). Grey spotted lines
indicate the samples that cannot be significantly differentiated by SIMPROF.Although was recorded as a
deep and cold-water associated genus in the IceAGE samples, it is known to occur commonly at shallow depths in the North
Sea (Beermann and Franke 2011) and the
Mediterranean Sea (Krapp-Schickel 1982). However,
an affinity with deep and cold water is recognised for the very widely distributed new
species (Tandberg and Vader 2018) and
.
seems to be limited to the
upper 1000 m, with conspicuous abundance at all stations from both BIOICE and IceAGE samples from south-west of the Reykjanes Peninsula (Figure 13). The most abundant amphilochid species,
, was sampled mainly from
shallow waters to the shelf edge (139-905 m, with one record from 1384 m). It is worth
highlighting, however, the distinct possibility that
might be a cryptic species
complex (Jazdzewska et al. 2018, Tandberg and Vader 2018).
Figure 13.
found at BIOICE and IceAGE stations plotted together (grey circles) with the distribution of
(red triangle) and
(green circle) at
these stations.
More detailed analysis of the family
yields two weakly marked clusters at low levels of similarity (cluster C -19%, cluster D -
4%; Figure 12). Overall, distribution patterns
observed for the reflected those seen for the
higher family-level analysis. In oedicerotid cluster C there is a subcluster of deep-sea
samples collected in the Norwegian Sea (C1 - at 62% similarity). This oedicerotid cluster
C1 largely corresponds with the family level cluster B. Samples from oedicerotid
subcluster C1 are dominated by (Hansen, 1888),
(Boeck, 1871),
(Sars M, 1858) and
(Goes, 1866). Shallower
samples are grouped in the oedicerotid cluster D but are also spread across various other
subclusters. Oedicerotid cluster D samples are consistently from the continental shelf
depths, 118 to 587m, but notably from different regions (Iceland Basin, Reykjanes Ridge,
Denmark Strait, north of Faroe Islands). Oedicerotid cluster D is dominated by
(Grube, 1864) and
Sars,
PageBreakPageBreakPageBreakPageBreak1892. This cluster is also
typified by the absence of four species which were dominant in oedicerotid cluster C,
namely , and
Eleven of the 21 species
of have abundances between one and
five individuals in the samples.
Figure 12.
Dendrogram of samples for the
(Bray-Curtis similarity, group average grouping method standardised and square root
transformed data). Abbreviations: RR - Reykjanes Ridge, NS - Norwegian Sea, DS -
Denmark Strait, IB - Iceland Basin, IrmB - Irminger Basin, WFI - west off Faroe
Islands, EFI - east off Faroe Islands, SFI - South of Faroe Islands, NFI - North of
Faroe Islands. Grey spotted lines indicate the samples that cannot be significantly
differentiated by SIMPROF.
Dendrogram of samples for the
(Bray-Curtis similarity, group average grouping method standardised and square root
transformed data). Abbreviations: RR - Reykjanes Ridge, NS - Norwegian Sea, DS -
Denmark Strait, IB - Iceland Basin, IrmB - Irminger Basin, WFI - west off Faroe
Islands, EFI - east off Faroe Islands, SFI - South of Faroe Islands, NFI - North of
Faroe Islands. Grey spotted lines indicate the samples that cannot be significantly
differentiated by SIMPROF.found at BIOICE and IceAGE stations plotted together (grey circles) with the distribution of
(red triangle) and
(green circle) at
these stations.In assessing the species distribution patterns
within the IceAGE material (for examples see Figure 14), a more diverse species assemblage is apparent in the northern
sampling localities. The two species of ,
(Spence Bate, 1857) and
(Sars, 1883) were present in
Norwegian Channel only, as were both species of , although
Sars, 1892 was
represented by only eleven specimens across two sites.
(Norman, 1889)
() was present at a single
Norwegian Channel site with 17 specimens. Boeck, 1871 was recorded at
three sites only in the Norwegian Channel, again with one sample represented by a single
specimen. A total of 72 specimens were recorded across two regions, the Norwegian Channel
and Denmark Strait, for (Boeck, 1871). A
similar split between the Norwegian Channel and Denmark Strait was seen for 217
individuals of (Hansen, 1888), yet
one of the two Norwegian Channel sites had the majority of individuals, with 110
specimens, while the second site had just three specimens. In the Denmark Strait sites,
specimens were more
evenly spread across sites. (Goes, 1866) and
(Sars M., 1858)
PageBreakwere present across the Norwegian Channel and Iceland-Faroe Ridge, as well
as the Denmark Strait to Norwegian Sea (Figure 14).
also showed high numbers of
individuals (42 specimens) at a single Norwegian Channel station, considering a total of
74 specimens reported across all stations.Distribution of selected
species showing different distribution patterns:
(green circle),
(lilac triangle),
(red star).genera and species have the
highest diversity in the North Atlantic. As the family has not received intensive study in
other regions, with the exception of the north east Pacific, it is difficult to know if
this is a biodiverse region for or an
artefact of taxonomic treatment. Generic level review would be meaningful to address the
taxonomic errors in the literature and to better understand relationships where few
characters separate groups (Barnard and Karaman
1991). In most genera a full complement of subadult, male, and female specimens
have not been assessed, and this is problematic for defining generic level characters,
i.e., (Just 1980). The distinction between intra- and
interspecific variation is not well studied. Some authors accept a 0.1 variation in ratios
as an acceptable species-level character even when less than five individuals were
available for study and large sample sizes show high degree of overlap and standard error
(Jansen 2002). The position of the eye(s) and
shape of the rostrum is cited as having high intraspecific variability, e.g., the widely
reported , while rostrum shape and size
is also used to separate species of (Jansen 2002). The family would benefit from a holistic generic and
species level review to more adequately represent evolutionary relationships.
General findings
The shelf-edge, especially in the Norwegian Channel, is particularly diverse. One
possibility for this might be that the two northbound water masses (deep and cold, shallow
and warm) mix in this zone, making possible habitats and more abundant food sources for a
larger and more diverse set of species. The diversity maximum for gastropod molluscs was
found to be between 400 and 450 m (Høisæter 2010).
Høisæter's data (2010) for gastropods showed a very similar pattern reported from the
Faroe Channel, but with a maximum diversity 200 m deeper along the Norwegian slope close
to the Norwegian Channel (North Sea Fan). This zone coincides with a high fluctuation in
temperature, where both, positive and negative values, were observed, indicating a varying
depth of thermocline. This diversity peak was previously interpreted as an overlap of the
upper and lower assemblages (Gage 2004). The
factors structuring bathymetric patterns in different ocean basins and slopes may differ
from those affecting other taxa in the same area.While substantial parts of Arctic waters north of Iceland, as well as the North Atlantic
south of Iceland, belong to the deep sea, reaching below 3000 m on the abyssal plains, the
BIOICE dataset includes only a few stations at these depths. The dataset used
here contains several samples below 1000 m. The Icelandic shallow water fauna is well
documented, particularly for crustaceans (Sars 1890-1895, Svavarsson 1997, Svavarsson et al. 1993)
though information declines with depth. In the case of peracarid crustaceans, in the
Arctic Ocean the numbers of individuals of each species is high, while overall diversity
is low. Conversely, for the North Atlantic , diversity
is high, while the number of individuals per species is comparably low (Svavarsson et al. 1993). Svavarsson et al. (1993) demonstrated considerable differences in
recent isopod faunal characteristics between the shallow and deep waters North of Iceland.
Here, the number of species declines by 50% at 1000 m and to one third of species at
depths greater than 2000 m. The degree of Arctic endemism is seen to increase with
depth.Other groups show similar depth patterns. In the case of deep-sea prosobranchs from the
North Atlantic (Porcupine Seabight and Abyssal Plain) Olabarria (2006) found depth to be a significant predictor of diversity with
rates of species succession increasing rapidly with increasing depth, indicating four
possible depths of faunal turnover: 700; 1600; 2800 and 4100 m. The study of Olabarria (2006) was based on data from 71 epibenthic
sledge samples between 150 and 4915 m. The turnover depths differed from other taxa in the
Porcupine Basin and also from other areas which would indicate a lack of global
consistency in such depth-related diversity patterns. The decrease in diversity observed
by Olabarria (2006) correlates with the permanent
thermocline from about 600 to 1400 m. Rex (1981)
and Etter and Rex (1990) found diversity maxima at
depths between 2000 to 3000 m for polychaetes, gastropods, protobranchs, and cumaceans.
Paterson and Lambshead (1995) found diversity
maxima at around 1800 m for polychaetes at the Hebridian Slope. Flach and Bruin (1999) observed increasing diversity with increasing
depth in molluscs. On the Scottish Slope macrozoobenthos diversity is low at 400 m and
highest at around 1400 m (Bett 2001).
PageBreakThe unimodal relationship between diversity and depth with a peak at
intermediate depths (2000-3000 m) is not universal and particular abiotic processes can
modify the trend (Ramirez-Llodra et al 2010). A
recent global biogeography "Global Open Ocean and Deep Seabed" (GOODS) describes 37
benthic provinces divided into four depth ranges (Ramirez-Llodra et al. 2010).
Conclusions
The sorting effort of two workshops on IceAGE expedition material has enabled the identification of more than 20,000
amphipod specimens to the family level from Icelandic and adjacent waters. Several
families were identified further to species level. Distribution maps of occurrences have
been provided in a preliminary investigation of regional amphipod family distributions.
Statistical analyses at the family level revealed a depth related pattern, which was
supported by species level data for two abundant families in the samples, the
and
. In all three datasets (family
level, and
) diversity was highest at slope
depths where due to upwelling effects, cold water mixes with warmer water and
phytoplankton/zooplankton are more abundant, supporting previous hypotheses that
thermoclines play an important role in shaping species diversity and distribution patterns
in the Icelandic benthic ecosystem (Høisæter
2010).
Authors: Adam D Leaché; Michelle S Koo; Carol L Spencer; Theodore J Papenfuss; Robert N Fisher; Jimmy A McGuire Journal: Proc Natl Acad Sci U S A Date: 2009-07-22 Impact factor: 11.205
Authors: Charlotte Havermans; Gontran Sonet; Cédric d'Udekem d'Acoz; Zoltán T Nagy; Patrick Martin; Saskia Brix; Torben Riehl; Shobhit Agrawal; Christoph Held Journal: PLoS One Date: 2013-09-25 Impact factor: 3.240
Authors: Hanieh Saeedi; James Davis Reimer; Miriam I Brandt; Philippe-Olivier Dumais; Anna Maria Jażdżewska; Nicholas W Jeffery; Peter M Thielen; Mark John Costello Journal: PeerJ Date: 2019-10-29 Impact factor: 2.984
Authors: Saskia Brix; Stefanie Kaiser; Anne-Nina Lörz; Morgane Le Saout; Mia Schumacher; Frederic Bonk; Hronn Egilsdottir; Steinunn Hilma Olafsdottir; Anne Helene S Tandberg; James Taylor; Simon Tewes; Joana R Xavier; Katrin Linse Journal: PeerJ Date: 2022-06-15 Impact factor: 3.061