MACROZOOBENTHOS OF THE SHALLOW WATERS OF PECHORA BAY (SE BARENTS SEA)

Macrozoobenthic communities provide vital ecosystem services including habitats and foraging resources for other species in all marine ecosystems. Although macrozoobenthos of deeper parts of the �echora �ea ��� �ar� ents �ea) have been studied in more detail, there is a lack of research in shallow waters of the �echora �ay. The study area lies within the Nenetsky �tate Nature Reserve, established in 1997, to protect important breeding and moulting grounds of waterfowl. Macrozoobenthos provide key foraging resources for waterfowl in the nature reserve, however, there is a mismatch between ornithological and macrobenthic data. �ight stations were studied along the Russky Zavorot �eninsula in the �echora �ay on a depth of 1.1–1.8 m within the near�shore zone of the Nenetsky �tate Nature Reserve in August 2016. A monodominant community of Limecola balthica with a biomass of 21.31 ± 0.32 g/m2 and 14 species in total was recorded across the area. The dominant species of the community correspond to those in the community of L. balthica recently described from the central estuarine part of the �echora estuary. A low biomass and poor species richness in the L. balthica community support the earlier published results for the northern part of the bay and indicate the dependence of the community charac� teristics on environmental factors. The paucity of macrozoobenthos in the area is likely attributed to extreme environmental conditions including the following: �1) the water column freezes to the bottom during winter in the shallows of the Pechora estuary or (2) the freshwater flow spreads under the ice, severely impacting salinity. Hence the community is comprised of eurythermal and euryhaline forms and is reduced in biomass. It is unlikely that the shallows of the Russky Zavorot �eninsula play an important role as feeding grounds for benthic preda� tors since a low in biomass barren community of a burrowing mollusc L. balthica does not provide enough forag� ing resources to feed stocks of waterfowl. The L. balthica�community could be used as an indicator of climate changes in the future – it is predicted that a reduction in sea ice volume will improve conditions for growth of L. balthica and may therefore lead to an increased body size and biomass of bivalves in the shallows.


Introduction
The Arctic Ocean has one of the most sensi� tive to environmental changes ecosystems on �arth ��piridonov et al., 2012). Declines in sea ice thick� ness in the Arctic Ocean, along with an increase in air temperature, ocean acidification and anthropogenic pressures from offshore industries lead to changes in the marine ecosystems of the Arctic regions �Kwok & Rothrock, 2009). The benthic fauna is often used to observe contemporary changes in the environ� ment, since various environmental factors including the availability of organic matter and water temper� ature, as well as human activities, have direct impact on biomass and the composition of macrozooben� thos assemblages �Denisenko et al., 2003;Hinz et al., 2009). Recent publications revealed that benthic invertebrates also tend to ingest and accumulate mi� croplastics from the water column �Courtene�Jones et al., 2017;La �eur et al., 2019). Hence, there is a need for expanding the current knowledge on mac� robenthos, especially in the Arctic regions.
The �echora �ea in the southeast ���) basin of the �arents �ea is characterised by shallowness, a specific hydrological regime and by an intense off� shore oil and gas exploration and production �Den� isenko et al., 2003). The macrozoobenthos of the �echora �ea accounts for approximately 35% of the benthic biodiversity in the �arents �ea and is gener� ally well�described for the deeper waters; however, there is a lack of data for the near�shore areas �Dahle et Denisenko et al., 2003;Kucheruk et al., 2003;�ukhotin et al., 2008;Denisenko N. et al., 2019). The macrozoobenthos of the �echora �ea is characterised by a high variability in spatial distribu� tion �«mosaic pattern») caused by alterations of sea� floor topography and sediment types. Presumably this is also applicable for the �echora �ay �Dahle et Denisenko et al., 2003).
The study area is within the �echora �ay, the large estuarine ecosystem that assures a huge pro� portion of continental runoff into the �arents �ea region. The �echora �ea receives approximately 2.5 million tonnes of terrigenous sediments annu� ally through the �echora estuary �Dobrovolsky & Zalogin, 1982). The �echora �ay is character� ised by a broad intertidal zone, with a tide height of 1. 1-1.5 m ��yshev et al., 2003;Denisenko N. et al., 2019). The ice thickness in winter reaches 1.5 m, freezing to the bottom of the shallow near�shore areas of the bay. �ediments in the bay are formed by clayey sands and are influenced by a continental runoff and permafrost abrasion �Denisenko N. et al., 2019). The latest review of macrozoobenthos of the �echora �ay was by Denisenko N. et al. �2019) and was based on samples collected in 1995. Twenty� two sites from a depth range of between 5 m and 18 m were studied in the north�east and central areas of the bay. Overall, the most common was a typi� cal estuarial assemblage with a strong dominance of Limecola balthica Linnaeus, 1758 �occurring at 9 of 22 sites). The L. balthica-community described by Denisenko N. et al. �2019) had a mean biomass of 130.3 ± 64.8 g/m 2 and comprised 34 species in total.
The study area lies within the 1 st zone of the Nenetsky �tate Nature Reserve named «�echora river estuary and a 2 km water territory surround� ing the Russky Zavorot peninsula». The Nenetsky �tate Nature Reserve covers the River �echora es� tuary and nearby islands. In total the state nature reserve covers an area of 3134 km 2 of which more than a half �1819 km 2 ) corresponds to marine areas �Nenetsky Zapovednik, 2019). To safeguard the area from rapidly developing industrial activities, a state nature reserve was established in 1997. The main aim of establishing the nature reserve was the protection of important habitats for waterfowl that stopover in shallow waters of the �echora �ea dur� ing their migration from reproduction sites in West �iberia to variable wintering sites ��ukhotin et al., 2008). The Nenetsky �tate Nature Reserve pro� vides nesting and feeding grounds, and forms part of the migration routes for 125 species of water� fowl and coastal birds. This includes species from the Red Data �ook of the Russian Federation and IUCN Red List of threatened species, such as: the yellow�billed loon Gavia adamsii Gray, GR, 1859; �ewick's swan Cygnus columbianus bewickii Ord, 1815; the lesser white�fronted goose Anser albifrons �copoli, 1769; the red�breasted goose Branta ruficollis �allas, 1769 �IUCN, 2019).
Common species, such as the king eiders �Somateria spectabilis Linnaeus, 1758), form abundant flocks of up to tens of thousands of individuals to feed and molt before migrating to wintering grounds �Krasnov et al., 2002). Marine ducks, including king eiders, are specialised benthic feeders and their main prey items are bivalves ��ukhotin et al., 2008). For� aging macrobenthos were studied near the coasts of the Dolgy Island ��ukhotin et al., 2008;. A mismatch between ornithological and macrobenthic data for the region was first noted by �ukhotin et al. �2008). However, macrobenthic assemblages have previously never been studied in the shallows of the continental shore of the Nenet� sky �tate Nature Reserve. For the open�sea, sandy shallows �5-7 m deep) of the Medynsky Zavorot �eninsula, between the mouth of �echora �ay and Dolgy Island, a community dominated by Limecola balthica �previously known as Macoma balthica Linnaeus, 1758) was described at several stations �Kucheruk et al., 2003). The biomass of benthos in this community was low �ca. 3 g/m 2 ) and species diversity was scarce �totally16 species and 1-8 spe� cies per sample), explained by a strong wave action.
Ultimately, there are no macrobenthic data available for continental shores of the �rotected Area. The present study aimed at fulfilling the knowledge gap on shallow�water macrozoobenthos of the Nenetsky �tate Nature Reserve by describing macrozoobenthic assemblages of the �echora �ay and comparing them to other Arctic estuarine com� munities. The results can be used as a baseline data for further conservation or spatial planning activi� ties in the area.

Material and Methods
Site description �enthic samples were collected between 24 th and 30 th August 2016 in the Nenetsky �tate Nature Reserve from the inner �southern) bay of the Russky Zavorot peninsula �Fig. 1A). The inner coastline of the Kuznetskaya inlet forms a shore of the �echora �ay and is comprised of wetlands, protected from the waves and covered by vegetation �Fig. 1C). Tides on the inner shore are regular, semi�diurnal with a range of 1-1.5 m. �ottom sediments were formed of homogeneous sand with silts across the whole sampling area �Appendix 1). Sampling �amples were taken at 8 stations in 3 replicates from each site in the depth range between 1.1 m to 1.8 m at low tide from a rubber motorboat with a hand shovel with capture volume of 0.05 m 2 . �edi� ments were washed over a mesh size of 0.5 mm with sea water, then pre-fixed with 4% formaldehyde.
In the laboratory, pre-fixed animals were man� ually sorted out of the organic debris of the samples for species identification, studied and re-fixed in 70% ethanol solution. Macrobenthic invertebrates were studied and photographed under a binocu� lar microscope Leica165C and identified with the maximum level of certainty. All the species names were given in accordance to the World Register of Marine �pecies �WoRM�). For each sample iden� tified taxa were counted and weighed on a Jewelry �cale ML�CF3 to mg, unidentified fragments were also weighed and recorded as «Varia». �olychaeta fragments were counted both in an� terior and posterior fragments and the bigger value chosen for each species. �ivalve molluscs were weighed with exoskeleton.

Calculations
Microsoft �xcel and software �A�T �version 2.17) were used for data analysis �Hammer et al. , 2001;Hammer & Harper, 2006). Non�transformed data were used. �tandard diversity indices were used to charac� terise diversity �dominance, �impson index, �hannon index, Chao2), all calculations were performed using �A�T software package �Hammer & Harper, 2006). �pecies accumulation curve or sample�based rarefaction was used to assess how samples repre� sent biodiversity �species richness) predicted in the area. The predicted species richness � ) computed with Chao2�type estimator following Colwell et al. �2004), species accumulation curve was plotted in PAST with 95 percent confidence intervals (Ham� mer & Harper, 2006): where H -samples, S obs -the total number of observed species and S 1 -the number of species found in exactly one sample.
Mean values ± standard error are given for bio� mass and abundance.
Determination of macrozoobenthic assem� blages was based on biomass values. �pecies with the largest contribution to biomass at each station were considered as dominants, second and third largest -as subdominants.
Classical hierarchical clustering based on paired group �U�GMA) algorithm and non�metric multidi� mensional scaling �MD�) both based on �ray�Cur� tis similarity index �Hammer & Harper, 2006) were used to reveal whether the benthos across the sites formed distinct communities �groups). The �imilar� ity percentage ��IM��R) was applied to assess taxa contribution to differences between the groups. The one�way analysis of similarities �ANO�IM) with �onferroni�corrected p�values following Hammer & Harper (2006) was used to estimate the signifi� cance of differences between the groups. The �ear� son coefficient (r) was applied to assess correlations with environmental variables.
Maps were designed in ��RI ArcMap 10.4.1 using the standard GI� tools provided by the soft� ware. The reference coordinate system WG�84 and Universal Transverse Mercator projection �zone 40N) were used.

Results
A total of 14 taxa of benthic invertebrates were identified in 24 samples from 8 stations, 12 taxa identified to species level (Table 1). The fauna was mainly comprised of crustaceans �5 species), poly� chaetes �4 species) and bivalves �3 species), and with a single species of �riapulida and Insecta. �ivalves were the dominant group in both total biomass and abundance. The mean biomass of the macrobenthic invertebrates in the area was 21.31 ± 0.32 g/m 2 and the mean abundance was 2131 ± 1825 individuals per m 2 �ind/m 2 ). �rimary data on abundance, bio� mass and number of species per sample are present� ed in Appendix 1. A taxonomic matrix of species with images of the specimens is in Appendix 2.
The �hannon diversity was low �H' = 1.26), mac� rozoobenthos was represented by a small number of taxa with a few individuals and astrong dominance of one species �Table 2). The predicted number of species was the same as discovered �Chao�2 rich� ness = 14.19). The biodiversity was represented close to equally between the stations: four species had 100% occurrence in the study area, and only one spe� cies had < 20% occurrence �Table 1).
The species accumulation curve approached the plateau at the level of 7 stations, reaching 14 species �Fig. 2). The discovered diversity of macrozooben� thos was therefore fully representative for the area.
The hierarchical clustering and MD� plots showed three groups of stations in the study area �Fig. 3). Group A consisted of stations 5, 7 and 8; group � included stations 2, 3 and 6; and station 1 formed a separate group.
ANOSIM analysis showed statistically significant difference between group A and � �Table 3, p < 0.05).    3. Groups of stations in the study area determined by the hierarchical clustering �carried out on macrozoobenthic biomass data) �A) and MD� ��). Three groups can be seen: A -stations 5, 7 and 8, �purple shading); � -stations 2, 3, 4 and 6 �green shading); C -station 1. The variation in biomass of three species of bi� valves �Limecola balthica, Yoldia hyperborea Gould, 1841 and Spio armata Thulin, 1957) assured difference between the stations as shown by �IM��R analysis �Ta� ble 4). The biomass of the bivalve L. balthica accounted for 83.7% contribution to the dissimilarity between the groups. Despite the statistical difference between the groups of stations, L. balthica remained the dominant species for all of the stations in the study area and dif� ferences were likely caused by natural variation in spa� were likely caused by natural variation in spa� tial distribution of the biomass of the common species. The macrozoobenthos in the study area was therefore formed by a monodominant community of L. balthica.
The species composition of each station and spatial distribution of the macrozoobenthos biomass �g/m 2 ) across the study area are shown in Fig. 4. The biomass of macrozoobenthos per station had no cor� relation with the depth range �r = �0.45, p = 0.2).

Discussion
The macrozoobenthos of the �echora �ea has been studied over nearly a hundred years �Zenk� evich, 1927; Dahle et al., 1998;Denisenko �. et al., 2003Kucheruk et al., 2003;�ukhotin et al., 2008;Denisenko N. et al., 2019). However, the vast majority of benthic surveys in this region were conducted on�board large research vessels at depths greater ~10 m. Hence there is a lack of macrozoobenthic data for near�shore and estua� rine zones. This is also true for most of the Arc� tic zones, where shallow estuarine ecosystems are usually out of focus of big surveys. The present study provides the first detailed report on macrozoobenthos of the continental shal� lows of the Nenetsky �tate Nature Reserve, al� though macrozoobenthos of the central and north� ern parts of the �echora estuary were discussed in a recent publication by Denisenko N. et al. �2019). The authors sampled twenty estuarine sites in the �echora �ay during the RV Geophysic cruise in spring 1995 and compared the �echora �ay macro� zoobenthos with that of the Ob bay in the Kara �ea �Denisenko et al., 1999). Recently Denisenko N. et al. �2019) reviewed their data.
Despite the small number of species found in the present study, the biodiversity of the macrozoo� benthos was representative for the study area since it matched the predicted species richness �Chao� 2 = 14.19, n = 14). �stuaries are typically character� ised by lower biodiversity of macrobenthic inver� tebrates compared to marine environments. Ratios between marine, estuarine and freshwater species in the composition of estuarine macrozoobenthos typically depend on the salinity of a particular site (Whitfield et al., 2012). The River Pechora estuary is characterised as a mesohaline zone with low spe� cies richness �Denisenko N. et al., 2019). The spe� cies richness is affected more by the granulometric sediment structure than organic matter content of sediments or water salinity variations within the �echora estuary �Denisenko N. et al., 2019). In the present study the fauna was comprised of marine species with a few brackish crustaceans. At least one brackish species was present at each station. Monoporeia affinis Lindstrцm, 1855 and Saduria entomon Linnaeus, 1758, occurred with the fre� quency of 0.38 �recorded for stations 7 and 8) and 0.25 �stations 1, 2 and 8) respectively.
Macrobenthos communities in the study area were dominated by Limecola balthica with the mean biomass of 21.31 ± 0.32 g/m 2 �varying from 1.33 g/m 2 on station 1 to 43.17 g/m 2 on station 2), and a total of 14 species. The observed differences in biomass between the stations were presumably a statistical artefact caused by natural variation in spatial distribution of biomass across the bay since the structure of dominance remained the same across the study area. �ediments also were the same across the study area supporting consistency of the benthic community. However, we acknowl� edge that an increasing research area would give a better understanding of the consistent patterns of species distribution across the bay.
In 1995, Denisenko N. et al. �2019) revealed five types of macrobenthic communities in the central and northern parts of the bay, as opposed to the only one observed in the present study. Of these five types, the community dominated by L. balthica was the most common overall and was characterised by the authors as typical for muddy� sand bottoms at depths of 5-10 m in the central part of the bay, with a strong dominance of L. balthica, a total of 34 species, moderate abundance and a biomass of 130.3 ± 64.8 g/m 2 �Denisenko N. et al., 2019). As for diversity, the majority of spe� cies found in the present study were also present in Denisenko's L. balthica�community, except for the following: Chironomidae gen.sp., Laonice cirrata M. �ars, 1851, Spio armata and Yoldia hyperborea. All these species are typical for shallow water and often found in terrigenous coastal muds. The closest to our study area site sampled in 1995 was site 24 characterised by the L. balthica�community with a biomass of 21.9 ± 1.8 g/m 2 .
The study area is characterised by extreme environmental conditions for macrozoobenthos, e.g. an ice thickness up to 1.5 m with most habi� tats at depths 1-2 m freezing to the bottom. The inner shore of Russky Zavorot �eninsula is isolat� ed from wind�drift convection and is most prob� ably impacted by the River �echora run off. Fur� thermore, in semi�isolated near�shore areas in the Arctic, continental stock in winter forms a fresh� water outflow that spreads under the ice, forming a layer of fresh water, forcing marine macrozoo� benthos to move deeper into sublittoral zones, as was also described for the Canadian Arctic ��llis, 1995). Therefore, the macrobenthic richness in the area is limited by ice thickness and freshwa� ter impact and is comprised of eurythermal and euryhaline forms, tolerant to fluctuations in both temperature and salinity. The sea ice thickness and under�ice freshwater impact are common limitation factors for intertidal and upper sublit� toral zones in the Arctic shallows �Mokievsky et al., 2016). Ultimately the community in the study area corresponds to the community with a domi� nance of L. balthica determined by Denisenko N. et al. �2019), though with lower richness and a biomass due to extreme conditions.
A further reduction of the Limecola�commu� nity was reported from the open shores of the �e� chora �ea �Kucheruk et al., 2003). �iomass values in this area were even lower than in the present study, whilst among driving factors of macroben� thic distribution authors emphasised not the ice thickness or freshwater input, but wind waves that disturb sea bottom down to several meters depth as it has been shown for shallowness near Dolgy Island �Denisenko �. et al. 2019). As in the present study, the dominant species were also represented by abundant juveniles only �up to 1000 ind/m 2 ) with a complete lack of adults.
Limecola balthica is an infaunal bivalve mol� lusc with a circumpolar distribution, common in the intertidal zones and estuaries, often dominant in soft bottom communities �Väinölä & Varvio, 1989). In the Arctic region, L. balthica is dominant within the �echora �ay as a relict species. A com� munity with dominance of L. balthica and Cyrtodaria kurriana Dunker, 1861 was also described from the shallows of �aydara �ay �Kucheruk et al., 1998), but it was absent in typical Arctic estuaries, such as the Ob �ay in the Kara �ea in 1995 �Den� isenko et al., 1999). Limecola balthica is evidently sensitive to the climate change, as was shown for stock from the western Wadden �North) �ea where an increase in water temperature resulted in a lower reproductive output and an earlier spawning period ��hilippart et al., 2003).
Among the most common waterfowl of the Nenetsky �tate Nature Reserve is the king eider that feeds on macrobenthic invertebrates, often molluscs ��run, 1971; Lovvorn et al., 2003;Mer� kel et al., 2007). It was shown for Dolgy Island in the �echora �ea that marine ducks including the king eider were predominantly feeding on mussels Mytilus edulis Linnaeus, 1758, that were unreach� able for traditional techniques of benthic research and therefore often underestimated ��ukhotin et al., 2008). Unlike the Dolgy Island research area, in the Russky Zavorot �eninsula any rock outcrops or fragments of mussel shells were not recorded during the present study. It is unlikely that the Limecola balthica�community in the shallows of Russky Zavorot peninsula provides enough forag� ing resources to sustain big stocks of waterfowl.
It has been repeatedly predicted in the lit� erature that increasing temperatures in the Arctic will affect benthic communities, which could lead to changes in species distribution and interaction, allow the introduction of new species, and en� able a decrease of arctic species alongside with an increase of boreal species in the composition of benthic fauna �Lambert et al., 2010; Josefson et al., 2013;Renaud et al., 2015). The �arents �ea has been identified as a hotspot for «atlantifica� tion» of seawater and the expansion of boreal spe� cies �Renaud et al., 2015;Vihtakari et al., 2018). It is likely that effects of climate change on the macrozoobenthos of the �echora �ay will appear in the foreseeable future. In the study area, the thickness of sea ice is a key limitation factor for macrozoobenthos, therefore reduction of the sea ice volume would likely improve conditions and habitat availability for the L. balthica�community which leads to increasing the body size and bio� mass of molluscs inhabiting shallows. To achieve a broader understanding of the biomass trends of L. balthica stock in Nenetsky �tate Nature Re� serve near�shore areas, the present study can be considered as a baseline with further regular ob� servations required.

Conclusions
A сommunity of macrobenthic invertebrates with a mean biomass of 21.31 ± 0.3 g/m 2 , strongly dominated by L. balthica, and comprising a total of 14 species, was described based on 8 stations in the shallows of the Russky Zavorot �eninsula at a depth of 1.1-1.8 m in August 2016. This is the first study of the benthos of the continental shal� lows of the Nenetsky �tate Nature Reserve. All studied sites were characterised by a muddy�sand substrate. The observed community represents a form reduced in richness and biomass of the L. balthica�community, described by Denisenko N. et al. �2019) from the central and northern parts of the River �echora estuary. At the periphery of its distribution, the community is attributed to sea ice thickness and freshwater impact and is therefore comprised of eurythermal and euryhaline forms. �pecies occurring in the shallows but absent in deeper habitats described by Denisenko N. et al. �2019) included Chironomidae gen.sp., Laonice cirrata, Spio armata, and Yoldia hyperborea.
It is unlikely that the shallows of Russky Za� vorot �eninsula play an important role as feeding grounds for benthic predators since a barren com� munity of L. balthica does not produce enough for� aging biomass. The state of the L. balthica�commu� nity can be used as an indicator of climate change in the future. We suggest that with a reduction in the volume of sea ice it is likely that conditions and habitat availability for the L. balthica�community will improve in the shallow waters, and this could lead to an increased size and biomass of bivalves. However, to gain a better understanding of L. balthica biomass dynamics in the �echora �ay, more regular observations are required in both near� shore and open water areas.
The studies undertaken up until now within the marine borders of the Nenetsky Reserve are still rather scarce and do not cover the whole range of shallow water habitats. As for many other marine reserves in the Arctic, there is an urgent need for detailed habitat mapping and diversity estimations.