Sinking Particle Fluxes at the Jan Mayen Hydrothermal Vent Field Area from Short-Term Sediment Traps

1. Introduction

Hydrothermal vent fields are defined as areas where highly mineralized and gas-saturated waters are discharged at mid-ocean ridges, volcanic arcs, and back-arc spreading centers [1,2]. These fluids are enriched with elements such as S, Si, Fe, Mn and other base metals (Cu, Zn, Pb, etc.) compared to seawater. Physical and chemical parameters of the environment change drastically in the discharge zone, resulting in development of a geochemical barrier where local accumulation of hydrothermally derived materials takes place, e.g., [1,2,3,4]. The study of hydrothermal plumes is necessary for the assessment of the ocean's chemical budget.

High temperature vents of solutions and gases are the most spectacular manifestations of hydrothermal activity. So-called black and white smokers produce particle-rich plumes that rise hundreds of meters above the seafloor [1,3,6,7,8,9,10,11,12,13]. The rising plume captures surrounding seawater, leading to the formation of metal sulfides and sulfates through oxidation. These particles, as a part of buoyancy and neutral buoyancy hydrothermal plumes, are carried by deep-sea currents and can be detected up to tens of kilometers from the emitting source, e.g., [13,14]. Buoyant plume turbulence causes transport and mixing that produce the neutrally buoyant plume fluids, thereby affecting the ocean over a wide range of temporal and spatial scales [16]. The mineral and chemical composition of the initial plume particles have been investigated in numerous studies, e.g., [6,7,10,17,18,19,20,21,22]. Hydrothermal systems are both an important source of some elements (e.g. Fe, Mn, Zn, Cu, etc.) that may be involved in biogeochemical processes in the pelagic regions of the ocean, and a precipitator of other elements and components (e.g. Mg, ${\mathrm{SO}}_4^{-}$, REEs, etc.) from seawater [5,23,24,25,26]. The study of particle fluxes sinking from plumes into the adjacent water column using sediment traps is of particular interest, as marine particles play a key role in the cycling of elements in the ocean. This type of research was initiated almost immediately after the discovery of the active hydrothermal systems in the ocean in 1977 [13,15,18,19,22,27,28,29,30,31,32,33,34].

Hydrothermal venting sites were discovered in the shallow southernmost part of the Mohns Ridge segment of the Arctic Mid-Ocean Ridge (AMOR) during the BIODEEP-05 cruise of the Norwegian research vessel G.O. Sars in 2005 [37]. All studied fields are collectively referred to as the Jan Mayen vent fields (JMVFs) [38]. Various components of the hydrothermal system were studied, such as topography, composition of venting structures, fluids and precipitates, dissolved gases, biodiversity of hydrothermal fauna, microbial communities, etc. [5,37,38,39,40,41]. To address the lack of information on the sinking particle fluxes at the Arctic hydrothermal vents we carried out first study of sinking material at the JMVFs using sediment traps (Figure 1). The present study discusses data on the vertical particle fluxes of two active vent sites at the Jan Mayen axial volcanic ridge (AVR) located inside the rift valley [42] in order to estimate the fluxes of hydrothermally derived particles to the seafloor, considering the lateral component and biogenic matter produced in the euphotic zone. We also discuss the trap fluxes in the context of similar available observations from other active hydrothermal systems. Therefore, here we report detailed geochemical studies of sediment trap material from the JMVFs at 71° N and 6° W of the southernmost Mohns Ridge, the sector of the northern Mid-Atlantic Ridge extending beyond the Arctic Circle latitude.

2. Materials and Methods

2.1. Study Area

The Mohns Ridge is a segment of the active ultraslow spreading AMOR between 71.2° N and 73.5° N with 1.6 cm/yr full spreading rate [43]. The southernmost segment of the Mohns Ridge is affected by the Iceland and Jan Mayen hotspots and is characterized by anomalously high magmatic activity and shallow water depths (700–550 m) compared to other sections of the ridge [39,44,45,46]. The characteristic structural elements of the Mohns Ridge are AVRs, which are confined to the rift valley [47,48,49,50]. Rift/volcanic structures with hydrothermal activity, such as the JMVFs, may occur within the AVR creating hydroacoustic anomalies above these fields due to gas/fluid release [39]. The Jan Mayen AVR at 71°18′ N is the largest and is located near the Jan Mayen hotspot, which is one of the most active areas of submarine neo-volcanism in the ocean. Its elevation above the rift valley floor is 2,100 m. The morphology of the Jan Mayen AVR is very different from that of the other AVRs in the Mohns Ridge. It is also separated from the Jan Mayen Rise by the Jan Mayen Fracture Zone that runs in a north-west to south-east direction (Figure 1 and Figure 2). From the northeast, the Jan Mayen AVR is also bounded by a transform fault. The structure of AVR is complicated by a series of faults and vertical displacements of the dumping type. There are three active vent fields and an one inactive have been discovered and studied at the Jan Mayen AVR summit [39]. Active hydrothermal vent fields include Troll Wall (TW) and Soria Moria (SM) discovered in 2005, and Perl & Bruce explored more recently in 2013; inactive vents include Gallionella Garden [5,39,40].

The bottom-tethered moorings were deployed at TW and SM hydrothermal vent fields (Figure 2), referred to as “TW mooring” and “SM mooring”. The TW vent field is located at a depth of 550 meters. It is confined to the fault bounding the eastern wall of the rift valley, which transects the large central volcano Frøya [39]. There are at least 10 major hydrothermal vents associated with several 5–10 m high white smoker chimneys within the TW vent field. Maximum fluid temperature was measured at 270°C [38,39].

The SM vent field is located 4.3 km to the south of the TW at a depth of 700 m, and is confined to the top of one of the small rifts [39]. The ridges extend in a quasi-meridional direction and are bounded to the east by the north-east-trending rift mountains. The height of the ridges does not exceed 110 m and their width is 250–300 m. At the SM vent field maximum vent fluid temperature reaches ~270°C as at the TW. However, hydrothermal fluids are emitted from two very different types of chimney. White smoker fluids emerge turbulently from numerous sulfide chimneys up to 8 to 9 m in height., while lower temperature clear fluids escape from irregular constructions made of barite, silica, and minor amounts of pyrite, sphalerite and galena [38,39].

The hydrography of the study area is described in detail in [35,36]. It is influenced by colder and less salty Arctic-derived waters flowing from the east coast of Greenland as the Jan Mayen Current, and warmer and saltier Atlantic-derived waters supply from the Norwegian Basin as a branch of the Norwegian Atlantic Current (the northern limb of the Gulf Stream system).

2.2. Field Observations and Sample Collection

Two short-term mooring systems with sediment traps were deployed and recovered during the 75^th cruise of the RV Akademik Mstislav Keldysh in June 2019 [51,52,53]. The deployment design consisted of a suite of measurement tools operating at different depths (Figure 3). Cylindrical sediment traps MSL-110 (IORAS, Moscow, Russia) [54] were deployed at three nominal depths of 30, 100, 200 and 400 m above the seafloor according to the locations of neutral buoyancy plumes reported in [55]. Autonomous platforms provided continuous sampling with periods of 101.7 and 95.5 hours at the TW and SM vent sites, respectively. Trap locations, depths, duration, sinking particle fluxes, and water depth of each deployment are listed in Table 1.

Quadrupled cylindrical MSL-110 sediment traps with inner diameter of each tube of 110 mm, i.e., four tubes per layer with a total collection area of 0.038 m², were deployed at each trap depth of the moorings. Ratio between diameter and collecting tube length is 1:6 [54]. The MSL-110 sediment trap design has previously been successfully utilized in the study of sinking particle fluxes in the Arctic Seas and the subpolar North Atlantic [56,57,58,59]. Collecting cups of the traps were filled with 0.45 µm filtered intermediate seawater made up to a salinity of 40 psu with NaCl. No poison was applied to the collecting cups during deployment.

Nortek Aquadopp Current Meter (Nortek AS, Rud, Norway) and SBE37SMP-ODO CTD-profiler (Sea-Bird Electronics, Inc., Bellevue, WA, USA) equipped with an SBE63 Optical Dissolved Oxygen Sensor (Sea-Bird Electronics, Inc., Bellevue, WA, USA) were deployed at 1 m below the 520-m trap depth at the TW-mooring to measure relative water motions and hydrographic parameters. RBRduet T.D loggers (RBR Ltd., 95 Hines Road, Ottawa, Ontario, Canada) were deployed immediately below the 570-m and 640-m trap depths at the SM-mooring to measure hydrographic data.

In order to document the water column properties, CTD profiling with Seabird SBE911plus (Sea-Bird Electronics, Inc., Bellevue, WA, USA) was carried out at two stations before deployment and after recovery of the moorings. To assess the concentration of suspended particulate matter (SPM), particulate organic carbon (POC) and chlorophyll a (Chl-a), water samples were collected using a Niskin bottles (OceanTest Equipment, Fort Lauderdale, Florida, USA) equipped sampling system SBE 32 (Sea-Bird Electronics, Inc., Bellevue, WA, USA). Samples were collected from the subsurface, chlorophyll fluorescence maximum, and near bottom layers according to CTD profiles, and were then divided into subsamples to determine total SPM, POC, and Chl a concentrations.

2.3. Processing and Analysis of Samples

The material collected by the traps, as well as water samples from Niskin bottles, were filtered under vacuum through pre-weighed nuclear pore membrane filters with a diameter of 47 mm and a pore size of 0.45 µm. Salts were eliminated by rinsing the filtered samples several times with MilliQ water at the end of filtration. The filters were dried at 50°C, and then weighed using a Sartorius analytical electronic balance with an accuracy of ±0.01 mg. The deposited mass was calculated from the difference between the final and initial weights of the filter. The total particle flux was obtained by dividing the deposited mass by the sediment trap collection area and collection time. The mass concentration of the SPM was obtained by dividing the deposited mass by the volume of the filtered seawater.

Subsamples for the determination of total particulate carbon (TC), particulate organic carbon (POC) and particulate inorganic carbon (PIC) and chlorophyll a (Chl-a) were filtered under a low vacuum onto Whatman GF/F glass microfiber filters of 47 mm diameter and a pore size of 0.7 µm precombusted at 500°C for 4 h.

Chl-a concentrations were determined fluorometrically according to [60,61]. Whatman GF/F glass microfibre filters with trapped material were extracted in 90% acetone (at 2°C in the dark for 24 hr). The fluorescence of the extracts was measured with a Trilogy laboratory fluorometer version 1.1 (Turner Designs, San Jose, CA, USA) before and after acidification with 1 N HCl. The accuracy of determination was ± 0.02 mg/m³.

Total particulate Si and Al were determined using the photometric method with a UNIKO 1201 device (Russia) according to [62]. The sample was fused previously in a muffle furnace (at a temperature of 950°C for 4 h) with a mixture of Na₂CO₃ and Na₂B₄O₇ and the addition of KNO₃. The fusion was dissolved in 3M HCl. Measurement of dissolved silicon concentration involves molybdate-blue spectrophotometry of the extract at a wavelength of 812, 815 nm. The colorimetric determination of Al is based on the ability of Al to form a red-colored complex with aluminon. Starch is used to preserve the red-colored complex as a colloidal suspension. The optical density of the solution is measured spectrophotometrically at a wavelength of 525 nm. The standard reference matter (SRM) of ocean sediments (OOPE-101 and OOPE-401, Russia) was used to assess the quality of the analysis. The discrepancies with the SRM passport values of the contents of the measured elements in SRM were 2%–5%. The content of lithogenic matter (LM) was calculated by multiplying the Al by a factor of 12.27 based on the average aluminum content in the upper continental crust (UCC) [63]. The silicon content exceeding the lithogenic content estimated by the Si/Al ratio in the UCC [63] was considered as exSi. The Si enrichment of SPM relative to lithogenic matter is usually explained by the presence of biogenic amorphous silica (bSiO₂) [64].

Particulate total carbon (PTC) and PIC were measured with a Shimadzu TOC-L analyzer with SSM-5000A module (Shimadzu Corporation, Japan). IC was measured directly by coulometry, which is the measurement of CO₂ following closed-system conversion of IC to CO₂ upon addition of phosphoric acid [65]. POC was determined by the difference between PTC and PIC. The analytical accuracy was 1.0 wt.%. From the PIC content, the particulate calcium carbonate CaCO3 was calculated by multiplying 8.33 (according to stoichiometric coefficients and atomic weights).

The elemental analysis of sinking material was carried out by inductively coupled plasma mass spectrometry (ICP-MS) using the Agilent 7500a quadrupole instrument (Agilent Technologies, Inc., Santa Clara, CA, USA). Filters with the weighed matter were placed in Teflon (Savillex™) vials, filled with a mixture of nitric, hydrofluoric and hydrogen peroxide ultrapure concentrated acids, placed in an ultrasonic bath for 90 min at a temperature of 70°C, and then were remained overnight. After that, the filters were removed from the vials and washed with a small amount of 0.1 N nitric acid. Then, with the addition of hydrochloric acid, the samples were evaporated three times on a hotplate. After the last evaporation the samples were brought to a volume of 35 mL with deionized water. A multi-element standard curve containing 10 ppb of indium was used to calculate trace element concentrations after the subtraction of the instrument blank. Digest blanks were subtracted from the sample and filter blanks were adjusted for a blank derived from the digestion acids and Teflon vials. The overall analytical precision was 5%–10% based on five replicate analyses of a single sample digest solution.

To assess the sources of material input, elemental concentrations were normalized to the average crustal composition, and enrichment factors (EFs) relative to the average UCC composition [63] were calculated using Eq.:

EF = (El/Al)_sample/(El/Al)_UCC,

(1)

where (El/Al) is the ratio of element to Al content in the sample or UCC. EF values close to 1 (less than 3) indicate a predominantly lithogenic origin of the material, whereas EF values greater than 3 indicate an additional source.

Based on the Fe and Al content data, the excess iron Fe_ex content was calculated:

Fe_ex = Fe_measured – Al_measured*(Fe/Al)_UCC,

(2)

where Fe_measured and Al_measured are the measured Fe and Al contents in the sample, (Fe/Al)_UCC is the ratio of these elements in the Earth's crust.

The cerium and europium anomalies (Ce_an and Eu_an respectively) in the samples were calculated using the following equations [66]:

Ce_an = Ceⁿ · 2 / (Laⁿ + Prⁿ),

(3)

Eu_an = Euⁿ · 2 / (Smⁿ + Gdⁿ),

(4)

where Elⁿ is the REE content in sample normalized to the content in NASC shale [67].

The ratio of light to heavy REEs was calculated using the formula:

LREE/HREE = (La^N + 2*Pr^N+Nd^N)/(Er^N+Tm^N+Yb^N+Lu^N),

(5)

This formula avoids biases associated with anomalous accumulation of REEs of variable valency [68]. All REEs have been normalized to the NASC shale composition.

Sinking material was examined using a VEGA-3sem TESCAN (Czech Republic) scanning electron microscope (SEM) with an energy dispersive X-ray spectrometer (EDS), Oxford INCA Energy 350 (UK). Mineral composition of individual particles was determined from the chemical composition based on stoichiometric ratios and atomic weights.

2.4. Satellite Research

To study the bio-optical characteristics of the sea surface water, MODIS-Aqua ocean color scanner data were retrieved from NOAA website (https://oceancolor.gsfc.nasa.gov/, accessed on 28 April 2022). 8-day composite Level 3 Mapped Chlorophyll Data at moorings locations with spatial resolution of 4×4 km was averaged. Satellite data were processed using SeaDAS 8.2.0 software (https://seadas.gsfc.nasa.gov/, accessed on 04 April 2022).

3. Results

3.1. Environmental Conditions at the Mooring Sites

The temperature, salinity and turbidity profiles were obtained before the deployment and after the retrieval of the moorings (see Figure 4). Most of the water column (from the bottom to ~140 m) was occupied by cold Arctic intermediate water (temperature -0.26–0.3°C, salinity 34.88–34.91 psu), and only the upper quasi-homogeneous layer 0–60 m warmed to 2–3°C (Figure 4). The background temperature in the deep water layers (up to 100 m from the bottom) varied from –0.13°C to 0.12°C and from –0.26°C to 0.13°C at the TW vent field and the SM vent field, respectively. CTD measurements during the deployment and retrieval of moorings revealed positive temperature anomalies near the bottom with an amplitude of 0.64°C in the TW and 2.49°C in the SM (Figure 4). Positive temperature anomalies were recorded in the 40–90 m layer from the bottom in the vicinity of the TW field and were maximally confined to the bottom (0–15 m) in the SM field.

An acoustic current meter data obtained from the TW-mooring (Figure 5) revealed fluctuations in current velocity ranging from 0.3 to 18.7 cm/s, with an average velocity of 6.1 cm/s. Regular variability in current speed and direction with a semi-diurnal cycle was observed, indicating tidal influence. The accumulated water mass transport (vector sum) during deployment time of the TW-mooring amounted to 8092 m. Based on this value, the average vector velocity of lateral particle transport is calculated to be 2.2 cm/s. At the TW mooring, several pronounced short-term above zero temperature anomalies with amplitude of up to 0.86°C were recorded in the near-bottom layer. The minima of the dissolved oxygen concentrations and the maxima of the sound backscattering by particles, obtained with the acoustic Doppler current meter, are associated with the temperature anomalies.

T.D-loggers in the near-bottom layer of the SM vent field also recorded above zero temperature anomalies, but with a much smaller amplitude (~0.25°C). The temperature anomalies were observed only once a day and were related to the tidal phase during the day (Figure 5). No similar temperature anomalies were detected 70 m higher, at the next level from the bottom (570 m).

In the SM vent field, a cold layer with a temperature of -0.26°C, which is 0.13°C below the background value, was observed near the bottom by CTD profiling prior to mooring deployment. The layer was approximately 50 meters thick. CTD-profiling after mooring recovery did not detect this layer. Fluctuations of water temperature of the same amplitude (up to 0.13°C) were also recorded by the sub-bottom T.D loggers. The period of variation was 1 day. This is characteristic of the diurnal component of the tides.

3.2. Suspended Particulate Matter and Chlorophyll A

SPM concentrations in the water column prior to mooring deployment ranged from a relatively high values of 0.23–0.30 mg/l in the upper (0–90 m) layer to a low value of 0.03–0.09 mg/l in deeper waters (see Figure 4). Upper mixed layer (0–25 m) was characterized by high Chl-a concentrations of 1.36–1.92 mg/m³ at the mooring sites and up to 5.4 mg/m³ in the adjacent area (30 miles to the west in the Arctic Front). The Chl-a concentrations exhibit a subsurface maxima at the depth of 25–35 m. With increasing depth to 100 m, the concentrations of Chl-a gradually decreased to 0.2 mg/m³ and below 100 m reached a minimum value (≤0.01 mg/m³).

3.3. Mass Fluxes and Major Element Composition

3.3.1. Total Particle Fluxes

The maximum vertical particle flux (279.6 mg/m²/day) occurs in the 150 m subsurface layer at the TW vent field (Figure 4, Table 1). Deeper down, from 270 m to the bottom, fluxes ranged from 85 to 130 mg/m²/day. Minimum fluxes (85–97 mg/m²/day) were recorded in the 450–570 m intermediate layer at both sites. Particle fluxes in the near-bottom layer (30 m from the bottom) were found to increase to 103.6 mg/m²/day and 103.9 mg/m²/day at the TW vent field and SM vent field, respectively. This may be an indication of both the hydrothermal supply and the possible influence of a benthic nepheloid layer. The study of the distribution of SPM, both by membrane filtration and by turbidity sensor at the CTD profiler, did not reveal the presence of a pronounced nepheloid layer near the bottom in the study area. This was also found in the work of M.D. Kravchishina et al. [41].

3.3.2. Major Phase Composition of Particles Including POС, PIC (CaCO₃), bSiO₂, and Lithogenic Matter

The sedimentary matter is dominated by biogenic components including CaCO₃ + POC + bSiO₂. These components reached ~99% in the upper layer (150 m) of the water column above the TW vent field (Table 1). The POC ranges from 35% to 56%. The maximum is found in the subsurface layers. High content of bSiO₂ (18.8–28.7%) was observed almost constantly throughout the water column, and maximum fluxes of 52.6 mg/m²/day and 37.5 mg/m²/day localized in the 150 m (TW mooring) and 270 m (SM mooring) subsurface layers, respectively. Subsurface water is also responsible for the maximum CaCO₃ flux. The flux of lithogenic particles increased below the euphotic zone and remains almost constant (13.6–19.7 mg/m²/day) from the layer of 270 m to the bottom at both vent fields, accounting for 12.8–21.8% of the sinking matter.

3.3.3. Electron Microprobe Analysis of Particles

In the studied water column, with the exception of the bottom layers of both fields, biogenic material (diatoms, coccolithophorids, foraminiferal tests fragments) and particles of rock-forming minerals of pelitic and fine silty grain size (feldspars, quartz, augite, andalusite, clay minerals, etc.) prevailed. Hydrothermally derived micron-sized minerals such as sulfides, sulfates, amorphous silica, etc., were abundant in the benthic layers of both vent fields.

3.3.3.1. Troll Wall

At a depth of 350 m, along with biogenic particles, a significant amount of mineral particles ranging in size from 2 µm to 20 µm were present. Particles finer than 2 µm often form isometric aggregates ≥50 µm in size, dominated by aluminosilicate clay particles mixed with algal skeletal fragments (Figure S1a). In addition to clay particles in the aggregates, EDS data revealed many unrounded grains of rock-forming minerals: forsterite, diopside, anorthite, K-feldspar, labrador, pseudorutile and titanomagnetite (Table S1). Deeper down, at a depth of 450 m, clay particles (up to 5 μm), organomineral aggregates (up to 20 μm) and feldspars (15–25 μm) were found along with forsterite and feldspar particles of irregular isometric shape with fresh fractures. An abundance of biogenic detritus and clay particles 5–10 µm in size was also observed in the bottom layer at 520 m. Of the main minerals that form the rock, particles of diopside and augite were found in this layer. The bottom layer contained a large amount of hydrothermally derived minerals:

• Gypsum microcrystallites up to ~140 µm in size formed by free growth is characterized by a columnar habitus with well-formed, flattened, elongated prismatic faces showing dissolution shapes (Figure S1b). The chemical composition of the gypsum is close to the stoichiometric composition and no significant concentrations of isomorphic impurities have been found in the gypsum (Table S2).
• Barite aggregates ~120 µm in size have lamellar and tabular habitus (Figure S1c). Flat barite crystals >10 µm with a crosscutting tabular shape of hexagonal type is identified in the aggregates. The tabular crystals form well-shaped rosette structures at both vent sites. According to the EDS data (Table S2), the barites demonstrate minor substitution of barium by strontium and calcium. Barite aggregates form a stable mineral association with sulfides.
• Iron, copper and zinc sulfide minerals form porous as bud-shaped masses with individual spherulites ≤0.5 µm in size, often of complex chemical composition with impurities (Figure S1d). Only pyrite forms regular octahedrons and cuboctahedrons up to 3 µm in size.
• Aggregates of non-crystalline Fe-Si oxyhydroxides of separate thin filaments of 0.5–1 µm diameter of isometric, sheaf and whorl shapes characteristic of the quenching process under conditions of high temperature difference (Figure S1e). Microspheres up to 0.5 µm are embedded in the filaments of the whorls. The EDS data indicates that the SiO₂ content of the aggregates ranges from 27 to 41 wt% and that they exhibit a significant iron impurity (38-45 wt%) as well as calcium (Table S3).

3.3.3.2. Soria Moria

The particle composition at depths of 270–570 m is notably similar to that of the upper and intermediate water layers (150-450 m) of the TW vent field. The prevailing constituents were biogenic matter, clay minerals, and particles of rock-forming minerals with a size of 10–15 µm. These included feldspars, quartz, augite, and andalusite. The latter are frequently characterized by an unrolled surface displaying fresh chipping or cleavage patterns (Figure S1f, Table S4). The particle composition of the 640 m near-bottom layer appears to be similar to that of the 520 m near-bottom layer of the TW field. In addition, biogenic particles and organo-mineral aggregates were also observed to be predominant. Among the mineral particles, clay minerals, single particles of rock-forming minerals and hydrothermally derived minerals were identified:

• Barite aggregates up to 50 µm in size with a lamellar and tabular habitus (Figure S1 g), and minor substitution of barium by strontium (Table S5);
• Sulfide minerals, represented by well-formed octahedrons of pyrite and sphalerite up to 6 µm (Figure S1h), as well as sulfide phases of complex composition in the form of dendrites and solid masses (Figure S1i);
• Aggregates of noncrystalline Fe-Si oxyhydroxides from individual thin filaments of similar diameter and shape as in the TW field (Figure S1j). EDS data indicates that the content of SiO₂ in aggregates is 32-34%. In addition to iron (34-37 wt%) and calcium, they also contain a significant admixture of phosphorus (up to 3 wt.%) (Table S6).

No gypsum microcrystallites were identified in the sinking material from the SM field.

3.3.4. Major and Trace Element Composition and REE

The trace element composition of the sinking matter is heterogeneous and exhibits a significant depth-related variability (Table 2). In both fields, the settling matter has similar elemental composition characteristics. The concentration of sulfide-forming elements, including Cu, Zn, Pb, Fe, and Mo, and other hydrothermally derived elements, such as Ba, Sr, Cd, Tl, and Eu, increases with depth and reaches a maximum value at a distance of 30 m from the bottom (depths of 520 and 640 m) (Table 2). With regard to other elements (both hydrothermally derived and rock-forming), the maximum content at 100 m from the bottom (depths of 450 and 570 m) is characteristic for Mn, V, Cr, Co, Ga, As, Rb, Y, Zr, Cs, Al, Hf, Th, as well as for most REEs (Table 3). The remaining elements (Si, P, Ti, Ni, Nb, Ta, U and Gd) do not exhibit a discernible correlation with depth. However, it can be observed that their content generally increases in the deep water layer compared to the upper layer (0-150 m).

The content of Al ranges from 0.77 to 1.78 wt.%, with an average of 1.25 wt.%. The lowest content of rock-forming elements, including Al, Ti, Zr, Hf, as well as Nb, Ta, Ni, Y, which are components of rock-forming and accessory minerals of igneous rocks, was identified in the sinking matter of the subsurface 150 m layer. The greatest values of their content are observed at depth of 100–200 m above the bottom, corresponding to depths of 450-350 m in the TW field and 570–470 m in the SM field (Table 2). A robust linear relationship was observed among the items within this group, where the coefficient of determination (R²) ranged from 0.78 to 0.93.

In order to ascertain the potential for a non-detrital origin for the Si concentration in the trapped matter, a plot of Si against Al was constructed (Figure 6a). A correlation coefficient of 0.60 has been determined for the pairwise relationship between Si and Al in the samples. Si normalized to Al vary from 8.5 to 15.3 while the average crustal ratio (Si/Al)_UCC = 3.83 [63], indicating the input of amorphous biogenic silica – bSiO₂. Concurrently, the enrichment factor (EF) for Si reaches 3.6-4.0 in the upper layers of the water column, predominantly as a consequence of silicon-containing phytoplankton. The Si/Al ratio in the bottom sediments varied from 1.3 to 3.9 [69].

The majority of elements exhibit EF values that are less than 3. The following elements have been identified as having an additional source other than rock-forming: Cu, Zn, Tl, Co, V, Ga, Cd, Sr, Mo, Pb, Eu, and Ba (Figure 7). The values of the EF for Cu, Zn, Sr, Ba, Mo, Cd, and Eu significantly increased at both fields in the lowermost depths. The SM field demonstrates a more than 10-fold increase in the values of EF elements (Cu, Zn, Sr, Ba, Mo, Cd) when compared to the overlying layer. This indicates an increase in the content of hydrothermally derived sulfides and sulfates in the bottom layer.

The Ba content reaches 0.4 and 1.2 wt.% in the bottom layer of the TW and SM vent fields, respectively. As the depth decreases, the Ba content in both fields declines and reaches values up to 0.1–0.06 wt.%. Sinking material throughout the water column is enriched in Ba relative to the average crustal ratio ((Ba/Al)_UCC = 0.008): EF varies from 3 to 11 in the upper and deep water layers and reaches 50-113 near the bottom. Ba exhibits a reliable correlation with a high pair correlation coefficient (R = 0.76-0.98) in the presence of certain elements: Zn, Pb, Mo, Tl, Eu, Sr, Cu and Cr (in descending order of R). The majority of these elements exhibit chalcophilic (Zn, Pb, Mo, Tl, Cu) or siderophilic (Cr) properties.

The most pronounced enrichment of sulfide-forming elements, including Fe, Cu and Zn, relative to the average crustal ratio (Fe(Cu, Zn)/Ti)_UCC, is observed in material trapped near the bottom (Figure 7 and Figure 8). With regard to Fe, the highest EF is observed in near-the-bottom layer at the TW vent field. For Cu, a significant enrichment of sinking material is observed throughout the deep water column of both fields. However, the maximum EF value is also reached in the benthic layer of the TW vent field. In contrast, the greatest enrichment of Zn is observed in the bottom layer of the SM vent field, where its content reaches 1.2 wt.%, which is four times higher than its concentration in the same layer of the TW field (Table 2). Fe shows the strongest correlations with Co, V, Ga and Cu, while Cu correlates most with Cd, Sr, Mo, Pb, Eu, Ba and Zn. Zn, in turn, shows the strongest correlations with Tl, Ba, Pb, Mo, Eu, Sr, Cr and Cu. The concentration of Cu and Pb in the near-the-bottom layer is about twice that of the overlying water column, while the concentration of Sr is about five times higher. The concentration of Pb is about twofold and that of Tl is ninefold higher in the sinking material at the bottom of the SM vent field than at the bottom of the TW field. With regard to Cd, the highest degree of enrichment (EF 10 times higher than the average crustal ratio (Cd/Al)_UCC) was observed in the benthic layer at the TW field.

The particulate Mn concentration in the water column is characterized by an enrichment of trapped matter, with the highest concentrations observed at a depth of 450-470 m at both vent fields. The concentration of Mn in the deep water column is slightly higher in the TW vent field than in the SM vent field. Mn shows robust correlations with Cs, As and Rb, as well as with some light rare earth elements (LREEs), including Ce, Sm, La and Nd, and Th. The highest concentration of labile elements, including As and Cs, as well as Rb, is observed at a depth of 100 m above the bottom (layer of 450 m) at the TW vent field (Table 3).

The concentration of LREEs in the sinking material shows a range of values, varying from 18 to 54 ppm at the TW vent field and from 25 to 38 ppm at the SM vent field. The concentration of LREEs reached its maximum value at a depth of 100 m above the bottom at the TW vent field. At the same time, the shale-normalized REE patterns of sinking material show both slight negative Ce anomalies, and positive Eu anomalies (Figure 9, Table 4), as well as a depletion of matter with REEs compared to NASC. The greatest depletion of rare earth elements (REEs) was observed at a depth of 150 m in the TW field, where the biogenic content reached its highest value.

3.3.5. Trace Element Fluxes

A calculation was performed to determine the vertical fluxes of a number of sulfide-forming elements (Zn, Cu, Mo, etc.), as well as those of Ba, Sr, and Cd, etc. (Table 5). The most remarkable feature is the flux of elements whose values increase in the lower layers at both sites, which are postulated to be associated with hydrothermally derived material. These elements include Fe, Cu, Zn, Sr, Mo, Cd, Ba, Eu, Tl and Pb.

4. Discussion

4.1. Environmental Conditions at the JMVFs Control Buoyant Hydrothermal Plume Transport

The CTD-profiler and moored temperature sensors in the benthic layers of both stations recorded positive temperature anomalies, indicating that the sediment traps in the benthic layer at both the TW and SM field were located within the area of influence of the hydrothermal plume, which was the purpose of the deployment. Particle settling in the lowest layer of the TW field occurred in the context of the dominant northeastward transport of bottom water, oriented at approximately 45°. This basically coincides with the spatial orientation of the rift valley in the study area [50] and also corresponds to the general direction of water mass transport by the Jan Mayen Current [35,36], taking into account the local bottom topography.

A comparison of the fluctuation in instrument depth at TW and SM moorings under the influence of currents revealed that this influence occurred in a synchronous manner at both fields (Figure 5). This observation indicates a close hydrodynamic regime at the lower layers of both hydrothermal fields. Consequently, in the vicinity of both sources, the hydrothermal plume propagates subhorizontally in a northeasterly direction (~45°) within a layer that extends no more than 100 m from the bottom. In the SM field, tidal motion appears to be responsible for the periodic inflow of cold deep water from depths exceeding the investigated depths by tens of meters. Thus, at the background station located at a depth of 2375 m, 50 km from the polygon, water exhibiting such characteristics was recorded at a depth of 850 m.

4.2. Comparative Study of Total Particle Flux and Major Particle Phase Composition at the JMVFs with Similar Available Data from Other Hydrothermal Systems

The particle fluxes obtained in the present study are generally slightly higher than the average flux observed in the study region during the June sampling period. For example, in the Lofoten Basin at a depth of 550 m during the year from August 2017 to July 2018, the fluxes showed a range of values, from 3 mg/m²/day in March to 191 mg/m²/day in September. However, fluxes did not exceed 10 mg/m²/day in June [78]. A two-year exposure of sediment traps on the northern segment of the Mohns Ridge in June 2019 revealed a single increase in total particle fluxes to 233 mg/m²/day [57]. Long-term studies in the Nordic Seas have shown that fluxes at 500 m a depth also did not exceed 200 mg/m²/day [79,80,81]. Furthermore, the flux maximum showed a notable intra-annual shift, occurring in June in 1989 and in August in 1990 and 1991, respectively. The only exception to this observation is the occurrence of a brief, transient flux maximum (430 mg/m²/day) in August 1991 [81].

The elevated fluxes observed in our study can be attributed to spring microalgae blooms during the sampling period. This is supported by the high chlorophyll concentrations observed in the surface layer of the study area (up to 5.4 mg/m³) and the maximum subsurface chlorophyll flux observed at the TW mooring (0.291 mg/m²/day at 150 m). Unfortunately, satellite data were not available for the study area during the sampling period due to dense cloud cover. However, a general satellite map of the entire Norwegian-Greenland Basin reveals a pronounced bloom in the ice-melt-free waters, with a more subtle manifestation observed on the Arctic front to the north and northwest of the study site (Figure 10). Long-term studies have consistently demonstrated that the Greenland Sea reaches a maximum level of primary production in June [80,82,83]. The annual increase in total particle flux at this site in June and July coincides with the melting of sea ice and the formation of a haline stratification, which facilitates the growth of spring phytoplankton [80].

Consequently, the primary driver of fluxes in the subsurface is the biological pump and the supply of biogenic particles from the upper active layer of the ocean. In the lowermost layer, there is a noticeable increase in the proportion of lithogenic material derived from hydrothermal vent fields. Available data on sinking hydrothermally derived particle fluxes are limited for meaningful comparative studies with other submarine hydrothermal systems. Nevertheless, the published data we have compiled (see Table 5) allow us to draw a couple of important conclusions. It is important to note that particle fluxes in this area are considerably lower than those observed at other hydrothermal vent fields on the Mid-Atlantic Ridge (MAR). For example, fluxes ranging from 280 to 19,200 mg/m²/day have been observed at the Rainbow field in the vicinity of the hydrothermal vent (Table 5 [13]). In the TAG and Broken Spur vent fields, the fluxes reached 5200 mg/m²/day and 1800 mg/m²/day, respectively [28,77]. At the Lucky Strike vent field beneath the Sintra vent, the flux was significantly lower, averaging 264 mg/m²/day [15], yet even these values are almost three times higher than those obtained in our study. As shown by the data presented, there is considerable variability in material fluxes at hydrothermal fluxes in different geographical regions. Furthermore, there is a notable temporal variability in fluxes, with a 70-fold difference observed over the 16-day trapping experiment at Rainbow vent field [13].

Consequently, the intensity of the hydrothermal manifestations observed in the fields studied and the influence of the hydrothermal plume on the aquatic environment are significantly lower than those observed in other known hydrothermal fields of the MAR, as previously reported [41]. In terms of the vertical particle flux values, the TW and SM fields are most similar to the Lucky Strike vent field influenced by the Azores hot spot. A comparison of the chemical and mineral composition of bottom sediments in these fields revealed a high degree of similarity [41].

The near-bottom layers, which are influenced by the delivery of matter from high-temperature hydrothermal fluids, are not clearly distinguished by the ratio of the main components of the flux. Of interest is a slight decrease in Al content in the flux (by 13–22% of the overlying layer), which is the main lithogenic tracer of rock-forming particle input. At the same time, the total flux remains unchanged and even increases by 15–20%.

4.3. Characterization of Major and Trace Elements in Sinking Matter to Identify Hydrothermal Contribution

Both moorings demonstrate an increase in excess iron (Fe_ex) with depth. The Fe_ex content varies between 0.13% and 1.01% at the TW field, and between 0.26% and 0.43% in the SM field. The maximum Fe_ex content is observed at depths of 450 and 520 meters within the TW field. The proportion of Fe_ex in the sinking matter varies from a minimum of 26% to a maximum of 63% of the total iron content. The EF values for Fe do not exceed the value of 3 at any water level, with the maximum value of 2.7 occurring at a depth of 520 m of the TW field. This indicates a significant proportion of lithogenic matter in sinking particles. The elevated iron concentration observed in the TW field plume is a consequence of the higher content of iron-silicate oxyhydroxides, with Fe concentrations reaching 45 wt.% (Table S3). The content of Fe-Si oxyhydroxides and their Fe concentration (up to 37 wt.%) were found to be lower in the sinking matter of the SM field than in the sinking matter of the TW vent field. In the composition of Fe-Si oxyhydroxides, Ca in both fields and P in the SM vent field are reported as simple substitutions. It is well established that Fe oxyhydroxide particles are capable of absorbing Ca and P from seawater [7]. Despite the fact that pyrite crystals were observed among the sinking particles from both sites, there is evidence from microscopic examination that Fe-Si oxyhydroxide precipitation is more prevalent than Fe sulfide precipitation.

The highest concentration of manganese (Mn) is observed in the 450-470 m layer, particularly in the particles of the TW plume (up to 1472 ppm). This may be attributed to the oxidation of Mn, which shows a higher concentration in the hydrothermal plume and its precipitation from the solution [84]. According to [5], Mn and abiogenic gases (CH₄, ³He) had similar elevated concentrations coinciding with density changes close to 500 m. In TW vent field samples, synchronously with the increase in Mn concentration, there is an increase in the concentration and EF of Cs, As, Rb and Ce, which are co-precipitated with Mn oxyhydroxide particles. This suggests scavenging processes [85] or the enrichment of these elements in basalts [46,86].

The concentration of Cd, Zn, and P in sinking particles is slightly elevated in the upper water layer above 150 m. Thus, high concentrations of these elements in the surface water are due to their occurrence in the biogenic particles [26] that are abundant in these productive waters with Chl-a concentrations of 1–2 mg/m³.

The high Al content and the close relationship between Si and Al content (R² = 0.71, n = 8) indicate a significant role of aluminosilicates in the composition of sinking material, which is supported by EDS data. A number of elements that are associated with lithogenic matter and included in rock-forming (groups of olivines, pyroxenes, plagioclases, etc.) and accessory minerals of igneous rocks are characterized by a vertical distribution similar to that of aluminum: Ti, V, Zr, Hf, Nb, Ta, Ni, most of REE (except Eu) and Th. The content of these elements reaches its maximum value at depths of 100–200 m below the surface water layer. This serves as a proxy for the absence of the benthic nepheloid layer, as no discernible increase in the concentration of these elements due to the resuspension of surface bottom sediments was observed. The elevated concentration of Ni and Cr (and their correlation with Ta, Y, and Nb) in the sinking material as well as in the background bottom sediments outside the hydrothermal fields [41] can be attributed to the influence of the Jan Mayen rocks [46,86], which are enriched in incompatible elements. In the near-bottom layers, in addition to organo-mineral aggregates of sinking particles from the euphotic zone of the water column, an abundance of minerals of hydrothermal genesis (barite, gypsum, Fe, Cu and Zn sulfides and Fe-Si oxyhydroxides) is observed. Elevated concentrations of Fe, Mn, Ba, Cu, Pb, Cd were previously observed in SPM of the JMVFs in the benthic and intermediate layers [84]. The sinking particles of the studied fields represents a mixture of different components and form a specific population of particles including the pelagic/hemipelagic background particles, volcanogenic particles and particles derived from Jan Mayen basalts and basalts associated with submarine eruptions, and hydrothermally derived particles [41,87].

The Ba content involved in barite formation is 2.5 times higher in the composition of sinking particles from the SM vent field than from the TW vent field. SEM examination demonstrated that barites occur in a typical rosette morphology consisting of cross-cut tabular crystals, with sizes varying from 50 to 120 µm (Figure S1 c, g, h, i). Hydrothermal barite is a primary precipitate of Ba-rich pure (mantle-derived) hydrothermal fluids interacting and mixing with sulfate-rich seawater [88,89,90]. Non-radiogenic Sr (as well as Ba) is leached from the oceanic crust (basalt). EDS microanalysis for achieve chemical characterization of barite has shown average contents of 52.79 ± 9.33% for Ba, 14.72 ± 0.78% for S, and 2.81 ± 1.67% for Sr on SM (Table S2) and 44.10 ± 1.85% for Ba, 17.17 ± 1.22% for S, and 2.84 ± 0.79% for Sr on TW (Table S5). More than 90% of barite consists of Ba and SO₄, and the Ba/S molar ratio of the crystals (range from 0.52 to 1.08) is close to that of pure barite. Using the stoichiometric ratio and subtracting the Ba contribution from settling from upper active layer particles, the vertical flux of hydrothermal barite can be calculated to be 694 and 2103 µg/m²/day at TW and SM vents, respectively.

The particles of the studied buoyant plumes are characterized by the following association of chalcophilic trace elements Zn, Fe, Cd, Cu, Pb, V and Tl. Among the sulfide minerals, zinc sulfides (sphalerite) are the most significant. Sulfides of Fe, Cu, etc. are less substantial. It has been observed by Stensland et al. [5] that copper and iron exhibit similar profiles with peaks at two distinct depths of water column: approximately 500 and 600 m. A similar pattern is observed in our samples, wherein Fe forms an association with Cu, as well as Co and V, and enriches the sediment at the same depths. The SEM data indicate that collomorphic (bud -shaped) porous microconcretions of Fe sulfides form a mineral association with sulfides of Zn and other sulfide-forming elements. Only pyrite and sphalerite form octahedra and cuboctahedra of 3–6 µm in diameter. Hydrothermally derived barite forms a stable mineral association with sulfides. The mineral barite has been previously observed in association with zinc sulfides in the bottom sediments of the JMVFs [41,87]. Using the stoichiometric ratio and subtracting the flux of Zn from the upper layers, it would be possible to estimate the flux of sphalerite, which was as high as 313 and 1676 µg/m²/day at the TW and SM vent sites, respectively.

The above-mentioned features in the particle composition of the studied buoyant plumes appear to be related to the temperature of the fluids in the studied vent fields. Consequently, the maximum fluid temperature of both hydrothermal vent fields reaches approximately 270°C [39,41], however, irregularly shaped, barite-silica chimneys with large flanges that emit low-temperature fluids are also found on the SM vent field [39].

4.4. REE in Particle Fluxes as a Traces of Hydrotermaly Derived Particles

In general, the REE spectrum of the trapped material is very similar to that of alkaline Jan Mayen basalts (Figure 8), as reported by Haase et al. [71], indicating a substantial contribution of this material. However, the total REE content is 4-6 times lower than that in basalts. This can be explained by the admixture of organic and carbonate matter and the admixture of hydrothermally derived sulfide minerals, which have the effect of reducing the REE content when introduced into the sinking matter. REE patterns can be influenced by minor aluminosilicate detritus due to the low REE content in the hydrothermal precipitates [91]. Negative cerium anomalies (Ce_an=0.82–0.89) suggest scavenging of REE by suspended matter in the buoyant plume from seawater. As previously reported, significant REE scavenging must mostly occur after deposition of suspended hydrothermal precipitates from the water column [85,91,92]. This is why the negative cerium anomaly isn't very pronounced. There is also one sample with a positive cerium anomaly (Ce_an = 1.17). It was sampled at the TW vent site at a depth of 450 m, where the Mn content is high (Table 2). This indicates the presence of a significant amount of hydrous manganese oxides in the sample. In general, both studied vent sites are characterized by a decrease in Ce_an values at upper layers of water column, indicating an increasing in the influence of dissolved REEs from seawater with decreasing depth, as reported by Dubinin [68,93]. A number of indicator ratios (light to heavy REEs ratio, Eu_an) suggest a potential hydrothermal influence of in the water layer up to 100 m from the seafloor [94].

The influence of hydrothermal buoyant plume is most pronounced at 30 m from the bottom, but is also evident at virtually all sampling depths in the water column. This may be due to the delivery of nanoparticles during the collapse of gas bubbles due to pressure reduction during surfacing [5]. The positive europium anomaly in the water column above 100 m from the seafloor may be partly due to the supply of Jan Mayen basaltic material with the Jan Mayen Current. Thus, basaltic tephra has been identified along the entire length of a bottom sediment core sampled in the southeast of the island [95], and is also abundant in the surface sediments of the Jan Mayen AVR [41,87].

4.5. Fluxes of Hydrothermally Derived Elements and Specific Features of Studied Hydrothermal Buoyant Plumes

The ratio of the individual element fluxes in the bottom layer to the overlying background fluxes is higher in the SM field, with the exception of Fe (Table 5), which is likely due to differences in the locations of the studied vent fields in relation to the AVR. In addition, the SM vent field characterize by two types of venting: white smokers and clear plumes with a wide temperature range from 50 to 270 ℃. The extensive yellow-brown Fe oxyhydroxide deposits and iron enrichment of bottom sediments observed in the TW vent field area have been previously documented in our work [41,53] and in the literature [39,45]. Furthermore, the enrichment of the water column with particulate and dissolved iron, delivered with gas bubbles above the TW vent field, has also been reported in [5]. It has been demonstrated that metal migration occurs above the hydrothermal plume, specifically on the hydrate-coated gas bubbles. As the bubbles rise, the hydrates are disrupted, resulting in the release of metal-rich nanoparticles. Further coagulation of nanoparticles can result in the enrichment of the sedimentary matter with metals such as Co, V, Ga, and Cu.

Comparison of the elemental fluxes with the available data for a number of other hydrothermal vent fields (Table 5, [13,15,28,29,31,77]) revealed that the fluxes of Fe, Cu, and Zn in the JMVFs are several orders of magnitude lower than those observed in other vent fields. This phenomenon has previously been demonstrated in the context of trace element concentrations in suspended particulate matter in the vicinity of the TW vent field [41]. At the same time, the Ba flux is comparable to other vent fields and even exceeds its magnitude at the Broken Spur vent field. Ba content in the sinking matter of the studied TW and SM vent fields is one of the highest among the known low- and medium-temperature hydrothermal systems. The ratio Ba in the flux is the closest to that of the Lucky Strike vent field [15].

The MAR hydrothermal vent fields, located at 14°45'–37°30' N, are also associated with basaltic volcanism and occur at variable depths, ranging from 850 to 3670 m. Primary fluid temperatures have been observed to range from 363°C to 152°C (Table 5). It has been demonstrated that the fluids of the Lucky Strike and Menezes Gwen vent fields exhibit compositional differences from other MAR hydrothermal fields located south of the Azores hotspot. These differences include a lower temperature, a high pH value (up to 5), a lower concentration of Cl, Na, Fe, Li, Cu and an enrichment of Ba, Zn, Pb, Cs, Rb. These characteristics are typically attributed to the elevated concentration of the aforementioned elements in the basalts that were initially emplaced near the Azores hotspot [41,96]. Nevertheless, despite the proximity of the P-T conditions of the JMVFs to those of the Lucky Strike and Menezes Gwen vent fields and to the basalts involved in hydrothermal particle formation, the fluxes and concentrations of these elements in the TW and SM trapped matter differ to a greater extent (Table 5). In the deeper waters (>3000 m) and at high temperatures (>350°C) of the Broken Spur and TAG vent fields, the flux of barium is minimal among all MAR vent sites. Increasing fluxes of Cu and Zn are observed here.

It is also possible that the observed differences in the magnitude of total fluxes and fluxes of hydrothermally derived elements may be attributed to variations in the conditions of sinking particles collected by sediment traps at different vent fields. The moorings were deployed at varying distances from the vents and from the bottom of the seafloor [13,15,28,29,31,77].

In the course of our study, the temperature of the water in the lower layer of the TW vent field exhibited a number of increases, which resulted in the occurrence of the plume impact on the traps on a number of occasions (Figure 5). In the SM vent field, such impacts were recorded on just four occasions during the running time of the trap. They were recorded once a day during high tide. This phenomenon can be attributed to a change in the direction of plume transport under the influence of the tide and a change in the temperature regime of the source, which depends on the phase of the tide. This phenomenon has also been observed at the Lucky Strike vent field [97]. At the same time, the fluxes of hydrothermally derived elements and minerals, particularly barium (Ba) and zinc (Zn), and their respective ores, barite and sphalerite, exhibited significantly elevated levels precisely at the SM vent field. In general, these facts indicate a higher particle release by white smokers at the SM vent field compared to the TW vent field. The data obtained emphasise the circulation of several immiscible phases and fluids of slightly different compositions in the studied hydrothermal systems within the Jan Mayen vent field area and possibly the different degree of maturity of the studied ore-generating hydrothermal systems.

5. Conclusions

This study examines sinking particle fluxes and their composition at two white smoker venting sites (Troll Wall and Soria Moria) of the JMVFs located at the AVR and formed volcanogenic massive sulphide deposits, dominated by barite-sphalerite mineralization. Our findings contribute to the limited number of studies investigating sinking particle fluxes from buoyant and neutrally buoyant hydrothermal vent fields in the ocean. The supply of hydrothermally-derived matter is accompanied by normal pelagic/hemipelagic sedimentation, which is dominated by biogenic particles, especially in the upper water layers, and basaltic particles from seafloor and Jan Mayen rocks. In general, the JMVFs exhibit relatively low plumes turbidity and hence low vertical fluxes of sinking particles, which is markedly lower than at other hydrothermal vent fields of the MAR and EPR (only 103 mg/m²/day against 264–11589 mg/m²/day [13,15,28,29,31,75,76,77]). The hydrothermal plume propagates subhorizontally in a northeasterly direction (~45°) within a layer that extends no more than 100 m from the bottom. The impact of hydrothermally derived particles is most evident at a depth of 30 m above the seafloor. Transportation of buoyancy hydrothermal plumes was significantly affected to tidal motion. Nevertheless, the influence of hydrothermally derived particles is observed for certain parameters in almost the entire water column from the bottom to the subsurface layers due to the relative shallowness of the Jan Mayen vent field area.

Hydrothermally derived particles were represented by a low- to medium-temperature mineral association including barite, sphalerite, pyrite, chalcopyrite, gypsum, and non-crystalline Fe-Si oxyhydroxides. The vertical flux of hydrothermal barite at the SM is ~3 times higher than at the TW: 2103 µg/m²/day and 694 µg/m²/day, respectively. The flux of sphalerite, the next abundant hydrothermal mineral, at the SM is ~5 times higher than at the TW: 1676 µg/m²/day and 313 µg/m²/day, respectively. Therefore, despite the proximity of the studied vent sites to each other and similar fluid features characteristic of volcanogenic massive sulfide deposits, differences in the composition of the settling hydrothermal matter of the TW and SM fields were clearly identified. These differences were observed in the elemental composition of the particles, with some specialization evident. SM vent field is distinguished by a more pronounced influx of sedimentary matter compared to TW field, accompanied by elevated fluxes of Zn and Ba. On the contrary, the TW field exhibits elevated values of Fe and Mn fluxes. This phenomenon is likely attributable to the presence of sources of varying configuration and intensity, with differing fluid temperatures, within the SM field. Furthermore, data obtained emphasise the circulation of several immiscible phases and fluids of slightly different compositions in the studied hydrothermal systems. In terms of the sinking matter composition, the TW and SM vent fields are most similar to the Menez Gwen and Lucky Strike vent fields affected by Azores hotspot. Our study yields important geochemical and mineralogical data regarding hydrothermally derived sinking particles that provide a unique perspective on the mafic-hosted hydrothermal settings affected by hotspots.