Organic Carbon Oxidation in the Sediment of the Ulleung Basin in the East Sea

Abstract

We characterized the biogeochemical organic carbon (C_org) cycles in the surface sediment layer of the Ulleung Basin (UB) of the East Sea. The total oxygen uptake (TOU) rate and the diffusive oxygen uptake (DOU) rate of the sediment were measured using an autonomous in situ benthic lander equipped with a benthic chamber (KIOST BelcII) and a microprofiler (KIOST BelpII). The TOU rate was in the range of 1.51 to 1.93 mmol O₂ m⁻² d⁻¹, about double the DOU rate. The high TOU/DOU ratio implies that the benthic biological activity in the upper sediment layer is one of the important factors controlling benthic remineralization. The in situ oxygen exposure time was about 20 days, which is comparable to the values of other continental margin sediments. The sedimentary C_org oxidation rates ranged from 6.4 to 6.5 g C m⁻² yr⁻¹, which accounted for ~2% of the primary production in UB. The C_org burial fluxes ranged from 3.14 ± 0.12 to 3.48 ± 0.60 g C m⁻² yr⁻¹, corresponding to more than 30% of the deposited C_org buried into the inactive sediment deep layer.

1. Introduction

The Ulleung Basin (UB) is located in the southern part of the East Sea (ES). The oligotrophic Tsushima warm current flows in through the narrow Korea Strait between Korea and Japan, and it separates into a couple of strong streams by physical and meteorological forces close to the ES mouth [1]. As the floating warm waters move in the surface layer to the north of the ES, the complicated oceanographic features (seasonal upwelling, eddies, and seasonal variation in mixed layer thickness) may enhance the primary production in the euphotic depth [2,3,4,5,6,7,8], and half of the particulate organic carbon (C_org) produced by primary production in the euphotic depth (<100 m) can be transported vertically to the deep-sea sediment [9]. Thus, the UB has been suggested as a ‘hot spot’ for biogeochemical cycles of C_org, not only in the water column but also in the sediment [5].

Oxygen, the most favorable oxidizer for C_org decomposition, is used as an electron acceptor for C_org degradation [10,11]. The by-products produced via C_org degradation in the porewater are reoxidized in the oxic zone of the sediment layer. Consequently, oxygen consumption in the sediment can be represented quantitatively and can thus be used as a proxy of the total benthic carbon mineralization [10,11]. The sediment oxygen uptake rate has been measured using an enclosure incubation method and/or O₂ microprofiling methods. The total oxygen uptake rate (TOU) measured by enclosure incubation in the laboratory or in situ represents the sum of oxygen consumption by benthic biology activities and carbon oxidation, which can be estimated from the initial rate of oxygen decrease in the enclosing chamber. Otherwise, the O₂ microprofiling method measures the diffusive oxygen uptake rate (DOU) via the sediment–water interface (SWI) using the high spatial distribution of O₂ in the diffusive boundary layer (DBL) near SWI by an oxygen microsensor (sensor tip < 1 mm) [12,13].

It has been difficult to measure the reliable sediment oxygen uptake rate for the deep-sea sediment (>1 km) because of unknown artifacts present during the sampling treatments [11]. The alteration of physical factors such as temperature and pressure during sediment recovery may allow for increasing sediment oxygen uptake. In addition, the inhomogeneous fauna density in cores also gives a significant bias to the results. Therefore, the ex situ sediment oxygen consumptions may have an uncertainty for the deep-sea sediment [11].

The main purpose of this study was to understand the biogeochemical C_org in the sediment of the UB, the most productive basin in the ES. To achieve our purpose, we simultaneously measured the TOU and DOU using an autonomous in situ benthic lander to quantify the C_org oxidation in the sediment. The partitioning sedimentary organic carbon fluxes were assessed to build the mass budget of C_org in the UB.

2. Materials and Methods

2.1. Study Area

The ES was expanded by the back-arc basin spreading between the Asian continent and Japan Island, which is one of the deepest marginal seas and is composed of three major basins (Ulleung, Yamato, and Japan) in the northwest Pacific (Figure 1) [14]. Because of the bowl shape of ES, with a relatively small spatial scale, the steep topography at the shelf–slope system allows for the mass transport of the sediment into the basin interior [14,15,16]. The UB, located in the southern part of the ES, is the entrance of the warm surface water of the Tsushima current. Along the Korean peninsula, temporal upwelling and warm eddies have been observed, and these may be important elements of primary production in the photic zone of the southwestern part of the ES [2,3,4,5,6,7,8].

2.2. In Situ Measurement

The in situ measurements were carried out in September 2018 and April 2019 in the northern part of the UB (Figure 1,

Table 1. Site location, water depth, temperature, salinity, and dissolved oxygen in bottom water.

	D3	U1
Latitude	37°00 N	37°24 N
Longitude	131°00 E	131°12 E
Water depth (m)	2154	2206
Bottom water salinity	34.07	34.06
Bottom water temperature (°C)	0.22	0.12
Bottom water DO (μmol L−1)	186	189

). The physical and biological parameters in the water column were measured before and after benthic deployment using CTD (SBE911Plus, Seabird Electronics Inc., Bellevue, WA, USA).

To measure the in situ sediment oxygen uptake rate (TOU and DOU), the autonomous benthic landers (KIOST BelcII and BelpII) were deployed. In brief, the benthic lander system was composed of a rectangular benthic chamber (BelcII) and miroprofiler (BelpII) [17,18,19]. The BelcII in situ incubation chamber was designed to measure the oxygen in the interior of the chamber to estimate the TOU. Additionally, the automatic water sampler can collect the incubated water samples in the chamber at preset time intervals to allow for benthic fluxes. Meanwhile, the BelpII consists of the four separate O₂ microoptode (Pyroscience, OXR50-HS-SUB & FSO2-SUBPORT, Aachen, Germany) systems and a custom-built motor-driven linear stage, allowing for the measurement of the high spatial distribution (<100 μm) of oxygen at the SWI [19].

Before the deployment of the benthic lander, we programmed each parameter for measurements into the microcontroller of the systems. The lander was deployed freely from a research vessel, and the distance from the ship to the system was continuously monitored using an acoustic pinger (Teledyne Benthos, 865-A) until touchdown. After finishing the in situ measurement (>24 h), the lander was ascended by releasing a weight. The incubated water samples of the BelcII were immediately filtered with 0.2 μm pore size filter paper (Milipore) and then stored in a deep freezer (−20 °C) until analysis in the laboratory [20].

2.3. Sediment Collection

To collect undisturbed sediment core samples, we carefully inserted a plexiglass tube (8 cm internal diameter, 40 cm long) into the sediment in the box corer. After sealing both ends of the core, it was temporarily stored in a refrigerator until core sectioning (<a few hours). The sediment core was sectioned at 1 cm intervals for measuring the C_org and the total nitrogen content (T-N) of the sediment and was stored in a deep freezer (<−20 °C) [18].

2.4. Laboratory Experiment

The in situ incubated water samples for nutrients were analyzed using a flow injection autoanalyzer (Bran+Luebbe, QuAAtro39 AutoAnalyzer) [20]. The water content of sediment was determined using weight loss after drying to a constant weight at 105 °C. The dry bulk density (DBD, g cm⁻³) of sediment was calculated using the water content of the sediment and the grain density [18]. The dried sediment samples were acidified with 1 M HCl to measure the total C_org and nitrogen content of the sediment, and those were quantified by a CHN analyzer (Thermo Finnigan, Flash EA 1112, USA).

2.5. Flux Calculation

The benthic fluxes (TOU and nutrients) were estimated from a least-squares linear regression using the concentration gradients with the time and depth of the benthic chamber, as follows:

$${\mathrm{F}}_{\mathrm{TOU},\mathrm{BNF}}=\left(\frac{\mathrm{dC}}{\mathrm{dt}}\right)\times \left(\frac{\mathrm{V}}{\mathrm{A}}\right)$$

(1)

where F_{TOU, BNF} is the TOU and the benthic nutrient fluxes (mmol m⁻² d⁻¹), dC/dt is the gradient of the linear regression line estimated from fitting the concentration (C) as a function of the incubation time (t) (mmol m⁻³ d⁻¹), V is the chamber volume (m³), and A is the chamber area (m²) [17,18].

Using the high spatial resolution of the oxygen profiles, the upper DBL boundary and the sediment surface position were determined [12,13]. The upper DBL boundary was determined from the intersection of the constant oxygen concentration in the overlying water and the extrapolated linear oxygen gradient in the DBL. The sediment surface (z = 0) position was identified from a distinct change in the linear slope between the DBL and the porewater concentration. The oxygen penetration depth (OPD) was determined from the O₂ profiles as the depth where the signal of the O₂ microsensor reached a constant.

Using the high resolution of the vertical O₂ distributions, the DOUs and volumetric oxygen consumption rates were estimated by applying the numerical model PROFILE [21]. The two boundary conditions (BCs) in PROFILE were as follows: the first BC was the O₂ concentration at the sediment–water interface, and the second BC was that the flux at the bottom was zero. The apparent sediment molecular diffusion coefficient (D_s) was calculated with a diffusion coefficient of O₂ in the in situ temperature and sediment porosity [22,23].

The replicate the in situ measurements, OPD and DOU were compared using the nonparametric statistical Wilcoxon–Mann–Whitney test with a 95% confidence level. The statistical analysis was performed using the RStudio package.

2.6. C_org Budget Calculation

The C_org mass budget was calculated with following equation, assuming a steady state [24]:

C_in = C_ox + C_burial

(2)

where C_in is the input flux of C_org into the sediment from the water column, and C_ox is the oxidation flux of C_org that was determined by multiplying TOU and the Redfield ratio (106/138). The burial flux of C_org (C_burial) at the sediment layer was calculated with the following equation:

C_burial = C_∞ × DBD × SR

(3)

where C_∞ is the mean C_org content (%) that shows the constant values of the lower layer (>30 cm), DBD is the mean DBD (g cm⁻³) in the lower layer, and SR is the sedimentation rate of UB (0.06 cm yr⁻¹) [18].

3. Results

3.1. TOU and DOU

The TOU measurement at D3 in 2018 failed due to a failure of a microcontroller unit in BelcII (Figure 2). The TOU measured at U1 in September of 2018 was 1.51 mmol m⁻² d⁻¹. With no significant spatial differences, the TOU values in 2019 were 1.89 and 1.93 mmol m⁻² d⁻¹ at D3 and U1, respectively (

Table 2. The results of TOU, DOU, and benthic nutrient fluxes. The errors for TOU and BNF were calculated from the standard error for linear regression. The DOU errors represent the standard error for the duplicate measurements (n = 6).

		TOU	DOU	NH4+	NOx−	PO43−	Si(OH)4
		(mmol m−2 d−1)
2018	D3	-	-	−0.09 ± 0.06	1.68 ± 0.54	0.09 ± 0.05	8.67± 2.18
				(R2 = 0.21)	(R2 = 0.58)	(R2 = 0.32)	(R2 = 0.69)
	U1	1.52 ± 0.00	0.77 ± 0.13	−0.05 ± 0.07	0.29 ± 0.07	0.06 ± 0.03	1.08 ± 0.39
				(R2 = 0.07)	(R2 = 0.78)	(R2 = 0.34)	(R2 = 0.6)
2019	D3	1.89 ± 0.00	-	−0.13 ± 0.07	1.41 ± 0.46	0.12 ± 0.04	6.98 ± 2.71
				(R2 = 0.33)	(R2 = 0.61)	(R2 = 0.51)	(R2 = 0.45)
	U1	1.93 ± 0.00	1.00 ± 0.15	0.3 ± 0.04	−0.16 ± 0.07	0.01 ± 0.0	2.23 ± 0.25
				(R2 = 0.88)	(R2 = 0.51)	(R2 = 0.57)	(R2 = 0.93)

). However, the TOU values at U1 for 2019 were about 30% higher than those measured in September 2018. The result measured at D3 in April 2019 matched well the previous results of the same season [18].

The temporal variation of the O₂ microprofiles at U1 in the autumn and spring were not statistically significant for OPD (p = 0.125) and DOU (p = 0.25), which yielded 11.7 ± 0.4 mm for 2018, 12.4 ± 0.5 mm for 2019, 0.77 ± 0.13 mmol m⁻² d⁻¹ for 2018, and 1.00 ± 0.15 mmol m⁻² d⁻¹ for 2019 (Figure 3). The ratios of DOU and TOU were about 0.5, which was comparable to other deep-water depth results [11]. The volume-specific oxygen consumption rate (OCR) was divided into three layers in the oxic layer. In 2019, the OCR value in the upper surface sediment (up to 22.8 mm: 0.0029 nmol cm⁻³ s⁻¹) was higher by a factor of 1.75 than in 2018 (up to 28.5 mm: 0.0017 nmol cm⁻³ s⁻¹), which was similar to TOU—i.e., the sediment oxygen consumption was higher in April than in September, which may show a seasonal variation.

3.2. Benthic Nutrient Flux

The evolution of nutrients (ammonium, nitrate, phosphate, and silicate) with time generally shows organic matter degradation in the surface sediment (Figure 4). From the linear relationship of nutrient concentration with time, the benthic nutrient fluxes (BNFs) were estimated and are listed in Table 2. Overall, the nutrients gradually increased with time, but the ammonium and phosphate suggested poor linear regression results (R² < 0.4). This result may be explained ad the incubation times (22 h) for the estimation of ammonium flux, which seems to be insufficient for concentration buildup in the chamber. The ammonium, nitrate, phosphate, and silicate benthic fluxes ranged from −0.05 ± 0.07 to 0.30 ± 0.04, −0.16 ± 0.07 to 1.68 ± 0.54, 0.01 ± 0.00 to 0.12 ± 0.04, and 1.08 ± 0.39 to 8.67 ± 2.18 mmol m⁻² d⁻¹, respectively. These results were comparable to previous in situ measurement results [18].

3.3. Vertical Distribution of C_org and Total Nitrogen Content

The C_org and T-N in the surface sediment (z = 0) at D3 and U1 were in the range of 2.53% to 2.73% and 0.33% to 0.36%, respectively (Figure 5). They were vertically decreased to ~10 cm sediment depth, and then both reached a constant content below 20 cm. The mean contents of C_org and T-N of the upper sediment layer (<10 cm) at U1 were 15% higher than at D3 (p = 0.002). However, the mean values of those below 20 cm of sediment depth were not statistically significant (p = 0.21). The same result was obtained for T-N, i.e., the upper surface T-N content at U1 was significantly higher than at D1, but below 20 cm of sediment depth. The results did not differ at both sites. The results for the sediment C_org and T-N were similar to previous results [20,25,26].

4. Discussion

The primary production (PP) in the UB ranged from 273 to 280 g C m⁻² yr⁻¹, which was higher than that of the Kuroshio current region [8,27]. The higher export C_org fluxes from the euphotic zone may be expected, because the benthic remineralization in the sediment of the UB can be significantly higher than other deep-sea sediments [9]. Indeed, the C_org content and the significantly higher rate of anaerobic respirations (NO_x⁻, Mn²⁺, SO₄²⁻) in the surface sediment layer of UB compared with other deep-sea sediments have suggested that the biogeochemical cycles of C_org and its anaerobic degradation pathways are important factors controlling the biogeochemical cycles of C_org [9,18].

4.1. Benthic C_org Oxidation in UB

The sediment oxygen uptake rate (TOU and DOU) has been widely used as a substitute for the C_org oxidation rate because it can represent the electron acceptors for total OM degradation [10,11,28]. The TOU, measured by the benthic chamber, normally represents the sum of the total remineralization rate of organic matter (OM), assuming that the O₂ in the incubated water is ultimately the consumption by OM oxidation, the reoxidation of reduced materials, and the benthic animal activities [11]. The DOU represents the oxygen uptake across the SWI followed by the concentration gradient between the water and sediment porewater, which is controlled mainly by the diffusion [10,11]. Therefore, the DOU may exclude the oxygen uptake mediated by benthic animal activities. The difference between the TOU and DOU (the fauna mediated oxygen uptake rate, FOU = TOU−DOU) is significant in the coastal sediment, where there is a higher biomass of benthic animals, which can account for about 50% of the TOU [11]. The depth attenuation of the FOU can be explained by the distribution of biomass with water depth and the shift in the benthic animal feeding type from very active irrigating specimens to a less active deposit-feeder according to the sediment carbon content [11]. By contrast, it is less important at more than 1000 m water depth sediment because of the relatively low benthic fauna biomass. Note that our FOU was higher than for other deep-sea sediments, which implies that the respiration of benthos and the reoxidation of reduced materials by bioirrigation and biological mixing could contribute to the TOU. Indeed, the volumetric OCRs in the oxic–anoxic interface were comparable with the upper oxic layer (Figure 3). Our results suggest that the high reoxidation of the reduced elements (i.e., NH₄⁺, Mn²⁺, Fe²⁺, and H₂S) at the oxic-anoxic interface can be enhanced by biological activities (bioirrigation and biological mixing), and thus, the thick Mn-oxide enrichment layer at the surface sediment of UB may be coupled with the cycles of Mn-oxide trapping via Mn²⁺ reoxidation. Consequently, these oxygen dynamics in the sediment layer can be an important step for controlling the reduced elements via organic matter oxidation pathways [9,10,11].

4.2. Benthic C_org Oxidation in UB

The partitioning C_org fluxes (C_ox, C_burial, and C_in) and burial efficiency (BE) are listed in

Table 3. Organic carbon oxidation (C_ox), burial flux (C_burial), input flux (C_in), and burial efficiency (BE) in the sediment of the UB.

	Cox	Cburial	Cin	BE
	(g C m−2 year−1)			(%)
D3	6.51	3.48 ± 0.60	9.99 ± 0.60	34 ± 6
U1	6.36	3.14 ± 0.12	9.51 ± 0.12	33 ± 1

. Assuming a steady state, the C_org oxidation will follow Redfield’s stoichiometric relationship between C and O₂. The benthic C_org oxidation rate (C_ox) at the sediment of the UB ranged from 5.1 to 6.5 g C m⁻² yr⁻¹, which was similar to previous results [18]. The C_ox at D3 and U1 were 6.61 and 6.36 g C m⁻² yr⁻¹, respectively, which corresponded to about 2% of the primary production in the southwestern part of the ES. The C_burial at D3 and U1 were given as 3.48 ± 0.60 and 3.14 ± 0.12 g C m⁻² yr⁻¹, respectively, which did not differ spatially and matched previous results well [18]. The C_burial fluxes accounted for ~1% of the annual mean primary production of the UB. Given that, at the UB, all deposited C_org is oxidized and buried into the sediment layer, the sedimentary C_org input fluxes (C_in = C_ox + C_burial) were 9.99 ± 0.60 and 9.51 ± 0.12 g C m⁻² yr⁻¹, respectively, which suggested that the benthic cycles of C_org corresponded to about 3% of PP. The C_org budget across the slope to the basin suggests that the lateral transport C_org and the segregation of organic matter for those transports determine their fate [18]. At the deep interior, far away from the slope of the UB, the refractory organic matter could be progressively enriched following preferential degradation during lateral transport, and thus, the laterally transported C_org may contribute in a similar order to the vertical fluxes of C_org. This result suggests that the sediment of the UB can act as the burial place of C_org [18]. The burial efficiency of C_org in the sediment layer (BE = C_burial/C_in × 100) was about 34% at both stations, which was comparable with the other high productive marginal sea sediments [29,30,31,32,33,34].

4.3. Oxygen Exposure Time (OET) in the UB

The in situ OET is the exposure time of C_org to oxygen at the location of deposition, which is important for understanding the mechanisms of C_org preservation in the sediment layer and/or the transport in the marginal sea [35,36]. We calculated the in situ OET at U1 by dividing the SR from the OPD (OET = OPD/SR), which were 20 ± 3 and 21 ± 4 y, which were similar to the California margin (1500–3500 m) and the Washington lower slope (600–2000 m) [35,36].

Kang et al. (2004) [37] reported that the O₂ concentration of the bottom water in the ES has decreased during the last century (1932–1996) with changes in the water mass structure. Given that the O₂ decrease in the bottom water will be progressing in the future, the OPD will be shallower, and thus, C_org preservation in the sediment will be more efficient. This implies that the sediment of the UB can act as the carbon reservoir and reactors. Our research is based on a fragmentary study, and a quantitative study on the biogeochemical cycles of the carbon-production-remineralization-transport-burial across from the slope to the basin system is needed to understand the function of the northeast Pacific margin sea in the global system.

5. Conclusions

In situ experiments and geochemical analyses were performed to understand the biogeochemical cycles of C_org in the benthic layer of the UB in the ES. The C_org oxidation rate of the sediment was consistent with previous results, which were relatively higher than other deep-sea sediments [18]. The high ratio of TOU/DOU suggests that the benthic biological activity in the surface sediment layer is an important factor for controlling the biogeochemical cycles of C_org. The large contribution of the O₂ consumption in the oxic–anoxic interface may be coupled with the Mn-oxide enrichment of the surface sediment layer. The relatively higher BE and OET compared to those of other marginal sea sediments suggests that the redistribution by lateral transport is one of the important keys in the biogeochemical cycles of C_org in the benthic boundary layer of the UB.