Distribution Characteristics and Dynamics of Marine Hydrogen in the Eastern Indian Ocean

Abstract

The ocean serves as a significant contributor of atmospheric Hydrogen (H₂) with indirect greenhouse effects. However, uncertainties persist regarding internal production and consumption processes of marine H₂, as well as controlling factors. Our study examined the spatial distribution and source-sink dynamics of marine H₂ in the Eastern Indian Ocean. H₂ concentrations in surface seawater exhibited a range of 2.95–21.96 nmol L⁻¹. High concentrations of H₂ were observed in the anoxic water in the Bay of Bengal. Rates of H₂ photo-production and microbial consumption in surface seawater ranged from 1.80 to 17.78 nmol L⁻¹ h⁻¹ and 1.02–9.18 nmol L⁻¹ h⁻¹, respectively. When considering the entire mixed layer, photo-production contribute to approximately 31%–43% of the total H₂ removal, with cyanobacteria potentially serving as another source in the mixed layer. Compared with the sea-to-air exchange, microbial consumption was the primary removal pathway of H₂ in seawater.

Key Points

• The distribution of H₂ was significantly affected by river input in the Eastern Indian Ocean

• Photo-production was an important source of H₂ in the mixed layer

• Microbial consumption was the primary sink for H₂ in the mixed layer

Plain Language Summary

Atmospheric hydrogen (H₂) can influence the environment and climate by consuming hydroxyl radicals (OH·) and indirectly raising greenhouse gas concentrations. Although the ocean serves as a significant source of atmospheric H₂, the biogeochemical processes governing its presence in seawater remain poorly understood. The Eastern Indian Ocean, characterized by a substantial inflow of freshwater, exerts a distinct impact on the local ecosystem. We conducted a field investigation in the Eastern Indian Ocean to clarify the sources, sinks, and controlling factors of H₂, including the Bay of Bengal with relatively higher primary productivity and the Eastern Equatorial Indian Ocean with low primary productivity, respectively. Our study involved the quantitative assessment of H₂ photo-production, microbial consumption, and sea-to-air exchange in seawater, along with the calculation of the H₂ budget in the mixed layer. This investigation enhances our understanding of H₂ cycling processes in seawater and contributes to the assessment of H₂ emissions from the ocean and their impact on the atmospheric budget.

1 Introduction

Molecular hydrogen (H₂) is the second most abundant reducing gas in the troposphere after methane (Ehhalt & Prather, 2001). Its residence time is estimated between 1.4 and 2.3 years (Rhee et al., 2006; Sanderson et al., 2003; Xiao et al., 2007). H₂ reacts with hydroxyl radicals (OH·), indirectly increasing the concentrations of CH₄ and the other greenhouse gases (Ehhalt & Rohrer, 2009; Popa et al., 2015; Schultz et al., 2003; Warwick et al., 2004). Consequently, H₂ is deemed as an indirect greenhouse gas influencing global climate. In general, the sources of H₂ in the atmosphere include the emissions from fossil fuel use, biomass burning, photochemical oxidation of volatile organic compounds, nitrogen (N₂) fixation, and ocean emission (Ehhalt & Rohrer, 2009). The ocean contributes to the global H₂ budget, estimated between 2 and 6 Tg yr⁻¹, with notable uncertainties (Ehhalt & Rohrer, 2009; Hauglustaine & Ehhalt, 2002; Novelli et al., 1999; Pieterse et al., 2013). These uncertainties arise from extrapolations based on limited data regarding the entire oceanic contribution.

The concentration of H₂ is typically supersaturated in surface seawater (Herr & Barger, 1978; Moore et al., 2009; Punshon et al., 2007; Schropp et al., 1987). The N₂ fixation has been widely recognized to be an important source of H₂ (Moore et al., 2009; Scranton, 1984). In this process, H₂ is produced as a specific byproduct in the following reaction:

$$${\mathrm{N}}_{2}+8{\mathrm{H}}^{+}+8{\mathrm{e}}^{-}+16\text{ATP}\,\to \,2{\text{NH}}_{3}+{\mathrm{H}}_{2}+16\text{ADP}+16{\mathrm{P}}_{\mathrm{i}}\infty $$$

Marine N₂ fixation is primarily driven by cyanobacteria in warm and oligotrophic seawater (Mahaffey et al., 2005; Shiozaki et al., 2010). Biological H₂ production mediated by anaerobic bacteria in low-oxygen areas has been considered a possible process (Herr et al., 1984; Schropp et al., 1987). Moreover, laboratory simulation experiments have shown that abiotic photo-production from chromophoric dissolved organic matter (CDOM) and low molecular weight organic matter such as acetaldehyde could lead to H₂ production (Punshon & Moore, 2008). However, there is still a lack of in-situ data regarding the photo-production of H₂ in seawater. The microbial consumption of H₂ is thermodynamically favorable and indeed occurs (Herr et al., 1981, 1984). Additionally, the high saturation of H₂ in surface seawater results in a net flux from the ocean to the atmosphere (Punshon et al., 2007).

The Eastern Indian Ocean is a semi-open ocean, and its marine ecosystem is influenced by a variety of land-based inputs due to its unique geographical location. River inflow notably impacts the ecosystem in the Eastern Indian Ocean, especially in the Bay of Bengal (Ghosh et al., 2024; Shetye et al., 1991). Many rivers in the Asian continent, such as the Ganges and Brahmaputra, feed large amounts of fresh water, nutrients and organic matter into the Bay of Bengal. These inputs markedly affect the bay's physical, chemical, and biological processes (Sengupta et al., 2006), which potentially influence the H₂ source and sink processes. For example, the latest research showed the concentration of microplastics would influence the production of H₂, while the Ganges is an important source of microplastics in the bay (Alam et al., 2023; Wei et al., 2022), which could enhance the concentration of H₂ in seawater. Additionally, the substantial freshwater input leads to strong stratification in the near-surface layer that inhibits vertical mixing, creating a sharp and intense oxygen minimum zone (Rao et al., 2016). The exchange of gases between the sea and air in the Eastern Indian Ocean is highly active, facilitated by the monsoon winds and elevated surface seawater temperatures. This may have a great impact on the atmospheric H₂ budget. In this study, we investigated the distribution characteristics and controlling factors of H₂ and calculated the sea-to-air flux of H₂ during autumn. Additionally, we conducted in-situ irradiation and incubation experiments to examine the photo-production and microbial consumption of H₂. These observations aim to provide important data support for the global scale emission of H₂ from the ocean and understand the regional characteristics and biogeochemical cycling process of H₂.

2 Methods

2.1 Voyage and Sampling

Our measurements were conducted on board the R/V “Shiyan 3” in the Eastern Indian Ocean from 27 September to 6 November 2020. The study area and detailed sampling stations are shown in Figure 1. The seawater samples of H₂ were collected from 64 stations by using 12-L volume Niskin bottles installed in a Seabird 911 conductivity-temperature-depth (CTD) rosette, from which seawater salinity and temperature were also recorded. Atmospheric H₂ samples were collected on the deck about 10 m above the sea level. To avoid pollution caused by ship sailing, all atmospheric samples were collected windward with a 10 mL airtight syringe while the ship was sailing. Samples for other environment parameters (chlorophyll-a, dissolved organic carbon [DOC], CDOM, nutrients, and cyanobacteria) were treated onboard for storage and brought back to the laboratory for analysis. The sampling and analytical methods are described Text S1–S3 in Supporting Information S1. In addition to sampling, in-suit incubation experiments were conducted at 11 stations.

Figure 1

Location of the sampling stations (red rhombus: Stations for microbial consumption and irradiation experiments; yellow triangle: Stations for irradiation experiments; green square: Stations for microbial consumption; BBW: Bay of Bengal water; EIW: Eastern equatorial Indian Ocean water) and Horizontal distribution of temperature, salinity, H₂, DOC, chlorophyll-a and cyanobacteria (pico-) in the surface seawater (about 5 m).

2.2 Analysis Method of H₂

In this study, all H₂ samples were measured onboard by Trace 3000R reduced gas detector (Ametek, USA). Before analysis, the instrument was calibrated using the H₂ standard gas with a mixing ratio of 500 ppbv, supplied by the State Center for Standard Matter (China). The calibration was repeated every day. The measurement of H₂ in seawater was conducted using the headspace equilibrium method (Xie et al., 2002). A 50 ml bottle was filled with a seawater sample collected in the field, sealed, and then the seawater sample was replaced with 8 mL of high purity N₂ (purity >99.99999%) using a gas-tight syringe. The treated sample was shaken at 300 r min⁻¹ for 5 min to ensure that the headspace in the bottle was in gas-liquid equilibrium; 6 mL of the equilibrium gas was extracted using a gas-tight syringe and injected through PTFE hydrophobic filter membrane into the instrument for determination. The measured value was converted to obtain the concentration of H₂ in seawater. The specific calculation formulas are shown Text S4 in Supporting Information S1. The detection limit was 0.02 nmol L⁻¹ with a relative standard deviation of less than 4.4%.

2.3 Incubation Experiments

2.3.1 Photo-Production

An in-situ irradiation experiment was selectively conducted to determine the photochemical production rates of H₂ and assess the influence of different wavelengths on this process. To avoid the influences of algae, microorganisms, and particles, seawater samples were filtered twice directly into pre-cleaned bottles through 0.22 μm polyethersulfone filters that were thoroughly rinsed with sample water (Park et al., 2021). Filtered samples were then transferred into 150 ml quartz tubes with Teflon stoppers for subsequent spectral treatments: (a) quartz tubes (full-spectrum solar radiation). (b) one layer of UF3 Plexiglas-wrapped quartz tubes (attenuating all UV and transmitting most of the visible light). (c) three layers of aluminum foil-wrapped quartz tubes (dark). The UV radiation results were obtained by subtracting the results of the PAR light from the results of the full-spectrum solar radiation. All quartz tubes were exposed to solar radiation in an incubator equipped with circulation equipment to maintain the in-situ temperature of the seawater samples. All samples were cultured for a duration of 6 hr, after which 50 mL aliquots were taken to determine the H₂ concentration. Measurements were taken at half-hourly intervals using a UV-Vis spectroradiometer (OL 756, Gooch and Housego, UK), with the incident photon flux density being recorded.

We assessed and integrated the photo-production process of H₂ resulting from UV and PAR across the water column. The light intensity for a specific type of radiation, denoted as X, at a depth of Z (PFD [X, Z]), can be derived from the surface light intensity (PFD [X, 0]) by exponentially attenuating it with depth, as prescribed by the Beer-Lambert equation:

$$$\text{PFD}\,[\mathrm{X},\mathrm{Z}]=\text{PFD}\,[\mathrm{X},0]\times {\mathrm{e}}^{-\text{kxZ}}$$$

Light attenuation underwater is primarily influenced by four main factors: water itself, CDOM, chlorophyll-a, and particles. The concentration of particular matter is generally low in the open ocean, and thus particles was omitted from our calculation (Pérez et al., 2016). Numerous studies have demonstrated a strong correlation between the UV attenuation coefficient and CDOM (Johannessen et al., 2003; Para et al., 2013), with CDOM recognized as the primary contributor to UV attenuation in the ocean (Pérez et al., 2016). Therefore, in our study, the UV attenuation coefficient (k_X, m⁻¹) is approximated as the sum of the absorption coefficient of pure water (Buiteveld et al., 1994; Pope & Fry, 1997) and the CDOM absorption coefficient at the corresponding wavelength (a_W [X] and a_CDOM [X], m⁻¹):

$$${k}_{\text{uv}}={a}_{\mathrm{W}}\,(\text{UV})+{a}_{\text{CDOM}}\,(\text{UV})$$$

The attenuation of solar radiation by phytoplankton mainly affects PAR. The attenuation coefficient of phytoplankton can be expressed by the concentration of chlorophyll-a ([Chl], mg L⁻¹), and the formula is as follows (Kloster et al., 2006):

$$${k}_{\text{phy}}=0.03\times [\text{Chl}]$$$

The attenuation coefficient of PAR waveband (k_PAR m⁻¹) can be expressed as:

$$${k}_{\text{PAR}}={a}_{\mathrm{W}}(\text{PAR})+{a}_{\text{CDOM}}(\text{PAR})+{k}_{\text{phy}}$$$

Combined with the attenuation formula of solar radiation, the photo-production rate at X waveband of depth Z (K [X, Z]) can be expressed as:

$$$\mathrm{K}(\mathrm{X},\mathrm{Z})=\mathrm{K}(\mathrm{X},0)\times \text{PFD}(\mathrm{X},\mathrm{Z})\,/\,\text{PFD}(\mathrm{X},0)=\mathrm{K}(\mathrm{X},0)\times {\mathrm{e}}^{-\text{kxZ}}$$$

The mixed layer depth was defined as the depth at which the density was 0.03 kg m⁻³ lower than surface water (to the reference depth of 10 dbar) (de Boyer Montégut et al., 2004).

2.3.2 Microbial Consumption

Surface seawater was collected to investigate the microbial consumption rates of H₂. Seawater samples were incubated on board in 500 ml airtight glass syringes onboard. Urea was added to the syringes to achieve a final concentration of 10 mmol L⁻¹, which served to inhibit the N₂ fixation (Rawson, 1985). As for the control group, the water samples underwent repeated filtration through 0.22 μm polyethersulfone filters for twice. Subsequently, all samples were placed in a lightproof incubator to prevent photoreaction and the in-situ surface seawater temperature was maintained by circulating surface seawater. After a 6 hr incubation, the concentration of H₂ was determined in both the experimental group and control group to calculate the microbial consumption rate of H₂.

2.4 Sea-To-Air Flux of H₂

The instantaneous sea-to-air flux (F) of H₂ was calculated by the two-layer model (Liss & Merlivat, 1986):

$$$F=k\left({\left[{\mathrm{H}}_{2}\right]}_{\text{surf}}\,\mbox{--}\,{\left[{\mathrm{H}}_{2}\right]}_{\text{eq}}\right)$$$

where F is the flux (nmol·m⁻²·hr⁻¹); k is the sea-to-air exchange velocity (m·s⁻¹); [H₂]_surf and [H₂]_eq represent the initial concentration of H₂ in surface seawater and the concentration of H₂ in seawater when atmospheric H₂ is at equilibrium with the seawater. Detailed calculation formulas are provided Text S4 in Supporting Information S1.

3 Results and Discussion

3.1 Oceanographic Background

Vertical profiles of temperature, salinity, chlorophyll-a, dissolved oxygen, DOC, and nutrients at a depth of 0–200 m along 87°E transect in the Eastern Indian Ocean are illustrated in Figure S1 in Supporting Information S1. The substantial influx of fresh water into the Bay of Bengal has led to a wide range of surface salinity (31.79–35.18) across the study area, limiting the transport of oxygen from mixed layer to subsurface seawater which resulted in the formation of low oxygen zones. It suggested that the study area exhibited different marine environments. Through K-means cluster analysis of the surface salinity (Table S1 in Supporting Information S1), the area was partitioned into the Bay of Bengal water (BBW) and the Eastern Equatorial Indian Ocean Water (EIW) to explore the distribution characteristics of and biogeochemical processes of H₂ in different environments. The concentration of DOC in surface seawater of the two regions showed a significant difference (t = 2.095, p < 0.005). Notably, the two highest concentrations of DOC in surface seawater both appeared in the Bay of Bengal (A03: 1.589 mg L⁻¹; E87-36: 1.466 mg L⁻¹). This was due to the abundant organic matter entering the Bay of Bengal along with fresh water. The other environment parameters of two regions are shown in Table S2 in Supporting Information S1.

3.2 Distributions and Controlling Factors of H₂

3.2.1 Horizontal Distribution of H₂ in Seawater

Horizontal distribution of temperature, salinity, H₂, chlorophyll-a, DOC and cyanobacteria (pico-) are shown in Figure 1. The average concentrations of H₂ in surface seawater were 10.66 ± 5.67 (3.81–21.96) nmol L⁻¹ and 4.98 ± 1.91 (2.75–9.98) nmol L⁻¹ in BBW and EIW, respectively. The concentration of H₂ in BBW significantly exceeded that in EIW (t = 3.133, p < 0.001). We compared the concentrations of H₂, chlorophyll-a and cell density of cyanobacteria in the surface seawater and found no correlation between them. However, a relationship was observed between the H₂ concentration and the cell density of Trichodesmium sp (r = 0.833, n = 7, p < 0.05, Table S3 in Supporting Information S1). Since H₂ is a byproduct of N₂ fixation while Trichodesmium is the important undertaker (Capone et al., 1997; Scranton, 1984), it suggests that N₂ fixation play a significant role in H₂ production. In addition, high concentrations of H₂ were observed in low salinity stations of the Bay of Bengal, exhibiting a significant negative correlation with salinity (r = −0.547, n = 64, p < 0.01, Table S4 in Supporting Information S1). This might be attributed to the river input, which provided large amounts of freshwater, nutrients, and organic matter, influencing the biogeochemical processes of H₂ in the Eastern Indian Ocean. To further explore the cause of high H₂ value in the BBW, the production and loss processes of marine H₂ have been conducted and will be discussed in the following sections.

3.2.2 Distribution of H₂ in Anoxic Seawater

The vertical profiles at most stations in EIW presented that the highest H₂ concentration occurred within the upper 50 m, with a rapid decrease below this depth and fluctuated slightly (Figure S1 in Supporting Information S1). Conversely, a distinct vertical distribution trend was observed in BBW, where higher H₂ values were consistently found in anoxic water layers (DO < 2.00 mg L⁻¹). For instance, concentrations of H₂ measured at a depth of 150 m at E87-37 station (20.84 nmol L⁻¹) and A03 (12.20 nmol L⁻¹) were 1.85 and 3.84 times higher than those in the surface seawater, respectively. Schropp et al. (1987) conducted a nutrient enrichment experiment on anoxic seawater, revealing elevated anaerobic production rates of H₂. The abundance of nutrients in the anoxic water of the Bay of Bengal (seen in Figure S2 in Supporting Information S1) likely promoted the growth of anaerobic bacteria, thereby facilitating the release of H₂.

3.3 Photo-Production of H₂

3.3.1 Photo-Production of H₂ in the Surface Seawater

The rates of H₂ photo-production driven by UV and PAR in surface seawater, along with cumulative solar radiation photon fluxes, are depicted in Figure 2. The photo-production under full-spectrum radiation (K_photo) was 13.31 ± 4.28 (7.53–17.77) nmol L⁻¹ h⁻¹ and 2.12 ± 1.47 (1.80–5.78) in the BBW and EIW. Dark control experiments did not yield significant H₂ production. The average cumulative solar radiation photon fluxes during irradiation experiments were similar between BBW (12.59 ± 3.51 E m⁻²) and EIW (12.55 ± 2.97 E m⁻²). Variances in H₂ photo-production rates between the two regions can be ascribed to differences in physicochemical properties, CDOM contents and compositions, and photosensitizer (nitrate) levels (Mack & Bolton, 1999). The absorption coefficient at 280 nm is used as a proxy to estimate CDOM concentration (Andrew et al., 2013). Our experiment results demonstrated a correlation between H₂ photo-production rate and both CDOM (r = 0.820, n = 8, p < 0.05) and nitrate (r = 0.846, n = 8, p < 0.01), respectively. The aCDOM (280) in the BBW (0.68 ± 0.07 m⁻¹) was higher than that in the EIW (0.53 ± 0.05 m⁻¹) and the nitrate concentration in surface seawater was also higher in BBW (0.033 ± 0.028 μmol L⁻¹) compared to EIW (0.024 ± 0.021 μmol L⁻¹). The elevated levels of CDOM and photosensitizer in BBW contributed to higher H2 photo-production rates.

Figure 2

(a) H₂ photo-production rates (K_photo) in ultraviolet radiation (UV) and photosynthetically active radiation (PAR) of surface seawater at each station with solar radiation photon fluxes (K_normalized) (the dotted line shows the boundary between the Bay of Bengal Water [BBW] and the Eastern Indian Ocean water [EIW]). (b) Modeled depth-specific (nmol L⁻¹ h⁻¹) of ultraviolet radiation (UV) and photosynthetically active radiation (PAR) mediated H₂ photo-production in the entire water column (depth-specific rates were integrated across the specified water column depth to obtain depth-integrated rates; the green and black dotted line represents the mixed layer depth and effective photo-production activity depth, respectively).

3.3.2 Effects of Radiation Quality on Photo-Production

The average rates of UV (K_UV) and PAR (K_PAR) in BBW were 5.87 ± 2.13 and 7.44 ± 2.16 nmol L⁻¹ h⁻¹, and those in EIW were 1.32 ± 0.52 and 1.53 ± 1.00 nmol L⁻¹, respectively. The average integral photo fluxes for UV and PAR during the experiments were calculated to be 0.46 ± 0.02 E m⁻², 12.2 ± 3.50 E m⁻² in the BBW, and 0.96 ± 0.38 E m⁻², 11.8 ± 2.73 E m⁻² in EIW, respectively. Despite UV radiation accounting for less than 10% of the total solar radiation, it contributed on average 43.2% ± 3.0% and 46.3% ± 8.4% of the total H₂ photo-production in BBW and EIW, respectively. In addition, we gave a normalized H₂ photo-production ratio, for UV: PAR of 45: 1 and 19: 1 in the BBW and EIW, respectively. It indicated that the UV efficiency of surface seawater in the BBW was higher than EIW. The ratio of the spectral slope (SR) of the shorter waveband (275–295 nm) to that of the longer waveband (350–400 nm) distinguished the origin of CDOM; smaller SR value indicated a higher proportion of terrestrial CDOM and vice versa for marine source. As shown in Table S5 in Supporting Information S1, the S_R in the BBW (1.617–1.824) was smaller than that in EIW (1.880–3.345), which indicated that the terrestrial CDOM was more prone to photochemical reaction and had a higher efficiency for H₂ photo-production.

3.3.3 Modeled Photo-Production Throughout the Mixed Layer

To scale the experiment results to the water column, we simulated the H₂ photo-production rates in the entire upper water for all irradiation stations (Figure 2b and Figure S3 in Supporting Information S1). Since the mixed layer is the depth horizon that regulates sea-to-air exchange, the following analysis focused on the photo-production within the mixed layer. We integrated the photo-production rates of UV and PAR within the depth to facilitate a more intuitive comparison of their contributions to H₂ photo-production in the mixed layer. To roughly assess the influence of H₂ photo-production in water column, an effective photo-production depth was defined as the depth at which solar radiation attenuates to 1% of the sea surface. If the depth of the mixed layer was shallower than the effective photo-production layer, the depth-integrated rate was calculated within the mixed layer depth. Otherwise, it was calculated within the effective photo-production layer depth. The average depth-integrated rate of UV and PAR in the mixed layer were 8.73 ± 5.74 (2.66–18.36) μmol m⁻² h⁻¹ and 19.3 ± 14.1 (4.77–41.1) μmol m⁻² h⁻¹, respectively, with average contributions to H₂ photo-production of approximately 31% and 69%, respectively. Compared with the process of H₂ photo-production in surface seawater, PAR emerged as the dominant driver of photo-production in the mixed layer.

3.4 Microbial Consumption of H₂

The microbial consumption rates of H₂ (K_micro) were calculated to be 6.70 ± 2.48 (4.22–9.18) and 1.39 ± 1.94 (1.02–1.91) nmol L⁻¹ h⁻¹ in the BBW and EIW. A positive correlation between microbial consumption rate and H₂ concentration was observed (r = 0.827, n = 6, p < 0.05) across the entire study area, indicating that higher H₂ concentrations were associated with faster microbial consumption rates. The consumption rates in EIW were comparable to those reported by Scranton (1984) (1.25–1.95 nmol L⁻¹ h⁻¹), while the consumption rates in BBW were generally higher. It could be attributed to the abundance of bacteria in the Bay of Bengal. A previous study found a strong positive correlation between bacterial abundance and chlorophyll-a concentration (Bird & Kalff, 1984). The average concentration of chlorophyll-a of microbial consumption stations in the BBW (0.20 ± 0.04 μg L⁻¹) was higher than that in the EIW (0.11 ± 0.07 μg L⁻¹) (despite the lack of statistical significance), indicating that bacteria in the BBW possessed higher activity and more efficient removal of H_2.

3.5 Sea-To-Air Flux

The average instantaneous sea-to-air flux of H₂ was 7.29 ± 6.04 (0.49–26.0) μmol m⁻² d⁻¹. The Eastern Indian Ocean was a net source of atmospheric H₂. Our flux estimates were higher than previous investigations in the Northern Atlantic Ocean (3.04 μmol m⁻² d⁻¹) and the Southern Atlantic Ocean (2.10 μmol m⁻² d⁻¹), which could be attributed to the higher concentrations of H₂ (the Northern Atlantic Ocean: 0.18–3.26 nmol L⁻¹, the Southern Atlantic Ocean:0.40–1.40 nmol L⁻¹) (Herr & Barger, 1978; Herr et al., 1984).

The sea-to-air flux of H₂ did not show a statistical difference between the two regions. It could be attributed to the effect of wind speeds and temperature of seawater on the sea-to-air exchange. A sensitivity analysis was conducted to evaluate the impact of these factors on the sea-to-air exchange, as shown in Figure S4a in Supporting Information S1. The results indicated that wind speed, seawater temperature, and H₂ concentration all affected the sea-to-air flux. Among these factors, wind speed exerted the most significant influence, followed by H₂ concentrations. A strong correlation was observed between wind speed and flux, with a 55% variation in H₂ flux (Figure S4b in Supporting Information S1). The higher wind speeds in the EIW (8.99 m s⁻¹ vs. 6.58 m s⁻¹ in the BBW) promoted this process. Although the concentration of H₂ in the BBW was significantly higher than that in EIW, it was insufficient to offset the effect of wind speed, resulting in comparable sea-to-air fluxes between the two regions.

3.6 Budget of H₂ in the Mixed Layer

We developed a budget model based on our research (Figure 3), to detail the source-sink dynamics of marine H₂ within the mixed layer. This model elucidated the biogeochemical process and rates of H₂. Since the H₂ photo-production rate varied significantly with depth, the average photo-production rates of the mixed layer were calculated as the quotient between depth-integrated rate across the mixed layer and the depth of the mixed layer. Turnover times for each removal pathway were calculated using the initial H₂ concentrations divided by their corresponding rates (Table S6 in Supporting Information S1).

Figure 3

Source-sink budget models for the H₂ within the mixed layer in the BBW and EIW region of the Eastern Indian Ocean (The red letters are potential source and sink of hydrogen). The numerical values in the figure indicate that the mean rates (standard deviation) of source and sink within the mixed layer in the study survey are expressed in nmol L⁻¹ h⁻¹.

Rates of microbial consumption and sea-to-air exchange within the mixed layer were 6.70 ± 2.48 and 0.0176 ± 0.001 nmol L⁻¹ h⁻¹ in BBW, and those in EIW were 1.39 ± 0.33 and 0.0067 ± 0.004 nmol L⁻¹ h⁻¹, respectively. It indicated that the majority of H₂ participated in the biogeochemical process, and only a small portion was emitted to the atmosphere. The total turnover times of H₂ were 0.10 ± 0.02 and 0.13 ± 0.05 days in BBW and EIW, respectively, suggesting the H₂ turnover in the Eastern Indian Ocean could be highly dynamic. The average rate of photo-production in the mixed layer in BBW (2.87 ± 0.86 nmol L⁻³ h⁻¹) was higher than that in EIW (0.44 ± 0.27 nmol L⁻³ h⁻¹), which could explain about 43% and 31% of the total removal of H₂ in the BBW and EIW, respectively. This high level of photosensitizer and CDOM resulted in a relatively higher contribution of photo-production in the mixed layer of BBW. However, it was evident that the photo-production could not maintain the turnover of H₂. Moore et al. (2009) observed a correlation between N₂ fixation rate (μmol N m⁻³ d⁻¹) and H₂ concentration (nmol L⁻¹) in the Pacific Ocean (N₂ fixation = 1.73 [H₂] ‒ 0.15, r = 0.98). Previous studies found that the N₂ fixation rate in the Eastern Indian Ocean is not lower than that in the Pacific Ocean (Shao et al., 2023). In addition, Price et al. (2007) estimated the N₂ fixation contribute at least 55% to total production of H₂ in the ocean, which suggested that the gap between sources and sinks of H₂ could be attributed to the N₂ fixation.

4 Summary

We conducted a systematic investigation in the Eastern Indian Ocean to get new insight into the occurrence and cycle of H₂ and assess the H₂ budget in the mixed layer for the first time. Elvated concentration of H₂ was primarily detected in region impacted by river input. In-situ incubation experiment revealed that photo-production rates of H₂ in the surface seawater were related to the CDOM and photosensitizer. The region with a high proportion of terrestrial CDOM had higher efficiencies of photo-production. The photo-production efficiency of the UV waveband was much higher than that of PAR, though the UV radiation only accounted for a fraction of the total solar radiation, it contributed more than 40% of the H₂ photo-production. Compared with the surface seawater, the proportion of H₂ photo-production in the UV waveband in the mixed layer was reduced, and the PAR was the dominant driver of photo-production. Microbial consumption was the main way of H₂ removal instead of sea-to-air exchange. Considering the budget in the mixed layer, the photo-production could not maintain the total removal of H₂. N₂ fixation could be another important source of H₂. These results clarify the in-situ concentrations and relative importance of different sources and sinks of the Eastern Indian Ocean, which improve our understanding of the cycling of H₂ in marine.

Acknowledgments

We thank the captain and crews of the R/V “Shiyan 3” for their assistance and cooperation during the investigation. This work was financially supported by the National Natural Science Foundation of China (Grant 42276042, 41876082 and 42225601), the Laoshan Laboratory (Grant LSKJ202201701), and the Fundamental Research Funds for the Central Universities (Grant 202372001 and 202072001).

Open Research

Data Availability Statement

Data presented in this paper are publicly available at Figshare via Jiang (2024).