Seasonality of marine calcifiers in the northern Barents Sea: Spatiotemporal distribution of planktonic foraminifers and shelled pteropods and their contribution to carbon dynamics

(cid:0) 3 ) and export production (mg m (cid:0) 2 d (cid:0) 1 ). The study area constitutes a latitudinal transect in the northern Barents Sea from 76 ˚ N to 82 ˚ N including seven stations through both Atlantic, Arctic, and Polar surface water regimes and the marginal and seasonal sea-ice zones. The transect was sampled in 2019 (August and December) and 2021 (March, May, and July). The highest carbon standing stocks and export production were found at the Polar seasonally sea-ice covered shelf stations with the contribution from shelled pteropods being significantly higher than planktonic foraminifers during all seasons. We recorded the highest production of foraminifers and pteropods in summer (August 2019 and July 2021) and autumn (December 2019) followed by spring (May 2021), and the lowest in winter (March 2021).


Introduction
The rapid increase in anthropogenic carbon dioxide (CO 2 ) in the atmosphere and the ocean uptake have changed and continue to change the water carbonate chemistry by reducing the pH, the carbonate ion concentration ([CO 3 2-]) and the calcium carbonate saturation state (ΩCaCO 3 ) (Feely et al., 2004).This process, known as ocean acidification, is thought to have irreversible consequences for marine calcifiers, such as planktonic foraminifers and thecosome (shelled) pteropods.In the past the reduction of calcification rates and biogenic calcium carbonate (CaCO 3 ) production, as well as damages to (aragonitic; Ar) shells have been attributed to ocean acidification and CaCO 3 undersaturation (Ω < 1) of the surface waters (Schiebel, 2002;Fabry, 2008;Hunt et al., 2008;Moy et al., 2009;Manno et al., 2017;Peck et al., 2018;Bednaršek et al., 2019 and references therein).However, damages to the aragonitic shell of the pteropod Limacina helicina have been observed even in supersaturated (Ω Ar > 1) conditions of Ω Ar = 1.5 (Bednaršek et al., 2014a;Bednaršek et al., 2019).Because of their sensitivity to ΩCaCO 3 , the calcareous shells of planktonic foraminifers and pteropods are considered biological indicators of ocean acidification (Orr et al., 2005;Fabry 2008; Moy et al., 2009;Bednaršek et al., 2012b).Furthermore, they have been reported to play an important role in the marine carbonate cycle and can affect the buffer capacity of the ocean by CaCO 3 production, export and dissolution (Schiebel, 2002;Ziveri et al., 2007;Langer, 2008;Bednaršek et al., 2012a;Buitenhuis et al., 2019;Subhas et al., 2022;Ziveri et al., 2023).
Planktonic foraminifers are protists with a shell made of calcite and are found in all oceans, from low to high latitudes.They mainly inhabit the upper 300 m of the water column and are transported passively by ocean currents (Hemleben et al., 1989).Previous studies have reported the absence of diel vertical migration in high latitudes (Manno and Pavlov, 2013;Greco et al., 2019;Meilland et al., 2020;Ofstad et al., 2020;Anglada-Ortiz et al., 2021).When they die, their shells sink and accumulate on the seafloor and in the sediment.They preserve in the sediment when the seabed is above the calcite compensation depth (CCD) or dissolve otherwise, hereby playing an important role in the marine carbonate cycle and alkalinity budget (Schiebel, 2002;Jonkers and Kučera, 2015).Even though the seasonal distribution of living foraminifers has been studied for a long time (Allan, 1960) there is only a limited number of studies focusing on Arctic areas and the southern Barents Sea (Ofstad et al., 2020) especially outside of the summer period.
Shelled pteropods are holoplanktonic gastropods found in all oceans.Their shells are made of aragonite, a metastable form of CaCO 3 , which is more sensitive to changes in the water carbonate chemistry than calcite (Bednaršek et al., 2012b;Manno et al., 2017).The presence of pteropod shells in the fossil record is restricted to sediments above the aragonite compensation depth, shallower than the CCD (Gerhardt and Henrich, 2001;Peijnenburg et al., 2020).However, they also play an important role in the carbonate cycle by exporting (mainly) inorganic carbon from the ocean surface (e.g.Anglada-Ortiz et al. (2021); Knecht et al. (2023); Ziveri et al. (2023)).To our knowledge, and similar to the foraminiferal fauna, the northern Barents Sea has never been studied to track the seasonality of the pteropod fauna.
The Barents Sea (Arctic Ocean) is a relatively shallow shelf sea (average water depth ~230 m) which currently experiences rapid warming, in both the atmosphere and the ocean (Dalpadado et al., 2014;Descamps et al., 2017).Coupled with a decline in sea-ice cover, the direct gas exchange with the atmosphere is predicted to increase (Bates and Mathis, 2009).The northern Barents Sea region of the Arctic is expected to be more affected by ocean acidification because of its already low carbonate saturation state (Ω) of calcite and aragonite and the higher solubility of CO 2 in cold waters (Chierici and Fransson, 2018).The study area is characterized by strong seasonal changes in light intensity and sea-ice cover.These parameters mainly drive the primary production and the availability of nutrients, together with surface stratification (Bluhm et al., 2015).Over the last few decades an increase in primary and secondary production have been observed in the Barents Sea and Arctic Ocean (Dalpadado et al., 2014;Arrigo and van Dijken, 2015;Lewis et al., 2020).In the northern Barents Sea, the primary production is characterized by a spring (phytoplankton) bloom, occurring between April and July, when the sea ice melts and retreats (Sakshaug, 1997;Lee et al., 2015).The spring bloom may be followed by a second bloom in late summer (Wassmann et al., 2019).These blooms are the most important food source for the zooplankton (Sakshaug, 1997 and references therein).Advection of Atlantic and Arctic/Polar Waters bring not only nutrients but phytoplankton and zooplankton to the northern Svalbard margin (Wassmann et al., 2019).
The marginal ice zone (MIZ) is a frontal system between Atlantic and Arctic/Polar Water (Sakshaug and Skjoldal, 1989) and characterized by high productivity and seasonality, mainly close to the sea-ice edge (Reigstad et al., 2002).This production is strongly linked to mixing of Atlantic and Arctic Water, meteorological and sea-ice conditions (Wassmann et al., 1999 and references therein).The MIZ has been expanding northwards since 1870 and accelerating since 1970 (Kinnard et al., 2008).The MIZ also affects the benthic community.A study by Saher et al. (2012) on the benthic foraminifers Nonionellina labradorica, a sea-ice edge indicator, showed that its distribution has been pushed northwards (100 km) as the summer sea-ice edge has moved northward during the last few decades compared to c. 40 years old data previously reported by Steinsund (1994).The seasonal ice zone (SIZ) is the transitional zone between the winter and the summer sea-ice edges (Wadhams, 1986), where the seasonally retreating and expanding sea ice generates a productive area between the open sea and the drifting pack ice (Wassmann and Reigstad, 2011).
Despite the vulnerability to ocean acidification and strong seasonality of the Northern Barents Sea, little is known about the distribution of marine calcifiers, their present state of calcification and how they would respond to ocean acidification.A recent study from the northern Svalbard margin reported that large (>500 µm) and medium-sized (250-500 µm) pteropods dominated the upper 50 m of the water column in late summer (September 2018), while medium (100-250 µm) and small-sized (<100 µm) foraminifers, dominated from 50 to 300 m at the same time (Anglada-Ortiz et al., 2021).The study also suggested that, in this region of the Arctic Ocean, pteropods compared to planktonic foraminifers contributed the most to the inorganic carbon standing stocks (66.6-96.5 %) and export production (56.7-98.4%)(Anglada-Ortiz et al., 2021).A study from the northern Barents Sea reported that adults and juveniles (>500 µm) of L. helicina dominated the assemblages from 0 to 300 m water depth in December 2019 during the polar night (Zamelczyk et al., 2021).
Our present study provides a seasonal quantification of carbonate contributions from foraminifers and pteropods from this remote and rarely studied Arctic region (Fig. 1).We estimate the seasonal and vertical distribution of the planktonic foraminifers and pteropods and their contribution to the inorganic and organic carbon standing stocks (µg m − 3 ) and export production (mg m − 2 d − 1 ) over a 650 km long southnorth transect from the central Barents Sea into the Arctic Ocean slope and Nansen Basin.These new data shed light on the contribution of the planktonic calcifying organisms to the carbon pump and their life cycle.This work will contribute to improve projections of environmental changes (e.g.ocean acidification) in the region and the reconstruction of past environments based on their fossil shells in sedimentary record.The sampling transect spans the Atlantic zone south of the Polar Front, over the marginal and seasonal ice zone north of the front comprising seven seasonally sampled stations.

Study area
The Barents Sea (annual mean Area = 1.47 10 6 km 2 , annual mean Sea Surface Temperature = 0.9 C, annual mean Sea Surface Salinity = 34.2(Sakshaug andSlagstad, 1991, Smedsrud et al., 2022)), is a shelf sea in the Arctic Ocean.It is influenced by both warm and relatively saline Atlantic Water flowing into the Arctic Ocean, and cold and relatively fresh Arctic Water coming in from the Arctic Ocean, this part being seasonally sea-ice covered (Sundfjord et al., 2020, Lundesgaard et al., 2022) (Fig. 1a).The Atlantic Water reaches the Barents Sea via the Norwegian Atlantic Current until it meets the southward flowing Arctic Water to form the Polar Front, where the first and southernmost station of this study is located (P1) (Fig. 1a).Environmental parameters, such as winds and currents, play an important role on mixing the water column south of the Polar Front, while north of the front a strong pycnocline between the light Polar surface Water and Atlantic Water exists during the productive season (March-October) (Sakshaug and Slagstad, 1991).North of the Polar Front, the stations P2 to P5 are influenced by the Arctic/Polar Water (Fig. 1a).These waters are created by different mixing processes, including surface cooling, sea-ice edge interactions, inflows of meltwater and Atlantic Water (Lundesgaard et al., 2022).The northernmost stations (P6 and P7), located at the northern Svalbard slope and Nansen Basin respectively, are influenced by cold Arctic/Polar Water, as well as Atlantic inflow through the West Spitsbergen Current (Fig. 1a).The strength of the Atlantic inflow varies seasonally, having its maximum during winter and minimum in summer (Vernet et al., 2019;Fer et al., 2022).Previous studies in the last decades have reported an increase in Atlantic (and warmer and more saline) inflow and an increasing abundance of subpolar organisms advected through the west Spitsbergen current to the northern Svalbard margin, termed "Atlantification" (Bjørklund et al., 2012;Polyakov et al., 2020;Anglada-Ortiz et al., 2021).
Station P1, south of the Polar Front, is ice-free year-round.The location of the sea ice edge changes seasonally and interannually, being at its maximum during March and at minimum during the month of September (Fetterer et al., 2017).In 2019, the sea-ice margin retreated from 75 N (1st March 2019) to 80 N (16th September 2019), and in 2021 from (below) 75 N (1st of March 2021) to 82 N (16th September 2021) (from Norwegian Meteorological Institute Ice service, 2022).The stations P2-P7 were seasonally sea-ice covered during the study period (see Fig. 1b), however the sea-ice edge in this region retreated to above 82 N in September and October 2018 (e.g.Anglada-Ortiz et al. (2021); Pieńkowski et al. (2021)).No fast ice (sea ice attached to land (Jacobs et al., 1975)) was recorded during the sampling period in the region (see Fig. 1b).

Sampling and sample analyses
Plankton samples were collected onboard the RV Kronprins Haakon during the seasonal cruises of the Norwegian national Nansen Legacy Project to the Barents Sea in 2019 and 2021 (Table 1).Seven stations were sampled along a latitudinal transect east of the Svalbard archipelago (28.8 -34 ˚E), from 76 • N to 82 • N covering the shelf, slope, and deep Nansen Basin, and crossing the Polar Front, the SIZ and MIZ (Fig. 1).The stations are numbered from south to north and classified as the Atlantic shelf station (south of the Polar Front) (P1), shelf stations P2-P5 (north of the Polar Front and in the MIZ), slope station (P6) and Nansen Basin station (P7) (Table 1 and Fig. 1).Data from the December cruise (absolute and relative abundance of planktonic foraminifers and pteropods) have been published in Zamelczyk et al. (2021).Data of normalized size, protein content, organic and inorganic standing stocks, and export production of the planktonic foraminifers and pteropods sampled in December 2019 are new to this study as are all other data from the other seasons in 2019 and 2021.
Immediately after the recovery, the samples were wet sieved though a cascade of sieves of mesh sizes 500, 250, 100 and 64 µm.Living specimens of pteropods and planktonic foraminifers from all size fractions obtained (>500 µm = large size fraction, 250-500 µm = medium size fraction, 100-250 µm = small-medium size fraction, and 63-100 µm = small fraction) were wet picked from the upper 100 m of the water column for protein extraction and measurements (see 2.3 Organic and inorganic carbon contribution) and frozen at − 80 • C. The rest of the samples were frozen at − 20 • C and were analyzed in the laboratory of the Department of Geosciences, UiT the Arctic University of Norway (Tromsø, Norway).
Each frozen sample was thawed and planktonic foraminifers con-taining cytoplasm and pteropod shells with the animal inside were wet picked and counted.The absolute abundance (individuals per cubic meter (ind m − 3 )) was calculated dividing the number of specimens by the volume of water sampled with the multinet.The volume of water was calculated by the equation: with D being the sampled depth interval.We classified the foraminifers by size fractions as follows: 63-100 µm as small, 100-250 µm as medium, and 250-500 µm as large.For pteropods, we have attributed each size fraction to the life stage of individuals as follows: 63-100 µm (early veliger stages), 100-250 µm (veliger or early juveniles), 250-500 µm (juveniles), and >500 µm (adults).
Based on the absolute abundances per season, station, and depth, and the average shell diameter of planktonic foraminifers and pteropods (see 2.3 Organic and inorganic carbon contribution), we calculated the average normalized size of a model organism of a planktonic foraminifer and a pteropod (see 3.2 Seasonal and spatial distribution of marine calcifiers).

Statistical analysis
The statistical analyses were performed using the ggplot2 package from H (2016) from the Rstudio (version 4.2.1)software.To study the relation between our dataset and the environment (salinity and temperature, nutrients, chlorophyll a, pH, calcite and aragonite saturation states we have performed a Principal Component Analysis (PCA) and fit the distribution of planktonic foraminifers and shelled pteropods (separately) and the water masses.Moreover, we have performed a multiple linear regression and an Analysis of Variance (ANOVA) to assess the effects of environmental parameters on the abundance of foraminifers and pteropods separately, and to understand which factors best explain their distribution.

Organic and inorganic carbon contributions
The organic carbon was estimated as the individual protein content (as reported by Meilland et al. (2016) and Schiebel and Movellan (2012)) of 148 specimens of planktonic foraminifers and 300 specimens of pteropods that were individually and randomly picked from all stations and seasons onboard and were frozen at − 80 C (see 2.2 Sampling and sample analyses).The individual protein content is used as a proxy to estimate the organic carbon content of the organism, where 1 mg of protein equals to 1 mg of organic carbon.We followed the BCA (bicinchoninic acid) protocol from Meilland et al. (2016) using the nanospectrophotometer (NanoDrop 2000®) at the Department of Arctic and Marine Biology, UiT the Arctic University of Norway (Tromsø, Norway).This technique does not affect their aragonitic and calcitic shells, allowing us to use them for further analyses, e.g.size measurements (diameter and mass) or scanning electron microscopy (SEM).
The carbon standing stocks (µg m − 3 ) of foraminifers and pteropods have been estimated by extrapolating their protein content and shell diameter, for the organic or inorganic contribution, respectively (see 3.3  Organic and inorganic carbon of marine calcifiers) and integrating their absolute abundances from the upper 100 m of the water column following the published literature (Schiebel and Hemleben, 2000;Schiebel, 2002;Bednaršek et al., 2012a;Meilland et al., 2016;Anglada-Ortiz et al., 2021).Similarly, the export productions (mg m − 2 d − 1 ) have been estimated using protein content and shell diameter, their abundances between 50 and 100 m (or 80-100 m), except for station P6 in December, which was 200 m, and their test sink velocity (Schiebel and Hemleben, 2000;Schiebel, 2002;Bednaršek et al., 2012a;Meilland et al., 2016;Anglada-Ortiz et al., 2021).

Environmental properties of water masses
In summer (August and July) and late autumn (December) we observed a wider range in temperatures and salinities, associated with the higher atmospheric temperatures and melting of sea ice, compared to winter (March) and spring (May) (Fig. 2a, 2b).In terms of temperature, the slope station (P6) and basin station (P7) (ice covered during all cruises, see Fig. 1b) varied less along the seasons than the other stations (see Fig. 2a).Moreover, we observed lower surface salinities in July, August, and December (Fig. 2b).The stations P5, P6 and P7 were associated with very closed drift ice during all sampling seasons, while stations P2 and P4 were associated with variable sea ice conditions, consisting of very open drift ice, open drift ice, and very close drift ice in August, July and May, respectively (Fig. 1b).In all seasons, the surface water (20-50 m) consisted of Atlantic and modified Atlantic Water at the stations P1, P6 and P7, while the other stations (P2-P5) were characterized by Polar Water and warm Polar Waters (Fig. 2c).

Seasonal and spatial distribution of marine calcifiers
A clear seasonal pattern of temporal and spatial distribution of the studied planktonic calcifiers has been identified.The overall highest seasonal absolute abundances of living planktonic foraminifers and pteropods (ind m − 3 ) were observed in August 2019, followed by July  Regarding the vertical distribution of marine calcifiers and their group relative abundance (planktonic foraminifers vs pteropods) and the environmental conditions, from now on we will refer to the stations with the following categories: shelf station south of the Polar Front (Atlantic station P1), shelf stations north of the Polar Front (Polar stations P2-P5), slope station (P6), and basin station (P7).The environmental parameters seem to follow a depth distribution, and in general the abundance of shelled pteropods is the highest when the temperature is low (Fig. S2).
In May, the highest abundance (50 ind m − 3 ) of calcifiers occurred at P2 at subsurface (50-100 m).In this season we observed an increasing presence of planktonic foraminifers (av 30%) when compared to March and to pteropods (av 70%) (Figs. 3 and 9a).The basin station was the only one dominated by planktonic foraminifers, while pteropods dominated the upper 300 m of the water column at the Polar stations and represented (approximately) half of both groups at the Atlantic and slope stations, (Fig. 9a) (see supplementary material for details).The foraminiferal community was dominated by small-medium sized organisms, while the pteropod community was dominated by veliger/ young juveniles (Figs. 3 and 4).
In August the highest (82 ind m − 3 ) abundances of calcifiers were found at station P5 at depth (100-150 m) and the lowest (4 ind m − 3 ) at stations P6 at depth (200-300 m) (Figs. 3 and 4).Opposite to the other stations, where high abundances were found at the surface (0-50 m) and decreasing with depth, the abundances at P5 (mainly pteropods) increase at depth (100-150 m) (Fig. 4).Almost no pteropods were collected from the slope (P6) and basin (P7) stations in this (or any) season (Fig. 4).In general, foraminifers dominated the upper 300 m of the water column at the Atlantic station, slope, and basin stations, while pteropods at the Polar stations, with exception of P4 (Fig. 9a, 9b).The planktonic foraminiferal community was dominated by small and smallmedium specimens, while pteropods by juveniles/young adults (Figs. 3  and 4) (see supplementary material for more details).

Protein content of foraminifers and pteropods
The protein content of 148 and 300 living foraminifers and pteropods, respectively, was correlated to the length of the organisms, being directly proportional for foraminifers and logarithmic for pteropods (Fig. 5).The protein-length of pteropods was better correlated than in the case of planktonic foraminifers (R 2 = 0.68 and R 2 = 0.2, respectively) (Fig. 5).The protein content of foraminifers, as well as their size, were significantly smaller in terms of values and variability compared to pteropods (Fig. 5).

Seasonal variability of planktonic foraminifers and pteropod size distribution
The normalized size of calcifiers based on their abundance, changed along the seasons.For both planktonic foraminifers and pteropods, we observed larger average sizes in December 2019, followed by August 2019, July 2021, and May 2021, and the lowest, in March 2021 (Fig. 6).The size range of foraminifers from 0 to 300 m and the upper 100 m was widest in December 2019 and May 2021 (Fig. 6a, 6b), and in the case of pteropods, in August and December 2019 and May 2021 (Fig. 6d, 6e).Below 100 m water depth, the highest size range of foraminifers was in March 2021 (Fig. 6g) and for pteropods, December 2019 (Fig. 6f).We did not observe larger organisms below 100 m that could suggest ontogenic vertical migration.

Discussion
In this study we have observed the highest abundance of planktonic foraminifers and pteropods in August and July, followed by December and May, and with a minimum in March.However, the largest diameter of calcifiers and the associated total carbon standing stock and export production were estimated for December, followed by August and July, May, and March.We find the highest production of foraminifers in summer in the Atlantic zones south of the Polar front and in the Arctic Ocean in the northern part of the MIZ (P1, P6 and P7; Fig. 3).For pteropods production is highest in the polar stations and along the MIZ and SIZ during most seasons (P2-P5; Fig. 4).

Pattern in abundance, seasonality and water masses
Due to difficulties of sampling and accessibility in the Arctic region, most studies have been carried out during the summer season.From all Arctic regions, planktonic foraminifers have been mostly studied in the Fram Strait (e.g.Carstens et al., 1997;Volkmann, 2000;Stangeew, 2001;Manno and Pavlov, 2013;Pados and Spielhagen, 2014;Greco et al., 2022).Here, the abundances of living planktonic foraminifers are 30-60 ind m − 3 in June-July-early August (Volkmann, 2000;Manno and Pavlov, 2013;Pados and Spielhagen, 2014), while the mean abundance of foraminifers in the Arctic Basin was 25.4 ind m − 3 (Tell et al., 2022).Carstens et al. (1997) reported different maxima in abundances along the Fram Strait in August of 1250 ind m − 3 and 100 ind m − 3 at 78 • and 80 • N, respectively.The abundances in the Barents Sea (6-12 ind m − 3 ) (Volkmann, 2000), are comparable to the current study (5-15 ind m − 3 , and 10-35 ind m − 3 in July and August, respectively).Ofstad et al. (2020) reported abundances in the southern Barents Sea in April (0-6 ind m − 3 ) comparable to May in the current study; while the highest were found in June (436 ind m − 3 ) and exceeding any of the abundances found in the summer months in the northern Barents Sea (Fig. 3).The higher values compared to this study could be attributed to a higher productivity in the southern Barents Sea compared to the northern part and/or influence of strong seepage of methane probably causing upwelling (Ofstad et al., 2020).The abundances found along the north Svalbard margin in September (2.3-52.6 ind m − 3 , Anglada-Ortiz et al., 2021) agrees with the values found in the northern Barents Sea in July and August (Fig. 3).
In general for pteropods, lower abundances were reported compared to the present study in the southern Barents Sea (Ofstad et al., 2020) and the northern Svalbard margin (Anglada-Ortiz et al., 2021), probably related to local differences in water masses and presence/absence of sea ice.Abundances from the Atlantic shelf station (P1) from August and July are comparable to the results from the southern Barents Sea in June and April, respectively (Ofstad et al., 2020).
The stations P1 (south of the SIZ), and P6-P7 (north of the SIZ) have generally the lowest (total) abundances in all seasons.Planktonic foraminifers are more abundant in the Atlantic influenced stations (P1, P6 and P7), while pteropods are more abundant in the Arctic productive stations P2-P5 (Figs. 9 and S3).The distribution of planktonic foraminifers observed in the current study is associated with temperature, with higher abundances in warmer waters (Atlantic influenced stations P1, P6 and P7) (Figs. 3,S2 and S3).Their vertical distribution does not follow a specific depth pattern, but it changes through seasons (Fig. 3).In spring and winter, their highest abundances are found at the upper  50-100 m of the water column, while in summer they are spread throughout the water column (Fig. 3) and potentially following the distribution of food.Their abundances and distribution are significantly explained (p < 0.05) by the temperature and nutrients (NO 2 -, NO 3 -and SiOH 4 ) (Table S6).Reported possible controlling factors of the distribution of calcifiers, foraminifers specifically, are temperature and chlorophyll (as a measure of surface productivity), but also sea-ice cover and therefore, inorganic nutrient availability (Volkmann, 2000;Pados and Spielhagen, 2014;Greco et al., 2019).Several studies found the highest abundances of planktonic foraminifers along the productive seaice margins in the Arctic Ocean (Carstens et al., 1997;Volkmann, 2000;Pados and Spielhagen, 2014).These studies were mainly carried out during the late spring or summer months (June-August) and some of them also included the dead (=empty) foraminifers.Our observations of the highest abundances of planktonic foraminifers and pteropods during the summer months (July and August) and at the stations located close to the sea ice edge and in the SIZ in all studied seasons concur well with previous data (Fig. 9a).
The absence (zero abundance) of planktonic foraminifers during winter (March) and the increasing values during spring (May) suggest two possible scenarios: planktonic foraminifers are either seasonally advected from the south by the Atlantic currents and/or during winter they are in a dormant stage resting within the sea ice (as reported by Nigam (2005); Ross and Hallock (2016); Meilland et al. (2022)).The repeatedly higher abundances found at the slope (P6) and basin (P7) stations, influenced by Atlantic currents, combined by the zero abundances found in March, suggest that both processes were at work and followed by their capacity to reproduce rapidly asexually, as observed in the Greenland Sea (Meilland et al., 2022).In the western Barents Sea (Storfjorden) planktonic foraminifers and shelled pteropods were found under ice in late winter (March 2003) (Werner, 2005).We believe therefore, a "nursery" role of the sea ice could exist during winter months for pteropods, but especially for foraminifers.This is the case for other groups such as copepods (Søreide et al., 2010).Specimens of N. pachyderma would overwinter as they do in Antarctica (Lipps and Krebs, 1974;Spindler and Dieckmann, 1986) and use a multigenerational strategy combining sexual and asexual reproduction to repopulate the environment successfully within a short time frame (Meilland et al., 2022).Recent laboratory experiments on living individuals of N. pachyderma captured from the Greenland Sea documented dormancy and inactivity stages (Westgård et al., 2023).
The low abundances of pteropods together with the smaller sizes in late winter might be due to presence of offspring from the late summer populations.The increasing proportion of larger organisms, as well as their normalized size may be indicative of their life cycle (Fig. 6).The pteropod species Limacina helicina, one of the most ubiquitous species in the Arctic, can be found from temperate to polar regions (Bednaršek et al., 2014b;Peck et al., 2016).It is most abundant in the Arctic stations P2-P5 likely following the spring and summer blooms of phytoplankton and zooplankton.Limacina helicina is considered an omnivore collecting food using their mucous webs (Lalli and Gilmer, 1989;Gannefors et al., 2005;Conley et al., 2018).At the same time, L. helicina, serves as an important food source for larger zooplankton, including the non-shelled pteropod Clione limacina, but also for fish, such as polar cod, and sea birds (Gannefors et al., 2005 and references therein).In our study L. helicina is most abundant in summer and autumn with large specimens, and less abundant and with juveniles in winter (March) and spring (May) (Fig. 4).The very low abundances found in March agree with the scarce presence (almost zero) reported during pre-spring bloom in a Canadian fjord (Wang et al., 2017).Our observed seasonal pattern is furthermore similar to other studies.In Kongsfjorden, Svalbard, L. helicina has a life span of one year, with one or two new generations per year (in spring and summer) (Gannefors et al., 2005).They reach their maximum abundance in late summer and can reach a maximum size of 13 mm (Gannefors et al., 2005;Wang et al., 2017).The highest flux of pteropods in deep sediment traps from the Norwegian Sea (Lofoten Basin at 69 ˚N, Bear Island at 76 ˚N and Fram Strait at 79 ˚N) was recorded in October (Meinecke and Wefer, 1990).Shallow sediment traps from the Fram Strait recorded a rapidly increasing flux of pteropods in summer (July-August) or early autumn (September-October) when it becomes stable until February (Busch et al., 2015).The distribution of shelled pteropods from our study is also associated with temperature, finding higher abundances in colder waters, (Arctic stations P2-P5) (Figs. 4 and S3).In general, they are mainly found in the upper 100 m of the water column (Fig. 4).However, in March we found veliger stages throughout the whole water column (Fig. 4).Their abundances are significantly explained (p < 0.01) by a combination of salinity, temperature, and nutrients (NO 3 -), thus showing association with Arctic waters (Table S6).

Seasonality in carbon standing stocks and export production
Despite the similar absolute abundances of planktonic foraminifers and shelled pteropods in the upper 100 m during summer months (August and July) (Figs. 3 and 4), foraminifers contribute on average 34% to the total (organic and inorganic) export production at 100 m, while pteropods, contributes c. 66% (Table S3).
The carbon standing stocks and export production is well correlated with the seasons.We suggest that the seasonality of carbon standing stocks and export production could be partially associated with the seaice edge, the MIZ and SIZ where we find the fresher polar surface water.The calcifiers follow the production of phytoplankton, specially diatoms (Wassmann et al., 1999) and the distribution of zooplankton such as copepods (Falk-Petersen et al., 1999).The highest values of export production recorded along the transect were found in the Arctic stations, P2-P5, where the MIZ was located during all sampling seasons (Figs. 8  and 9).The ice edge, the MIZ and SIZ have been previously described as the most seasonally productive zone for phytoplankton and other organisms that will likely be consumed by foraminifers and pteropods.In particular, the distribution pattern of the foraminifers along the transect in relation to productivity and sea ice distribution is relevant for studies that use foraminifers as proxies to reconstruct past climate and environment.The spatial and temporal variability of foraminifers are also key to better reconstruct past productivity in the fossil record based on the abundance and flux of their shells.In the northern Barents Sea, we have observed the highest foraminiferal export productions in early summer (July, 3.5 ± 3.38 mg CaCO 3 m − 2 d -1 ) followed by late summer (August, 2.32 ± 1.93 mg CaCO 3 m − 2 d -1 ) (Table S2).This is later than the peak phytoplankton bloom in the ice-covered northern Barents Sea (Wassmann and Reigstad, 2011), which results in an even more delayed foraminiferal export production (Fig. 8).This late foraminiferal production peak could also be because 2021 was a particularly cold year, keeping a larger (in terms of area) sea-ice cover in the study area and for a longer time than in 2019 (Fig. 1).The Arctic Ocean in general, and our study area in particular, have been reported as extremely variable in degree of sea-ice cover and light availability, resulting in a very strong seasonality and variability of biological production.
The seasonal chlorophyll concentration (=chlorophyll a) has been measured at all stations and previously published by Vader (2022).The highest values are found in July, followed by August, and May (Fig. S3).Planktonic foraminifers and pteropods are heterotrophs, feeding on both phytoplankton and smaller zooplankton.We would therefore assume that the higher production of these organisms would occur after the phytoplankton bloom.This has been observed in modelled seasonal distribution of mesozooplankton by Wassmann et al. (2019).However, the production of the calcifiers could be increasing at a slower rate (compared to smaller zooplankton) and their maximum delayed: the July-August maximum may have developed from the spring bloom, while the still high production combined with the larger sizes in December, from a potential late summer phytoplankton bloom.In May and July we observed the highest carbon standing stocks and export productions at the stations closest to the ice edge (P2-P4) and at the time of maximum spring and early summer phytoplankton productivity (Figs. 7 and 8).Moreover, we need to acknowledge the interannual variability in the Barents Sea region.It is still unclear if years with a higher influence of Atlantic Water (e.g.2018) could develop a higher production the following year that would hamper the comparison between years.In September 2018 we observed larger carbon standing stocks and export production north of Svalbard (Anglada-Ortiz et al., 2021) than in August and December.We could hypothesize that the carbon standing stock and export production in this region increase until October where it reaches its maximum and subsequently starts decreasing.However, we could also attribute the higher carbon standing stocks from Anglada-Ortiz et al. (2021) to that 2018 was a warmer year than usual, with no ice cover at 82 • N in late summer (September) retreating further to 83 • N in October (Rasmussen et al., 2018;Husum et al., 2020).In contrast to what we have observed during this seasonal study, in 2018 pteropods were found along the North Svalbard margin in the Arctic Ocean.Given the northward location of the MIZ in late summer 2018, the Arctic zone had spread far north and most likely the production moved along following the retreating sea-ice edge.

Conclusions
We identified a clear seasonal pattern in terms of production, size distribution and species abundances and export production of planktonic foraminifers and pteropods, observing the highest values in summer and autumn, and the lowest, in winter (March), as follows: • In winter (March 2021), with the largest sea-ice extent and with the edge of open and close drift ice located at its southernmost position (76.4 • N), is when the lowest abundances of calcifiers were found.
The negligible abundance of planktonic foraminifers (<0.4 ind m − 3 ), and the low abundance of pteropods (early veligers) resulted in the lowest carbon standing stock and export production.• In spring (May 2021) when the sea ice started retreating and where the sea-ice edge between open and close ice drift was located at P2, the abundance of foraminifers and pteropods slowly increased and hence, the carbon standing stock and export production increased compared to late winter.The pteropod community was dominated by both veligers and early veligers, while the planktonic foraminifers, by small and medium sized specimens.• In summer months, with decreasing sea-ice cover along the transect (P4 very open drift ice in July, and at the edge of open and close drift ice in August) the abundance values reached their highest.The significant abundances of large planktonic foraminifers (>250 µm) and the increased abundance of juvenile pteropods in August 2019 resulted in a higher carbon standing stock and export production compared to July 2021 (they do not differ strongly from the values found in May 2021).• In late autumn (December 2019), the sea ice covered all stations except the Atlantic station P1 and the southernmost polar station P2, which were at the edge of close and very close ice drift.The abundances in general did not increase, but the relative abundance of adult and juvenile pteropods (>500 µm) did and reached their maximum of all the seasons.In December, we observed the highest normalized size from all the seasons, and hence the highest average carbon standing stock.The average export production was slightly higher than in August.
Furthermore, we found the highest carbon standing stocks and export production of the calcifiers in the seasonal ice zone SIZ (P2-P4) during all seasons closely following the productivity patterns of phytoplankton and other zooplankton.The pteropod community dominates the total carbon standing stock and export production at all seasons, representing on average 83% of both estimates.The foraminiferal distribution pattern was explained by the combination of food availability and temperature and association with Atlantic Water, while the distribution pattern of pteropods was explained by the combination of temperature, salinity, and food availability and association with Arctic Water.
The abundances of marine calcifiers in the northern Barents Sea are expected to change under conditions of "Atlantification" and ocean acidification.The abundances of shelled pteropods will probably decline during years of increased Atlantic inflow, while foraminifers could be increasing.Decreased pH in the water column could result in a lower contribution from pteropods to the carbon standing stocks and export production.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1. (A) Location map of study area, including bathymetry and main currents (Atlantic, red arrows, and Polar ("Arctic"), blue arrows)) and strength ("current width") from R package Vihtakari (2020), and location of the Polar Front (black dashed line) from Loeng (1991).(B) Sea ice extent during the sampling months from the Norwegian Ice Service-MET Norway and bathymetry from NOAA National Centers for Environmental Information (NCEI); International Bathymetric Chart of the Arctic Ocean (IBCAO); General Bathymetric Chart of the Oceans (GEBCO).

Fig. 5 .
Fig. 5. Protein content (µg) of foraminifers (left) and pteropods (right) relative to shell length (µm) with the equations used to estimate organic content (see 2.3 Organic and inorganic carbon contribution).Note different scales on the x-and y-axis.

Fig. 7 .
Fig. 7. Panel A: Total carbon (organic and inorganic) standing stocks (from 0 to 100 m depth, µg m − 3 ) from planktonic foraminifers and shelled pteropods in five different seasons, each of them represented by a different color (August 2019: pink; December 2019: light pink; March 2021: light blue; May 2021: green; and July 2021: orange).Panel B: Detailed standing stocks at each station during the different seasons and information about sea ice cover (close and very close drift ice: white; open and very open drift ice: light grey; open water: grey) and seasonal ice zone (blue) (note different y-axes at panel B).

Fig. 8 .
Fig. 8. Panel A: Total carbon (organic and inorganic) export production (at 100 m depth, mg m -2 d -1 ) from planktonic foraminifers and shelled pteropods in five different seasons, each of them represented by a different color (August 2019: pink; December 2019: light pink; March 2021: light blue; May 2021: green; and July 2021: orange).Panel B: Detailed export production at each station during the different seasons and information about sea ice cover (close and very close drift ice: white; open and very open drift ice: light grey; open water: grey) and seasonal ice zone (blue) (note different y-axes at panel B).

Fig. 9 .
Fig. 9. Panel A: Depth integrated abundance (ind m − 3 ) of all size fractions of planktonic foraminifers (dark blue) and shelled pteropods (purple) from the upper 300 m of the water column (with exception of station P6 in December, which only considers the upper 200 m), and sea-ice edge from the Norwegian Ice Service-MET Norway (in light blue).The size of the circles represents the total absolute abundance.Panel B: Distribution and results of the two-way ANOVA test of planktic foraminifers (upper panels) and pteropods (lower panels) in the Arctic (P2 -P5) and Atlantic (P1, P6 and P7) influenced stations during all seasons.***p < 0.01.