The Influence of the Atlantic Water Boundary Current on the Phytoplankton Composition and Biomass in the Northern Barents Sea and the Adjacent Nansen Basin

Abstract

The modern Arctic is characterized by a decreased ice cover and significant interannual variability. However, the reaction of the High Arctic ecosystem to such changes is still being determined. This study tested the hypothesis that the key drivers of changes in phytoplankton are the position and intensity of Atlantic water (AW) flow. The research was conducted in August 2017 in the northern part of the Barents Sea and in August 2020 in the Nansen Basin. In 2017, the Nansen Basin was ice covered; in 2020, the Nansen Basin had open water up to 83° N. A comparative analysis of phytoplankton composition, dominant species, abundance, and biomass at the boundary of the ice and open water in the marginal ice zone (MIZ) as well as in the open water was carried out. The total biomass of the phytoplankton in the photic layer of MIZ is one and a half orders of magnitude greater than in open water. In 2017, the maximum abundance and biomass of phytoplankton in the MIZ were formed by cold-water diatoms Thalassiosira spp. (T. gravida, T. rotula, T. hyalina, T. nordenskioeldii), associated with first-year ice. They were confined to the northern shelf of the Barents Sea. The large diatom Porosira glacialis grew intensively in the MIZ of the Nansen Basin under the influence of Atlantic waters. A seasonal thermocline, above which the concentrations of silicon and nitrogen were close to zero, and deep maxima of phytoplankton abundance and biomass were recorded in the open water. Atlantic species—haptophyte Phaeocystis pouchettii and large diatom Eucampia groenlandica—formed these maxima. P. pouchettii were observed in the Nansen Basin in the Atlantic water (AW) flow (2020); E. groenlandica demonstrated a high biomass (4848 mg m⁻³, 179.5 mg C m⁻³) in the Franz Victoria trench (2017). Such high biomass of this species in the northern Barents Sea shelf has not been observed before. The variability of the phytoplankton composition and biomass in the Franz Victoria trench and in the Nansen Basin is related to the intensity of the AW, which comes from the Frame Strait as the Atlantic Water Boundary Current.

1. Introduction

The Arctic, particularly the northern part of the Barents Sea, has been identified as extremely sensitive to climate change [1]. The evidence of noticeable warming and its impact on the ice cover and thickness is well documented [2,3,4,5]. This significant change in the ice cover is primarily associated with increased Atlantic water (AW) inflow. The term “atlantification” was introduced to describe the complex transformations of the physicochemical properties of water during the interaction of Arctic water masses with AW [6,7,8,9,10,11]. A vital feature of the Arctic region is the pronounced seasonality of solar energy supply and the influence of atmospheric temperature on the rate of ice melting [4,5]. In addition, thermal convection from the depths occurs in the western Arctic (the Nansen Basin), which does not weaken in winter [12]. This can lead to a decrease in the thickness of the ice and, as a result, a reduction in the contribution of melted cold-water masses in the summer and an increase in the heat content of the upper water layer [1]. Ice is becoming seasonal [13], and if similar trends persist, it will be so across most of the Arctic Ocean in two decades [14,15].

Sea ice determines the physical processes in the water column; it regulates the transfer of heat and wind load at the atmosphere–ocean boundary and, as a result, determines the hydrography of the water column [16]. Replacing multi-year ice with seasonal ice changes its thickness and physical properties. The incoming light energy depends on the thickness of the ice [17,18], and the physical ice structure determines the phytoplankton composition of its inhabitants. When the ice melts, these species are transferred to the phytoplankton communities of open water [19,20,21,22,23,24,25]. First-year ice prevails in the western Arctic in the Barents Sea [26], and a similar trend has been noted north of Svalbard up to 82° N near the Yermak Plateau [27]. In the Nansen Basin, seasonal ice was observed up to 83° N [28]. The primary modulator of ice concentration in this area is the intensity of the AW inflow [7,10,29].

The increased influence of AW transforms the vertical hydrophysical structure of the Arctic Ocean and the Barents Sea [8,9,10,30,31]. The AW flow can distinguish the processes of advection and convection. When encountering the cold Arctic waters, the AW warm streams dive below the cold waters in the northern part of the Barents Sea and much further north [29]. The temperature difference causes convection. In the Nansen Basin, heat convection maintains the state of open water until February [12]; it enhances the vertical transport of saltier and nutrient-rich AW [32,33]. These processes restructure the ecosystems and replace native species with North Atlantic invasive species [34,35,36,37,38]. The phenomena associated with atlantification at the ecosystem level are accompanied by a change in the phenology of phytoplankton blooms and the northward movement of boreal species of all trophic levels from phytoplankton to fish [39,40,41,42,43,44,45]. It is believed that the transition from ice cover to open water also contributes to increased primary production [46,47,48], as ice limits photosynthesis by blocking the sunlight. However, intense phytoplankton growth also occurs under the ice [49,50,51,52,53], which can lead to the removal of nutrients from the water. When the ice melts, a layer of water on the surface, with low salinity and low nutrient concentration, does not contribute to the phytoplankton bloom. In addition, the thermocline is formed in ice-free water under the influence of atmospheric warming. At the upper mixed layer (UML), the growth of phytoplankton is limited by low concentrations of silicon and nitrogen [28]. The challenges in forecasting phytoplankton growth under ice, in MIZ, and in open water make it difficult to accurately predict the effect of ice melting on ecosystem structure and productivity [47,54,55] and, consequently, the ecosystem dynamics under climate shifts in the high Arctic. The Barents Sea ecosystem is heterogeneous both in space and in time. The AW flows play a fundamental role in creating this heterogeneity; the main AW flow passes through the southern part of the sea, and significantly less warm water enters the northern part [56,57]. The northern part of the sea is covered with ice in winter. On the other hand, warm AW prevents ice formation in the southern part, and open water is usually observed here all year round. Seasonal movement of the ice edge affects the structure and dynamics of the ecosystem. Previous studies have described the seasonal dynamics of phytoplankton in the southern and central parts of the Barents Sea [58,59,60]. In particular, in the central parts, the spring phytoplankton bloom in the MIZ has been associated with the intensive growth of colonial diatoms of the genus Chaetoceros, Fragilariopsis, and Thalassiosira, with some contribution from Phaeocystis pouchitii [59]. In addition, a high abundance of dinoflagellates characterizes the summer season [59,60].

A feature of the southern part of the sea is that in the AW water mass, blooms of the coccolithophorid Emiliania (Gephyrocapsa) huxleyi occur annually during summer (July and August), reaching densities of 12 × 10⁶ cells L⁻¹ in the UML [61,62,63,64,65,66]. These blooms have been found in surface waters with temperatures between 7.14 and 11.7 °C, low silicate and nitrogen concentrations, and a nitrogen to phosphorus ratio below the Redfield ratio [65]. The northern part of the sea has been the least explored due to its inaccessibility [67,68,69].

Except for isolated observations [28,70], the available data on the effect of AW on phytoplankton in the Nansen Basin and the northern Barents Sea shelf, east of 34° N, are insufficient. This lack of comprehensive data hinders our understanding of these areas’ complex physical, biogeochemical, and ecological processes.

The main driver of the productivity of Arctic ecosystems is the nutrient-rich AW. This study aims to test the hypothesis that phytoplankton composition and productivity are influenced by the AW flow. Specifically in the northern part of the Barents Sea, the phytoplankton composition and its productivity are determined by the Atlantic Water Boundary Current (AWBC), which is subject to spatial fluctuations due to the meandering of its flow. Therefore, this study aims to provide valuable information on phytoplankton dynamics in the high Arctic marine ecosystems and, as a result, improve our understanding of how such systems may respond to future environmental changes.

2. Methods

2.1. Sampling

Water samples collected on 3–6 August 2017, and 15–21 August 2020, during the 68th and 80th cruises of RV Akademik Mstislav Keldysh, were used to study the phytoplankton composition and biomass. In 2017, the research area was limited to 80° N, and the boundary of the ice cover extended to the northern part of the Barents Sea (Figure 1(upper)). In 2020, this boundary was much further north in the deep Nansen Basin (83° N) (Figure 1(lower)).

The stations were selected to include ecologically heterogeneous areas (deep water and shelf, MIZ, and open water) and geographical areas (north of the Barents Sea and the Nansen basin). In 2017, the sampling stations were located on the northern Barents Sea shelf (Figure 1(upper)) (Table S1) in the western part adjacent to Eastern Svalbard (stations 5553, 5554, and 5556) and the eastern part—coastal waters of Franz Josef Land (stations 5560 and 5561). The other four stations were positioned in the deepwater Franz Victoria Trench at its western (stations 5557, 5557A, and 5558) and central part (station 5559). Three days later, a second sampling was conducted at station 5557 (station 5557A) to assess the community’s stability.

In 2020, the research was conducted in a variety of locations, including the Franz Victoria trench (stations 6864 and 6866), the offshore zone adjacent to the Franz Victoria trench (stations 6854 and 6856), the offshore station as close as possible to AWBC, the AWBC flow (6862), and the marginal ice zone (MIZ) of the deepwater Nansen Basin (stations 6860 and 6861) (Figure 1(lower)) (Table S1). Stations 6854 and 6866 were located in the western part of the Franz Victoria trench; station 6864 was in its central part, and station 6866 was on the shelf in its eastern part.

2.2. CTD Data

An SBE 32 Carousel Water Sampler equipped with 10 L Niskin bottles was used, along with an SBE 911plus CTD (Sea-Bird Electronics, Inc., Bellevue, WA, USA) with a temperature sensor (measurement accuracy up to 0.001 °C), a salinity sensor (accuracy up to 0.001), and a pressure sensor (0.001 hPa). Vertical temperature and conductivity profiles were carried out at each station from the water surface to the seafloor. Sampling was conducted at different depths using CTD data and the data from the fluorescence sensor.

2.3. Phytoplankton Analyses

The samples were fixed with neutralized formaldehyde (the final concentration was 1%). A sedimentation method and sample decanting were used to concentrate the cells [72,73]. Phytoplankton samples were collected and processed from all 18 stations (n = 133) (2017—stations 5553–5561; 2020—stations 6854–6868) (Figure 1 (upper,lower)). Species identification and cell count were carried out under an Ergaval light microscope (Carl Zeiss, Oberkochen, Germany) in an aqueous preparation with an increase of 16 × 20 and 16 × 40. Cells with linear dimensions of less than 20 µm (nanoplankton) were counted in a 0.05 mL chamber. A 1.0 mL Nauman chamber was used to count large-size cells (microplankton). The biomass was determined using the volumetric method, equating the cell to any geometric shape [74,75,76]. Cells with a linear size of less than 2 µm were not considered when calculating phytoplankton’s total biomass. Formulas Menden-Deuer and Lessard [77] were used to convert wet-weight biomass into carbon units. Cells with an unknown taxonomic affiliation ranging from 4 to 10 microns were classified as “small flagellates”. The species identification was based on morphology and was carried out using the following sources [78,79,80]. The ecological (neritic, oceanic, pantalassic species) and phytogeographical (Arctic, boreal, boreal-Arctic species) characteristics of the species are given by [81,82]. A cell abundance of 1.0 × 10⁶ cells L⁻¹ and biomass of 1000 mg m⁻³ was used as the bloom level [83]. The dominant species had the highest biomass at a given station; the subdominant species was second after the dominant species.

2.4. Nutrient Analyses

Analyses of phosphate phosphorous (${\mathrm{P}-\mathrm{P}\mathrm{O}}_4^{3-}$), dissolved silicate (Si) and nitrate (${\mathrm{N}-\mathrm{N}\mathrm{O}}^{3-}$), nitrite (${\mathrm{N}-\mathrm{N}\mathrm{O}}^{2-}$), and ammonia (${\mathrm{N}-\mathrm{N}\mathrm{H}}^{4-}$) nitrogen were carried out with a segmented continuous-flow TechNicon II autoanalyzer or with colorimetric methods [84,85] on board the vessel using DR 5000™ UV-Vis Laboratory Spectrophotometer manufactured by Hach-Lange GmbH (Düsseldorf, Germany). Total dissolved inorganic nitrogen (N) was estimated as a sum of nitrate, nitrite, and ammonia.

2.5. Optical Measurements

Measurement of photosynthetically available radiation (PAR) at the sea surface and underwater in visible light in the spectral range of 400–700 nm was performed using devices manufactured by LiCOR (LI-COR Environmental, Lincoln, Nebraska, USA). The optical measurements of the PAR profiles were carried out only during daylight.

The diffuse PAR attenuation coefficient $K_d$ was calculated using the exponential law of variation of incident downwelling irradiance $E_d\left(z\right)$ with depth z [86] as follows:

$$E_d\left(z\right)=E_d\left(0\right)·e^{-K_{d}z}$$

$E_d\left(0\right)$—surface irradiance.

The euphotic depth ($z_{eu}$) was calculated as the depth where $E_d\left(z\right)$ was reduced to 1% of surface irradiance as follows:

$$z_{eu}=-ln\left({\displaystyle \frac{0.01}{K_d}}\right)$$

2.6. Statistical Methods

In the case of a comparison between two samples, a t-test was used. The level of significance was always set at p < 0.05. The PAST 4.13 software https://www.nhm.uio.no/english/research/resources/past/ accessed on 15 July 2024) was used.

3. Results

3.1. Environmental Conditions

3.1.1. Sea Ice

A unique feature of the studied area in 2017 was the presence of broken sea ice in the northern part of the Barents Sea; only the Franz Victoria Trench was relatively free of ice (Figure 1(upper)). In 2020, the compact ice cover retreated far to the north, and there was open water area in the Nansen Basin up to 83° N (Figure 1(lower)).

3.1.2. Water Masses and the Vertical Water Column Structure

The water masses were determined according to [69] with modifications for the Nansen Basin and the north of the Barents Sea (Table S2). Fundamental features of the vertical distribution of water masses were that the Polar water masses (PSW) were located closer to the surface, and the saltier, denser Atlantic water masses (AW) were located below (Figure 2 and Figure 3;

Table 1. Surface temperature and salinity, as well as thermocline depths, pycnocline, and halocline, in 2017 (north of the Barents Sea) and 2020 (the Nansen Basin).

Station	T, °C Surface	S, PSU Surface	Thermocline Depth, m	Pycnocline Depth, m	Halocline Depth, m
5553	4.65	34.16	14	15	14
5554	4.46	33.28	5	5	6
5556	0.06	32.37	13	14	5
5557	3.41	33.70	24	5, 20	5
5557A	3.53	33.58	20	20	20
5558	2.65	33.81	15	15	16
5559	−0.315	31.99	4, 15	4, 15	3, 15
5560	2.24	33.64	14, 18	9, 19	6, 20
5561	2.70	34.30	22, 26	22, 26	21
6854	2.695	32.98	14	15	15
6856	2.97	32.82	16	17	18
6860	−1.39	30.91	9	8	8
6861	−1.41	31.17	13	10	10
6862	−0.17	32.09	10	10	10
6864	3.31	33.01	23	21	21
6866	3.72	33.69	20	20	20

and

Table 2. Salinity, density, minimal temperature in the PSW core, and maximal temperature in the AW core as well as their depths in 2017 (north of the Barents Sea) and 2020 (the Nansen Basin).

Station	PSW				AW
	Depth, m	S, PSU	T, °C at the PSW Core	Density (σ), kg m−3	Depth, m	S, PSU	T, °C at the AW Core	Density (σ), kg m−3
2017
5553	54	34.92	1.90	27.92	90	35.015	2.556	27.94
5554	43	34.47	−1.27	27.75	148	34.88	1.219	27.94
5556	45	34.46	−1.75	27.74	235	34.90	1.394	27.94
5557	53	34.50	−1.71	27.77	171	34.755	0.526	27.88
5557A	60	34.50	−1.71	27.77	118	34.71	0.434	27.85
5558	60	34.50	−1.73	27.77	177	34.76	0.617	27.88
5559	49	34.52	−1.47	27.78	178	34.77	0.618	27.89
5560	37	34.48	−1.50	27.75	142	34.78	1.107	27.86
5561	47	34.58	−1.52	27.83	284	34.90	0.075	28.02
2020
6854	40	34.22	−1.31	27.53	133	34.87	2.01	27.86
6856	43	34.08	−1.73	27.43	245	34.92	2.43	27.87
6860	40	34.15	−1.77	27.49	213	34.92	2.55	27.86
6861	63	34.20	−1.77	27.53	175	34.90	2.66	27.83
6862	42	34.20	−1.42	27.52	138	34.87	2.22	27.85
6864	41	34.17	−1.43	27.49	115	34.84	1.77	27.86
6866	75	34.56	−1.10	27.80	106	34.73	−0.01	27.89

). Notably, at stations 6860 and 6861 in the MIZ zone in 2020 and station 5559 in 2017, a thin layer of meltwater (mPSW) with water temperatures <0 °C and salinity below 32.0 was registered. This thin layer, observed near the surface, was formed above the halocline, a layer of water in which the salinity changes sharply with depth. In the MIZ, saltier and warmer PSW are located below mPSW, and thermocline, halocline, and pycnocline are formed between them (Table 1). In open water, the surface is heated, the temperature rises above 0 °C, and a seasonal thermocline occurs, the depth of which varies from 5 to 26 m. We defined the depth with the lowest temperature as the PSW core, which varies from 37 to 60 m in 2017 and from 40 to 75 m in 2020; no differences in the location of the PSW core in 2020 and 2017 were found (p = 0.6) (Table 2). The temperature in the PSW core was <0; the exception was station 5553, where the temperature in the PSW was >0 °C. The average temperature in the PSW core was −1.63 °C in 2017 and −1.52 °C in 2020; however, the difference between the two values was not statistically significant (p = 0.5).

The saltier and warmer AW was below the PSW, and we defined the depths with maximum temperatures as the AW core; it was located below 161 m and 171 m in 2020 and 2017, respectively (Table 2). These differences are not significant (p = 0.71). In 2020, no AW was found at station 6866. In 2017, when moving north, AW transformation occurred; it deepened, and the temperature in the core decreased (Figure 2 (upper) and Figure 3, Table 2). In 2020, the highest temperature in the core of AW was observed in the Nansen Basin. When moving south, it decreased; at station 6866, it became less than 0 °C.

3.1.3. The Photic Layer

The PAR intensity on the surface changed by more than an order of magnitude during measurements (

Table 3. PAR (μM photons·m^–2·s⁻¹) at the water surface and at the depth of the euphotic zone $z_{eu}$ in 2017 and 2020.

Station	Surface PAR	zeu, m
		1%
	2017
5553	570	38
5556	776	25
5557	451	48
5557A	459	43
	2020
6860	184	35
6861	169	46
6862	148	30
6854	279	39
6856	144	43
6864	139	60
6866	59	64

). The average depth at which light intensity was 1% of that on the surface did not differ significantly between 2017 and 2020 and was equal to 38.5 ± 9.9 and 45.3 ± 12.6 m, respectively.

3.1.4. Nutrient Concentration

The Si, N, and P concentrations in 2017 were significantly higher in AW than in PSW, and differences in N concentrations were insignificant in 2020 (

Table 4. Comparison of the average Si, P, and N concentrations and their ratios in UML, PSW, and AW in 2017 and 2020; bold p values indicate significant differences.

Nutrients		2017			2020
		Mean:	p	p	Mean:	p	p
Si	UML	0.27	0.01		0.51	0.0006
	PSW	1.24		0.0000001	1.99		0.0000001
	AW	3.99			4.71
P	UML	0.07	0.0006		0.08	10−14
	PSW	0.35		0.0001	0.49		0.00045
	AW	0.63			0.73
N	UML	0.63	0.00088		0.01	0.00000025
	PSW	6.02		0.000001	5.81		0.71
	AW	12.64			5.40
N:P	UML	10.08	0.2		0.77	0.0000003
	PSW	20.52		0.6	11.41		0.005
	AW	24.12			15.05
Si:N	UML	0.42	0.11		336.84	0.03
	PSW	0.25		0.000001	0.35		0.0115
	AW	7.09			0.43
Si:P	UML	3.67	0.8		24.61	0.23
	PSW	4.06		0.038	3.92		0.0001
	AW	7.09			6.48

). Except for N:P, the nutrient ratio also showed significant differences in 2017. Deficient Si, N, and P concentrations were recorded above the seasonal thermocline (UML) compared to PSW in 2017 and 2020; all nutrient ratios in 2017 were not statistically significant; in 2020, there were no differences in Si:P.

At all stations in the offshore zone, the nutricline was located in the photic zone at a depth below the pycnocline occurrence (Tables S3 and S4). In the AWBC area, the nutricline was located above a depth of 35 to 45 m. In the MIZ of the deep pool, the nutricline was fixed at a depth below 20 m. In 2017, in the northern part of the Barents Sea, the nutrient concentrations were close to limiting in UML. At some stations, the concentration of nitrates in the upper 20 m water layer was close to zero (station 5558). The observations of Si and N dynamics in the southern part of the Franz Victoria trench in 2017 indicate significant changes in their concentrations in the UML over a very short time (Table S3). For three days, the concentration of Si at station 5557 (5557-A) decreased, while the concentration of nitrates increased four times.

3.2. Phytoplankton

3.2.1. Phytoplankton Composition

A comparative analysis of the phytoplankton composition throughout the studied water area revealed a high degree of similarity between the communities in both years of research. The total number of species was 121 and 136 in 2017 and 2020, respectively (Table S5). When comparing the total number of species per station (Table S5), we found 45 species at MIZ stations and 41 species in open water, with no significant differences (p = 0.69). However, on the shelf (38 species) and in deep water (58 species), the differences approach significance (p = 0.05). The average number of species per station in 2017 (34) was significantly different from 2020 (51 species) (p = 0.02). In both years, the highest number of species was characteristic of the genera Gymnodinium and Protoperidinium. Diatoms with the genera Thalassiosira and Chaetoceros were in second place. In both years, almost the same species were recorded for each genus.

3.2.2. The Main Ecological and Phytogeographic Complexes of Phytoplankton

In the summer, the phytoplankton of the studied area can be distinguished into three main complexes: 1—a complex of species associated with multi-year ice; 2—a complex of species related to first-year ice; and 3—a complex of Atlantic open water species (Table S6). Complex 1 was typical, mainly for the MIZ deep water near the boundary of the long-term pack ice. Complex 2 included the neritic boreal-Arctic species of the genus Thalassiosira and the cryopelagic species, Porosira glacialis. These species produced diatom blooms in the MIZ of the studied area both on the shelf (Thalassiosira spp.) and in the deep water (P. glacialis). Complex 3 included the most characteristic Atlantic species—spring bloom, oceanic, and warm-water indicator species. Cold-water Atlantic species that form spring blooms in open water (Eucampia groenlandica, Phaeocysis pouchettii) are found only in the AWBC flow (station 6862), the adjacent section of the northern shelf (station 6854), the Franz Victoria Trench (stations 5557-A and 5558), and partially in the deep water (station 6860). Oceanic large-size diatoms from the genus Rhizosolenia were most fully represented in the deep water at depths from 5 to 40 m (station 6860) and 34 to 90 m (station 6861). These species were also found in isolation in the main AWBC flow in the upper 40 m water layer in 2017 in the Franz Victoria Trench. Large-size diatom E. groenlandica bloomed at stations 5557-A and 5558. Species of the genus Chaetoceros (C. atlanticus and C. danicus) have been recorded only on stations 6861 and 6862. The presence of Atlantic species of the genus Rhizosolenia and Chaetoceros in the lower depths of the Franz Victoria trench in 2017 indicated the influence of AW since these species belong to the Atlantic complex, i.e., they are the indicators of AW (Table S6). The main indicator species of warm Atlantic waters is the coccolithophorid Emiliania (Gephyrocapsa) huxleyi, found in small abundance in the surface layer of open water on the northern shelf, in the Franz Victoria trench, and in the AWBC flow. At a depth of 602 m in the AWBC zone, the abundance of this species reached 1.6 × 10⁴ cells L⁻¹. In the MIZ, this species was represented only by individual coccoliths. Warm-water Atlantic species from the genus Ceratium (Tripos) were more common in the summer of 2017, indicating the depth of penetration of warm Atlantic waters into the high Arctic.

3.2.3. The Dominant Species of Phytoplankton

In both years, the dominant species amongst the phytoplankton of the MIZ, both on the shelf and in the deep water, were neritic boreal-Arctic diatom species associated with first-year ice (

Table 5. Maximum abundance (the upper numbers, cells L⁻¹) of dominant phytoplankton and the depth (the lower numbers, m) of this maximum in August 2017.

Species	Shelf				The Frantz Victoria Trench
	MIZ	Open Water					MIZ
	5556	5561	5553	5554	5560	5557	5558	5559
Porosira glacialis	108 5	6 30	0	0	0	6 52	6 52	0
Thalassiosira spp.	6.1 × 105 22	192 55	0	40 90	12 5	12 52	600 100	66 15
Eucampia groenlandica	100 20	0	0	0	0	1.4 × 105 52	9.7 × 105 45	20 80
Thalassiosira hyalina	120 22	1.3 × 105 30	0	0	0	0	240 100	40 15

Notes: Thalassiosira spp. = Thalassiosira gravida + Thalassiosira nordenskioeldii + Thalassiosira rotula.

and

Table 6. Maximum abundance (the upper numbers, cells L⁻¹) of dominant phytoplankton and the depth (the lower numbers, m) of this maximum in August 2020.

Species	Deepwater AWBC	Slope		Deepwater AWBC		The Frantz Victoria Trench
	Open Water			MIZ		Open Water
	6862	6854	6856	6860	6861	6864	6866
Porosira glacialis	130 245	20 38	1.2 × 103 50	3.3 × 105 10	264 3	4 571	0
Thalassiosira spp.	1.8 × 104 30	4.8 × 103 38	100 292	1.0 × 104 10	136 3	60 111	50 384
Eucampia groenlandica	1.6 × 104 30	3.0 × 104 38	0	260 10	0	0	0
Thalassiosira hyalina	9.6 × 103 30	32 200	44 140	100 20	0	8 111	0
Phaeocystis pouchettii	2.6 × 106 38	6.0 × 104 38	1.6 × 104 40	8 200	0	0	0

Notes: Thalassiosira spp. = Thalassiosira gravida + Thalassiosira nordenskioeldii + Thalassiosira rotula.

). In the waters of the shelf, these were species of the genus Thalassiosira (T. gravida, T. rotula, T. nordenskioeldii, and T. hyalina). A bloom of a cryopelagic species (a large-size centric diatom, Porosira glacialis) in the MIZ was observed in the deep basin (station 6860). Spores of this species (130 spores L⁻¹) were found at a depth of 245 m (station 6862) in the AW flow (temperature 2.41 °C and salinity 34.92) (Table S5). The species mentioned above formed abundance and biomass maxima at the boundary between melted ice and PSW at a depth of halocline occurrence and high-lying nutricline (Tables S3 and S4). In 2017, in the western part of the studied area, at shelf station 5556, located among broken ice, a subsurface maximum of abundance and biomass at a depth of 22 m was formed by T. gravida, T. rotula, and T. nordenskioeldii. The cell abundance of this complex was very high and reached 6.1·10⁵ cells L⁻¹ (Table 5). In the eastern part of the studied area (station 5561), the deep maximum was formed by another species, T. hyalina.

The maximum phytoplankton abundance was observed in the deep water in the MIZ at a depth of 10 m at station 6860 in a layer of sharply gradient halocline due to the intensive growth of Porosira glacialis (Table 6). In the upper 10 m water layer, along with the diatom flora of multi-year ice (Melosira arctica, Fragilariopsis cylindrus, Fossula arctica (Fossulaphycus arcticus), and Nitzschia frigida), there were also Atlantic species—diatom Rhizosolenia hebetate var. semispina, Eucampia groenlandica, Pseudo-nitzschia delicatissima, and chrysophyta Dinobryon balticum. At another station (6861), located west of station 6860, Porosira glacial also dominated the phytoplankton (Table S5). However, its abundance was lower compared to station 6860. At this station, the phytoplankton community had not yet reached bloom abundance, and its composition and environmental ranges (temperature and salinity) were similar to those at station 6860. In addition, high concentrations of Si, N, and P were observed, providing potential conditions for future blooms.

In 2020, Eucampia groenlandica was found in the AWBC flow and the adjacent shelf waters (station 6854); however, in this case, this species was subdominant, and the dominant species in the AWBC flow was the haptophyta Phaeocystis pouchettii (Table 6). This species’ maximum abundance was higher than diatoms (E. groenlandica and Thalassiosira gravida). All maxima occurred at the same depth on the adjacent shelf (station 6854). The dominant phytoplankton species here were similar to station 6862. At another slope station 6856, located to the east, at the entrance to the Franz Victoria trench, Porosira glacialis was the dominant species of phytoplankton, with the maximum abundance noted at the lower boundary of the photic zone. The cold-water Monochrysis sp. was found in small abundance throughout the water column up to a depth of 100 m. The species dominating the main AWBC flow, E. groenlandica and P. pouchettii, were not noted in the phytoplankton community of station 6866, highlighting the unique characteristics of the phytoplankton community at different stations.

The absence of depth maxima of abundance diatoms in August 2020 in the Franz Victoria trench and the dominance of small flagellates and Monochrysis sp. in the community were recorded (Table S5). The growth of phytoplankton was limited by the complete absence of nitrates in the entire photic zone.

3.2.4. Biomass

In August 2017, the average wet-weight phytoplankton biomass in the northern part of the Barents Sea in the upper 50 m water layer varied widely from 8.0 to 2060.8 mg m⁻³ (

Table 7. Phytoplankton biomass (mg m⁻³) of the upper 50 m water layer (the upper numbers indicate the limits of fluctuations; the lower numbers represent the average wet-weight biomass) in August 2017.

Station	Total Biomass	Diatoms	Dinoflagellates	Coccolithophores	Chrysophyta	Small Flagellates
5553	73.9–256.5 165.2	0	72.8–241.7 157.3	0–0.06 0.03	0	0.99–13.9 7.44
5554	8.7–119.9 64.5	0–0.14 0.34	8.1–114.6 61.3	0–0.82 0.41	0	0.53–3.61 2.7
5556	4.6–4580.9 2318	3.8–4555.8 2299.3	0–25.6 15.7	0	0	0.8–4.47 3.22
5557	9.0–28.9 23.25	0–3.28 0.62	0.08–26.5 17.25	0–1.65 0.7	0	0.6–18.2 4.4
5557A	4.8–737.6 371.2	0.01–720.8 360.4	1.5–3.5 2.47	0	0	1.3–15.1 8.2
5558	9.1–4991.8 1879.3	0–4848.0 1818.0	5.9–21.4 11.5	0–0.29 0.1	0–3.7 0.8	1.2–122.1 47.4
5559	0.2–18.4 8.9	0–0.75 0.3	0–11.4 6.8	0	0	0.13–6.3 1.46
5560	2.6–14.5 9.4	0.01–0.2 0.1	1.2–9.3 5.8	0–0.82 0.2	0	1.2–5.1 3.3
5561	4.9–956.0 482.5	0.87–940.6 472.2	2.4–7.4 5.1	0	0	1.6–7.7 4.8

). The highest values were noted for the MIZ southeast of Svalbard (station 5556). Here, the biomass was formed by the neritic boreal-Arctic diatoms of the Thalassiosira complex. An equally high biomass (1879.3 mg m⁻³) was recorded in open water in the central part of the Franz Victoria Trench (station 5558). The Atlantic diatom Eucampia groenlandica contributed to the total biomass of phytoplankton here. The minimum average weighted biomass (<9.4 mg m⁻³) was typical for the northernmost station (station 5559) in the Franz Victoria Trench and for station 5560, off the western coast of the Franz Josef Land archipelago. In the studied northeastern Barents Sea areas, the community’s biomass was formed by dinoflagellates and small flagellates. The contribution of other systematic groups to the total biomass was lower.

In August 2020, the average wet-weight biomass in the northern part of the Barents Sea was lower than in 2017 (

Table 8. Phytoplankton biomass (mg m⁻³) of the upper 50 m water layer (the upper numbers indicate the limits of fluctuations; the lower numbers represent the average wet-weight biomass) in August 2020.

Station	Total Biomass	Diatoms	Dinoflagellates	Coccolithophores	Chrysophyta	Small Flagellates
6854	4.97–71.99 37.0	0–44.2 7.0	0.08–45.8 21.4	0	0–13.5 4.54	0.49–19.47 3.23
6856	2.9–109.7 46.0	0–105.0 11.7	1.4–32.7 10.0	0	0–32.1 10.0	0.23–7.78 4.1
6860	2.76–29,876.2 3792.0	2.0–29,855 3782.6	0–70.8 6.49	0	0	0.04–5.31 1.61
6861	1.6–21.6 9.52	1.54–28.3 8.47	0.01–1.69 0.33	0	0	0.11–1.19 0.7
6862	9.33–556.6 117.8	0.02–142.2 37.6	8.29–40.3 15.44	0–0.62 0.11	0–493.7 60.5	0.3–19.3 3.74
6864	3.26–34.23 18.7	0–0.80 0.17	0–28.22 4.01	0–6.17 0.35	0–6.18 1.8	1.3–21.7 9.5
6866	1.4–4.63 2.66	0.01–0.79 0.36	0.03–2.48 0.82	0	0–1.2 0.64	0.11–1.47 0.68

). The biomass was based on small flagellates and dinoflagellates. At slope stations 6854 and 6856, dinoflagellates were represented by both cold-water Arctic complex (Gymnodinium spp., Gyrodinium spp.) and Atlantic (Protoperidinium islandicum, P. grani, P. brevipes, and Prorocentrum minimum (P. cordatum)) complexes (Table S5). In the Franz Victoria Trench and on the central Barents Sea shelf, the community’s biomass was formed by dinoflagellates of the cold-water Arctic complex.

The highest value of average wet-weight biomass was observed in the deep basin (station 6860) due to the high biomass of diatom Porosira glacialis (30,792 mg m⁻³, 799 mg·C·m⁻³). Such high biomass occurred at depths of 10 m at temperatures <0, S = 33.2, and low Si and N concentrations (Table S7). The dominance index was approaching 100%. At station 6861, also located in the MIZ, the weighted average biomass of phytoplankton was only 9.5 mg m⁻³ with the dominance of the same diatom, P. glacialis.

In the main flow of AWBC, diatoms (E. groenlandica, T. gravida, T. rotula, T. nordenskioeldii, T. hyalina) and the haptophyta Phaeocystis pouchettii played a key role in forming the community’s biomass (Table 8). A bloom of Thalassiosira spp. was recorded at a depth of 22 m with temperature <0 °C, S = 34.4, and low Si and N concentrations (Table S6). A high biomass of T. hyalina was observed at a depth of 30 m at temperature <0, S = 34.5, and with relatively high concentrations of all nutrients (Table S7). The dominance index of diatom species was >90%.

The large diatoms Eucampia groenlandica had a high biomass at a depth of 45–52 m at temperature <0 °C, S = 34.5, and relatively lower Si concentrations (Table S7).

P. pouchettii’s contribution to the phytoplankton biomass was maximal under seasonal thermocline; relatively high biomass of this species has been recorded at depths of 38 m at temperatures <0 °C, S = 34.1, and high nutrient concentrations (Table S7). The dominance index was close to 62%.

The average wet-weight total phytoplankton biomass in MIZ (1501.5 mg m⁻³) and open water (104.8 mg m⁻³) for both years of research showed significant differences (p = 0.01).

4. Discussion

The ice cover’s interannual variability (2017 and 2020) was observed in the northern part of the Barents Sea and the Nansen Basin. In 2017, the Nansen Basin was covered entirely with ice; in 2020, open water was observed up to 83° N. The ice cover was similar in 2018; in 2019, the ice cover occupied entire areas of the Nansen Basin [87]. Sea ice plays a fundamental role in forming an ecosystem’s structural and functional organization [88], and it should be expected that the retreat of sea ice will lead to its transformation. Zones of sea ice, MIZ, open water, and features of the interaction of PSW and AW determine the multiplicity of structural changes in the phytoplankton community, as confirmed by our research.

4.1. Analysis of Water Masses

The retreat of the Arctic Sea ice and the replacement of multi-year ice with seasonal ice leads to a change in its physical properties. The main driver of such changes is the intensity of AW flow intake and their interaction with PSW, which creates a spatial heterogeneity of hydrophysical structures. When PSW and AW meet, they transform; some researchers attribute such water masses to the Barents Sea Water [43,89]. A two-layer structure is developed north of the polar front, with PSW at the top and AW below. The MIZ zone complements the picture, where a layer of meltwater with a characteristic halocline forms near the surface and a seasonal thermocline forms in an open water area. The complexity of the vertical structure and transformation of water masses in the flow creates classification difficulties [69,89,90,91].

The meandering of the AW flow, mainly north of Svalbard, significantly influences the position of the AWBC. The absence of warm AW found in the Franz Victoria Trench in 2020, which was present in 2017, had a noticeable impact on the structure and productivity of phytoplankton.

The intensive flow of meltwater leads to a thickening of the layer above the halocline [1]; the nutrient concentration above the halocline is low, and under the halocline, it is high (station 6860, Table S3). When the halocline was located inside a photic zone, intensive phytoplankton growth reaching the bloom level developed on the halocline, as in station 6860. Changes in the layer thickness above the halocline are relatively mobile [28]. This vertical dynamic of the halocline modulates the irradiance from a limiting value to inhibiting phytoplankton growth and, as a result, regulates the phytoplankton biomass.

4.2. Hydrochemistry

Different water masses are saturated with nutrients at different values; AW is rich in Si, N, and P; the concentration of these nutrients in PSW is significantly lower, and the concentration of Si and N in meltwater is close to zero (Table S3). The formation of a seasonal thermocline leads to the depletion of Si and N in the UML in the first place, which was a consequence of the spring phytoplankton bloom [92,93]. In this case, the ecosystem moves from a light-limited ecosystem, observed under the ice, to one limited by Si and N in open water [94]. The interannual variability of the intensity of the AWBC flow regulates not only hydrography but also the nutrient flux. This effect is not only due to their advection [95] but also due to vertical transport resulting from convective mixing [12]. Our research has shown that in 2017, the N concentration was not zero; in 2020, it was often close to zero (Tables S3 and S4). N availability appears to be the main growth-limiting factor in the photic zone, and this nutrient in the High Arctic determines productivity [93,96]. However, at some stations in 2017, the silicon concentration at the thermocline boundary remained very low and insufficient for the growth of diatoms [97,98]. This may explain the shift in the structure of phytoplankton towards small flagellates. Phosphorus concentrations have always been higher relative to silicon and nitrogen, likely explained by the increased export of this nutrient from AW [99].

4.3. Analysis of the Phytoplankton Composition, Their Abundance, and Biomass, as Well as the Ecological and Geographical Characteristics of the Species

The seasonal changes in ice edge movements to the north significantly contribute to the ecosystem’s temporary heterogeneity. The MIZ plays a primary role in forming the structure and productivity of the autotrophic part of its pelagic ecosystem, where phytoplankton blooms occur in the spring period [58,59]. However, the relationship between the time of sea ice retreat and the peak of phytoplankton biomass is nonlinear [100]. This phenomenon makes it possible to explain why in situ observations do not always detect phytoplankton blooms.

In high latitudes, the spring bloom period shifts to July–August [59,87,101]. Our mid-August 2017 and 2020 observations allowed us to estimate the phytoplankton growth in the shelf zone, the AWBC flow, and the Nansen Basin. In all the biotopes mentioned above, conditions for the bloom developed in August (Table S7). In the northern part of the Barents Sea, in addition to MIZ, phytoplankton grew intensively in open water under certain circumstances. In the summer of 2017, such a phenomenon was observed in the Franz Victoria Trench. Phytoplankton blooms in the central and southern parts of the trough were produced by the Atlantic diatom Eucampia groenlandica. The high biomass of this species is registered practically at the lower boundary of the photic zone and at the upper boundary of the nutricline. In the MIZ and other areas of the northern Barents Sea shelf, the dominant bloom species were diatoms associated with first-year ice from the genus Thalassiosira (T. gravida, T. rotula, T. nordenskioeldii, and T. hyalina). Species of the genus Thalassiosira developed spring diatom blooms in open water in the North Atlantic and in the Barents Sea, south of the polar front, particularly in the waters of the warm and salty AW current [102,103]. Large diatoms, Eucampia groenlandica, were also registered in this water area; thus, the impact of AW on this species’ blooms is indisputable.

AWBC, which is mainly directed to the east along the edge of the northern shelf of Svalbard, later deviates to the northeast, towards the zone of multi-year pack ice [16]. At the same time, the individual flow of this current penetrates south into the Barents Sea through a system of periglacial trenches [85], such as the Franz Victoria Trench between Svalbard and the Franz Josef Land archipelago. Our results have established that in different years, the structure of phytoplankton in the trench showed sharp differences. At stations located at the same site, Eucampia groenlandica bloomed in mid-August 2017; in August 2020, the bloom was missing. In this context, two questions had to be answered to understand the reasons for the interannual differences: (i) the relationship between AWBC and the E. groenlandica bloom in the Franz Victoria Trench in 2017 and (ii) the reason for the absence of the bloom in 2020.

Our analysis of the phytoplankton composition allowed us to find that the stations located in the Franz Victoria Trench (station 5558, 2017), in the AWBC zone (station 6854, 2020), and in the deep water (station 6860, 2020) are similar (Table S5). The same species characterized phytoplankton in the AWBC and the Franz Victoria Trench; Eucampia groenlandica and P. pouchettii were the dominant species. These species are characteristic of AW, and P. pouchettii forms blooms in the polar waters of the Atlantic [104]. In August 2017, the diatom E. groenlandica was the dominant species, and the haptophyte P. pouchettii was the subdominant. In contrast, in August 2020, P. pouchettii was the dominant species, and E. groenlandica was the subdominant, indicating a later stage of succession (Table 5 and Table 6). Another possible explanation for the dominance may be the interannual differences in the light condition; the haptophyte is better adapted to low irradiance because it has a higher light absorption coefficient [86].

Observations of Eucampia groenlandica bloom in the trench have shown that the phytoplankton composition is subject to very high variability over a relatively short time. Thus, at stations 5557 and 5557-A, located in the same site, a radical transformation of the phytoplankton took place within three days—the phytoplankton community corresponding to cold Arctic water changed to the community characterized by AW origin (Figure 3; Table 1), in which E. groenlandica diatom bloomed (Table S7). It should be noted that such high abundances of large diatoms, as Porosira glacialis and E. groenlandica, in the High Arctic have not been recorded before. Such intensive growth can be explained using some hypotheses. The first “light” hypothesis states that large diatoms can change their specific light absorption coefficient due to the movement of chloroplasts [105]. This mechanism of adaptation to light allows them to successfully acclimate to diurnal changes in light intensity and vertical migrations with unstable water columns. The second “storage” hypothesis is based on the fact that large diatoms have a competitive advantage over small cells due to their ability to accumulate nutrients in vacuoles and use these reserves when they are scarce in the environment [106,107]. Therefore, large cells dominate in periodically changing conditions, which is possible with variable vertical structure.

In this area, diatoms play a crucial role in carbon exports in May–June, and in August, small flagellates take over the export flux [108]. Our research has revealed that in addition to the small flagellates that dominate the seasonal thermocline layer, there are certain areas where the growth of large diatoms was found in August.

The dominant species on station 6854 in the AWBC flow was Eucampia groenlandica. The same species was part of the diatom complex in the MIZ in the deep-water station 6860 and was found in the upper 10 m layer of meltwater during the Porosira glacialis bloom, further evidence of the AWBC’s role in the bloom formation in the MIZ in the Nansen Basin.

In August 2020, no phytoplankton blooms were observed in the Franz Victoria Trench (station 6866) (Table 8). Against low abundance and biomass background, dinoflagellates of the cold-water Arctic complex and the chrysophyte Monochrysis sp. dominated the community. According to our observations, the latter is typical for the cold Arctic waters of the North Atlantic and the Nansen Basin. Monochrysis sp. was also the dominant species at the entrance to the Franz Victoria trench (station 6856), along with the diatom Porosira glacialis. Such phytoplankton composition, which differs sharply from the AWBC community, was formed due to the inflow of low-salinity surface water from the ice melting zone in the north. Since 2020 was abnormally warm in the Arctic over the entire observation period, the multi-year ice boundary was shifted far north above 83° N (Figure 1). The increased air temperature activated the process of ice melting in combination with the effects of AWBC [7]. Moving south, the meltwater covered the PSW, and the sharp pycnocline prevented nutrient penetration into the UML, limiting phytoplankton growth.

High AWBC activity was observed in the deep basin during the same period. The bloom of diatom Porosira glacialis at 83° N was a direct result of AWBC exposure [28]. The analysis of the distribution of Atlantic oceanic indicator species in the studied water area further supports this. The highest number of Rhizosolenia and Chaetoceros species was noted in the Nansen Basin, reinforcing the impact of AWBC. These species were practically absent on the shelf and in the Franz Victoria Trench. This fact may also support the assumption of AWBC meandering. During its high activity in 2017, the AWBC deviated towards the Barents Sea shelf, resulting in the bloom of the diatom Eucampia groenlandica in the Franz Victoria Trench. On the contrary, in 2020, AWBC manifested itself most vividly to the north, causing a bloom in the MIZ at 83° N.

The AWBC structure in the Nansen Basin remains a subject of ongoing research. The current has two branches: the southern Svalbard and the northern Ermak Plateau [109,110]. While some reports suggest that these branches merge, the exact location of this convergence is yet to be determined. Our data indicate that the structure of the diatom component at station 6861, located in the MIZ to the west of station 6860, differs from other areas of the deep basin we have studied. For instance, two oceanic Atlantic species from the genus Chaetoceros (C. atlanticus, C. danicus), the cosmopolitan dinoflagellate Prorocentrum micans, and the cold-water coccolithophorid Coccolithus pelagicus, typically found in the Greenland Sea [111], were only found at station 6861. This fact suggests that in August 2020, the northern branch of the AW extended up to station 6861; this branch passes through the Ermak Plateau and follows east, providing deep convection [112].

The few data available in the literature on the structure of the summer phytoplankton community of the northern shelf of the Barents Sea relate mainly to the Orly Trough (Kvitøya Trough) [19,68,69]. The results of these studies have highlighted the critical role of AWBC in forming a summer phytoplankton community in this area. Here, at 0–50 m depth, Atlantic species of the genera Chaetoceros and Rhizosolenia and coccolithophorids Emiliania huxleyi, with an abundance of 5.0 × 10⁴ cells L⁻¹, were found. In the summer, a high degree of interannual variability of the phytoplankton structure in the same area located at the entrance to the trench was observed as a change in the dominant species. From July–August 1984 to 2003, the community was dominated by cold-water Thalassiosira spp. and species of the Atlantic complex, Phaeocystis pouchettii and Chaetoceros gelidis. In August 2003, the 20 m surface water layer was already characterized by indicator species of warm AW coccolithophorid, Emiliania (Gephyrocapsa) huxleyi, which indicated the intensification of AWBC.

As for the bloom of the neritic boreal-Arctic diatom, Porosira glacialis, in the MIZ in the Nansen Basin, in our opinion, the presence of spores of this species at a depth of the AW core both in the AWBC and in the deep water may indicate the ways of transporting this species to the High Arctic. Further, getting into the surface layers with convection flux from the depth, the spores of this algae find favorable conditions for their growth on the lower surface of the first-year ice solder formed in spring at the edge of multi-year ice [22]. When first-year ice melts in summer, P. glacialis enters the subglacial layer of low-salinity water on the halocline, where its intensive growth occurs. The presence of meltwater is a necessary condition for the bloom development in the MIZ [113].

5. Conclusions

• The ice cover in the western Arctic tends to decrease and has interannual variability. In the summer of 2017, the north of the Barents Sea was ice-free, and the Nansen Basin was covered with ice; in 2020, open water was recorded in the Nansen Basin, up to 83° N. Our analysis of water masses north of the Barents Sea and the Nansen Basin showed that the vertical structure of the water masses is as follows: the PSW, poor in nutrients, is located on the top, and the AW, rich in nitrogen and silicon, is located below. A layer of meltwater is added to the MIZ from above.
• We note that MIZ is an order of magnitude more productive than open water. In the MIZ zone of the northern shelf of the Barents Sea in 2017, a high biomass of phytoplankton was formed by cold-water diatoms Thalassiosira spp. (T. gravida, T. rotula, T. hyalina, T. nordenskioeldii) associated with first-year ice. A distinctive feature of deep water in the high Arctic in summer was the bloom of large diatoms, Porosira glacialis, registered on the halocline between meltwater and PSW in the MIZ of the deep basin (2020).
• A seasonal thermocline characterized open water. In the UML, the concentrations of silicon and nitrogen were close to zero. Deep maxima of phytoplankton abundance and biomass were recorded here under seasonal thermocline; Atlantic species—large diatom Eucampia groenlandica and haptophyte Phaeocystis pouchettii—formed these maxima and were observed in the area of action of the Atlantic Water Boundary Current (AWBC) in the Nansen Basin (2020) and the Franz Victoria trench (2017). The high biomass of E. groenlandica (4848 mg m⁻³, 179.5 mg C m⁻³) was found on the nutricline boundary at a depth of 45–52 m and had not been observed before in the High Arctic.
• The high phytoplankton biomass in the Franz Victoria trench and the Nansen Basin is related to the intensity of the AW, which comes from Frame Strait as the Atlantic Water Boundary Current. Such a high biomass of large diatoms might suggest that the high Arctic can be very productive when ice retreats.
• These results are essential in understanding the Arctic ecosystem’s dynamics during climate change. We predict that the MIZ will be the most productive zone when the ice edge moves north. At the same time, large diatoms can grow intensively in the high Arctic, significantly contributing to biomass formation. A necessary condition is the presence of warm and nutrient-rich AW.