Preliminary study on tracing 14C in groundwater-derived dissolved organic carbon: transport, transformation, and seawater recirculation in Jeju Island, South Korea

Submarine groundwater discharge (SGD) is a key but understudied pathway in the terrestrial-oceanic dissolved organic carbon (DOC) cycle. In this study, fresh and saline groundwater samples were collected from two sites on Jeju Island, South Korea: Hwasun and Bangdu Bay on the western and eastern coast, respectively. DOC concentrations in fresh groundwater were extremely low (5 ± 3 μM), with pre-aged radiocarbon values ranging from –469 to –407‰. In the brackish zone of Bangdu Bay, distinctive spectroscopic signals, elevated DOC concentrations, and enriched Δ¹⁴C-DOC values along the groundwater-coastal continuum indicate active recirculation of coastal water. Overall, our findings demonstrate that SGD not only delivers aged terrestrial DOC to coastal waters but is also modified by coastal processes, underscoring its potential significant yet complex role under the influence of coastal pollutions.

1 Introduction

Submarine groundwater discharge (SGD) is an important pathway for the transport of water, organic matter, nutrients, and trace elements to coastal oceans (Burnett et al., 2006; Luijendijk et al., 2020; Zhu et al., 2025). Globally, approximately 44 million km³ of groundwater is estimated to be stored within the upper ~10 km of the continental crust (Luijendijk et al., 2020; Ferguson et al., 2021). SGD consists of both fresh groundwater and recirculated seawater, and biogeochemical reactions within the subterranean estuary modifying the chemical composition of the discharging water (Santos et al., 2021). Although fresh SGD to the ocean accounts for only ~1% of the annual river discharge (Taniguchi et al., 2019), the total SGD-derived nutrient fluxes can exceed river inputs (Santos et al., 2021). However, most previous investigations do not distinguish between fresh and saline SGD, despite substantial differences in their nutrient sources, residence times, and biogeochemical signatures (Santos et al., 2021). Therefore, differentiating the chemical characteristics of fresh and saline SGD is essential for understanding mixing processes and nutrient transformation within the subterranean estuary.

SGD has also been proposed as a key component of coastal carbon budgets (Moore, 2010), especially as a recent study have highlighted its role in delivering dissolved organic matter (DOM) and CO₂ to diverse coastal ecosystems (Tomer et al., 2025). Due to prolonged anoxic conditions and limited photodegradation, the degradation of DOM in groundwater systems differs from that in marine or riverine systems (Abarike et al., 2024; Moore et al., 2024). Dissolved organic carbon (DOC), which constitute the major fraction of DOM, plays a critical role in this process. McDonough et al. (2022) estimated that up to 13 Tg of photolabile and biolabile DOC is released from the groundwater systems and can be rapidly degraded in the ocean. Additionally, coastal zones are often characterized by high biological productivity driven by heavy nutrient loading (Oh et al., 2023). Overall, the nutrients and DOM supplied via SGD play a significant role in the coastal ocean carbon cycle.

Jeju Island, located off the southern coast of the Korean Peninsula, provides an ideal natural setting for studying SGD (Figure 1). The island hosts ~1000 artesian springs and wells along its coastline, with fresh groundwater supplying about 90% of the island’s total water resources. Due to its volcanic geology, Jeju Island is characterized by high permeability that facilitating SGD discharge (Kim et al., 2003). Furthermore, Jeju Island has minimal sustained river flow, making SGD a major source of nutrients and DOM to adjacent coastal waters (Hwang et al., 2005; Kim et al., 2011; Kim and Kim, 2017; Kim et al., 2022a). Spatial variability in water optical properties across Jeju Island likely reflect contrasting bedrock and hydrological conditions (Hahn et al., 1997). In the west, coastal groundwater flows directly to the sea through the basal-parabasal zone and is overlain by low-permeability layers (Kwon et al., 2021). In contrast, the east features a more complex aquifer system, where basal groundwater mixes with saline water due to the Seogwipo Formation’s low-permeability bedrock being overlain by permeable volcanic rocks, facilitating seawater intrusion (Lee et al., 2023). Therefore, on the eastern coast of Jeju, almost all groundwater discharge is attributed to recirculating seawater, while fresh groundwater contributes about 25% of the total submarine SGD on the western coast of Jeju (Kim et al., 2011).

Figure 1

Figure 1. (A) Geographic location of Jeju Island. The dash arrow indicates the Changjiang diluted water and solid black arrows indicate the Tsushima Warm Current (modified from Bai et al., 2014). (B) Map of Jeju Island with sampling sites indicated. (C, D) Enlarged figures of the sampling sites (RW, Rainwater; FGW, Fresh Groundwater; SW, Seawater; and SGW, Saline Groundwater).

Understanding the processes and chemical characteristics of DOC in brackish water is thus essential for evaluating the role of SGD in the coastal system. The average DOC concentration in groundwater on Jeju Island (35 ± 13 μM, Song et al., 2018; Kim et al., 2022b) is slightly lower than that in the deep Pacific Ocean (<39 μM, >1000 m water depth; Druffel et al., 1996) and the deep East/Japan Sea (<50 μM, >1000 m water depth; Ryu et al., 2023). Radiocarbon (¹⁴C hereafter) in DOM provide valuable insights into the coastal carbon cycle. However, only a limited number of studies have measured ¹⁴C in groundwater DOC globally (Wassenaar et al., 1991; Downing and Striegl, 2018; Thomas et al., 2021; McDonough et al., 2022). Most of those studies utilized Δ¹⁴C-DOC (where Δ¹⁴C represents the fractionation-corrected ¹⁴C/¹²C ratio relative to a standard: Broecker et al., 1959; Stuiver and Polach, 1977) to determine the age of groundwater DOC (Wassenaar et al., 1991). A few studies have reported Δ¹⁴C-DOC values in brackish waters, with the results showing significant spatial variation, ranging from an average of –176 ± 102‰ in Korean estuaries (Kang et al., 2024) to 133 ± 74‰ in the York River (Raymond and Bauer, 2001). Notably, Kang et al. (2024) suggested that aged DOC may be exported through groundwater discharge from the Seomjin Estuary in South Korea.

To improve our understanding of submarine groundwater system in Jeju Island, we examined the ¹⁴C age and molecular transformation of DOM in the SGD-coastal continuum of Jeju Island. This study presents the first Δ¹⁴C-DOC measurements from groundwater on Jeju Island. Here we measured DOC concentration, chromophoric dissolved organic matter (CDOM), fluorescent dissolved organic matter (FDOM), and Δ¹⁴C-DOC values in samples of Fresh Groundwater (FGW), Saline Groundwater (SGW) and Seawater (SW) collected from both the western and eastern coast of Jeju Island. The dataset generated in this study contributes to more comprehensive understanding of the role of groundwater in coastal carbon cycling and its potential modification by coastal waters.

2 Materials and methods

2.1 Study site and sample collection

Jeju Island is a dormant volcanic island with an area of ~1830 km², located in the southern sea of Korea (Figure 1). The island is primarily composed of permeable basaltic rocks formed by Cenozoic volcanism and includes the Mountain Hallasan shield volcano, which has an elevation of 1950 m. The volcanic bedrock allows rainwater to rapidly penetrate and recharge the groundwater, which is then transported to the coastal region through an aquifer system (Fenta et al., 2020). Jeju Island experiences an East Asian monsoon climate, with most precipitation occurring in the summer (annual rainfall: 1440–1690 mm; Kim et al., 2013).

Details of the sample collection are provided in Kim et al. (2013) and Kim et al. (2022). Briefly, sampling campaigns for FGW, SGW, and SW were conducted in the Hwasun area (southwestern Jeju Island) in April 2023, and in Bangdu Bay (eastern Jeju Island) in September 2023 (Figure 1; Supplementary Table S1). FGW sample from Hwasun area was collected from coastal artesian wells using HDPE bottles, whereas FGW sample from Bangdu Bay was taken directly from a spring outlet located inside the bay. SGW samples were obtained from shallow pits (~50 cm deep) dug into beach sediments. The first two volumes of seeping groundwater were gently discarded using a plastic beaker, and freshly recharged water was collected. A precipitation sample was collected during rainfall in June 2023 from the rooftop of a building at Jeju National University, located near the center region of Jeju Island, approximately 33km from both sampling sites (Figure 1). To minimize the influence of dry deposition, the rainwater sample was taken from the center of the roof, two hours after rainfall began. All samples were collected in Nalgene HDPE bottles pre-cleaned with 10% hydrochloric acid. In total, 7 samples were collected for radiocarbon analysis. Salinity and Oxidation-Reduction Potential (ORP) were measured on-site using a portable YSI Pro Plus sensor (YSI Inc., OH, USA). The measurement accuracies for salinity and ORP were ±0.1 PSU and ± 20 mV, respectively.

2.2 Nutrients analysis

Water samples were vacuum filtered through pre-combusted (550 °C for 4 hours) Whatman GF/F filters (pore size: 0.7 µm; Whatman Inc., NJ, USA) within 1 hour of each sampling campaign. Concentrations of dissolved inorganic nutrients (NO³⁻, NO²⁻, NH⁴⁺, PO₄³⁻, and Si(OH)₄) were analyzed photometrically using an auto-analyzer (New QuAAtro39, SEAL Analytical, Southampton, UK). Accuracy was verified prior to sample analysis using certified reference materials (Lots CO, CB, and BZ; KANSO Technos Co., Ltd. Japan), and the measured values were consistent with the certified values (within 2%). The detection limits for NO³⁻, NO²⁻, NH⁴⁺, PO₄³⁻, and Si(OH)₄ were 0.02, 0.04, 0.08, 0.05, and 0.06 µM, respectively.

2.3 DOC and TDN concentrations and Δ¹⁴C-DOC analysis

For DOC concentration and radiocarbon analysis, samples were filtered through pre-combusted (500 °C for 5 hours) GF/F filters and transferred to pre-combusted, 1 L Amber Boston Round glass bottles (Fisher Scientific, PA, USA), acidified to pH ~2 using 6 M HCl (Walker et al., 2017) and stored at room temperature. A previous study found no difference in radiocarbon results between acidified and frozen (at –20 °C) samples (Oh et al., 2025). DOC concentrations were determined using high-temperature catalytic oxidation (HTCO) at 680 °C with a total organic carbon analyzer (TOC-L, Shimadzu, Japan) equipped with an ASI-L auto-sampler. Prior to analysis, the system baseline was stabilized with carbon-free distilled water until the signal remained consistently below the detection limit (< 2 μM for DOC and< 3 μM for TDN). Accuracy was verified using deep-sea reference samples (DSR; 41–42 μM for DOC, University of Miami), with results in good agreement with consensus values (within 2%).

For radiocarbon analysis, DOC was converted to CO₂ using high-energy UV oxidization. Each sample was introduced into a quartz reactor, and sparged with high-purity N₂ gas (99.999%) at a steady flow rate for approximately 60 minutes. The samples were then irradiated with UV light (1200 W) for 6 hours. The resulting CO₂ was recovered by continued sparging with N₂ gas, cryogenically purified, quantified by pressure measurement, and sealed in Pyrex tubes. Sample tubes were sent to the Xi’an AMS Center of Institute of Earth Environment, Chinese Academy of Sciences. CO₂ gas samples graphitized offline and radiocarbon measurements were performed on the Mini Carbon Dating System (MICADAS, Ionplus AG) at the Xi’an AMS Center of Institute of Earth Environment, Chinese Academy of Sciences, with a reported precision of 0.2–0.3% and an accuracy of 3–4‰, based on long-term standards (Zhou et al., 2007; Fu et al., 2015). Blanks were determined by oxidizing ultrapure water. To obtain sufficient carbon for blank analysis, three blanks combined for one measurement. All samples were corrected using a process blank (n=9), which contained 11 ± 9 μg C with a Δ¹⁴C value of –326 ± 139‰ (n=3). The average difference between the original and blank-corrected Δ¹⁴C values was 5 ± 11‰. Notably, the associated uncertainty of fresh groundwater was highly variable due to error propagation (Figure 2A). Conventional ¹⁴C age is calculated with the half-life of 5700 years (Godwin, 1962).

Figure 2

Figure 2. (A) Relationship between dissolved organic carbon (DOC) concentration and Δ¹⁴C-DOC values. Solid and open symbols represent Hwasun Bay and Bandu Bay, respectively, with the trend line indicating the linear fit. (B) Δ¹⁴C-DOC values of water from the Yangtze River (YT) and its estuary (YTE, Qi et al., 2020), the Yellow River (YR) and its estuary (YRE, Qi et al., 2020), the Geum Estuary (GE, Kang et al., 2024), the East China Sea (ECS, Han et al., 2022), precipitation in Seoul (SP, Cha et al., 2020), coastal water from Jeju Island (Jeju, Han et al., 2022), the Seomjin Estuary (SE, Kang et al., 2024), the East/Japan Sea (EJS, Ryu et al., 2023), and the North Pacific (Druffel et al., 1996). Error bar is given in gray lines.

A two-endmember ¹⁴C mass balance was used to trace the contributions of aged and recently produced DOC, using the following equations:

$\begin{array}{ll}{[DOC]}_{sample}={[DOC]}_{background}+{[DOC]}_{exce} & (1)\end{array}$

$\begin{array}{ll}{\Delta }^{14}{C}_{sample}\times {[DOC]}_{sample}={\Delta }^{14}{C}_{background}\times {[DOC]}_{background}+{\Delta }^{14}{C}_{exce}\times {[DOC]}_{exce}& (2)\end{array}$

where [DOC]_sample, [DOC]_background, and [DOC]_exce represent DOC concentrations in sample, the background aged DOC, and added DOC, respectively. Δ¹⁴C values correspond to the radiocarbon measurements for each pool.

2.4 CDOM and FDOM analysis

Samples for CDOM and FDOM analysis were filtered through pre-rinsed polycarbonate filters (Nuclepore™, Whatman Inc., NJ, USA). Absorbance spectra and excitation-emission matrix (EEM) fluorescence were measured on the sampling day using a spectrofluorometer (Aqualog, HORIBA Jobin Yvon, NJ, USA). CDOM absorbance was converted to the Napierian absorption coefficient at 350 nm (a₃₅₀) following Helms et al. (2008), serves as a proxy for CDOM concentration. Spectral slope coefficients for 250–600 nm (S_250–600, nm⁻¹), 275–295 nm (S_275–295), and 350–400 nm (S_350–400) were derived from the absorption spectra by fitting them to an exponential decay function. The slope ratio (S_R) was calculated as the ratio of the slope in the shorter wavelength region (275–295 nm) to that in the longer wavelength region (350–400 nm). Specific UV absorbance at 254 nm (SUVA₂₅₄, unit: L mg C⁻¹ m⁻¹) was determined by normalizing UV absorbance at 254 nm to the DOC concentration.

Fluorescence EEMs were recorded across excitation wavelengths from 250–600 nm with emission at 5 nm intervals. A PARAFAC (PARAllel FACtor) model was applied to four EEM datasets using the Solo+MIA software (Eigenvector Research Inc.,WA, USA) and validated through split-half analysis and core consistency diagnostics (Bro and Kiers 2003). The fluorescence index (FI) was calculated as the ratio of emission intensities at 470 nm and 520 nm at an excitation wavelength of 370 nm (Cory and McKnight, 2005). The humification index (HIX) was determined as the ratio of the summed emission intensities from 435–480 nm to the sum of 300–345 nm and 435–480 nm at an excitation of 254 nm (Ohno, 2002). The biological index (BIX) was calculated as the ratio of emission intensities at 380 nm and 430 nm, with an excitation wavelength of 310 nm (Huguet et al., 2009).

3 Results and discussions

3.1 Aged DOC supply coastal water from fresh groundwater

In both the Hwasun and Bangdu areas, the highest DOC concentrations were observed in seawater (54 and 123 μM, respectively), while the lowest were found in fresh groundwater (5 ± 3 μM, n=2; Figure 2A). A linear correlation was found between Δ¹⁴C-DOC and DOC concentration (R² = 0.79). These results indicate that the DOC concentration and Δ¹⁴C values across the SGD-coastal continuum are influenced by mixing between recently produced DOC in seawater and aged DOC from groundwater (Figure 2A). Corresponding to the low DOC concentrations, the most depleted Δ¹⁴C-DOC values (–469 ± 131‰ and –407 ± 87‰) were also observed in fresh groundwater. These values are more depleted than those recorded for the oldest seawater in the Yellow River and East China Sea (Figure 2B). In seawater, Δ¹⁴C-DOC ranged from –346 to –105‰ across the two sites, with higher DOC concentrations and Δ¹⁴C values observed in Bandu Bay. A two-endmember ¹⁴C mass balance calculation based on Equations 1 and 2 indicates that the added DOC has a modern ¹⁴C of ~85‰ (54μM×–346‰+69μM×X =123μM×–105‰, X = 85‰). This estimated value falls within the reported Δ¹⁴C-DIC range of 36–132‰ (Ryu et al., 2023), suggesting that the added DOC signal resembles new primary production. This implies that the differences in DOC concentration and Δ¹⁴C values between the sites may be driven by variations in primary production, a conclusion further supported by the observed CDOM distribution pattern (discussed in the following section). Previous studies have also shown that DOC concentrations are higher in summer than spring in Jeju coastal waters due to microbial activity (Song et al., 2023). Recently produced particulate organic matter has been decomposed to DOC during the summer. Therefore, the differences in DOC and Δ¹⁴C-DOC values of seawater between the two sites likely reflect seasonal variations. Notably, DOC concentrations in fresh groundwater were generally lower than those in the open ocean, and these values deviated significantly from the regression line of the Keeling plot (Supplementary Figure S1).

In the East/Japan Sea, the lowest Δ¹⁴C-DOC value was observed at subsurface depths (–396 ± 31‰; Ryu et al., 2023), whereas in the open ocean, the most depleted values typically occur at greater depths. The subsurface water mass with depleted DOC displayed characteristics of the Tsushima Warm Current (TWC), which originates in the Northwest Pacific (Figure 1). However, the relatively homogeneous Δ¹⁴C-DOC signal during photochemical oxidation this region differs from that reported for the North Pacific (Ryu et al., 2023), suggesting the potential contribution of other aged DOC sources.

The Δ¹⁴C-DOC values from large rivers, such as the Yellow and Yangtze Rivers show wider variation (–227 to –57‰; Xue et al., 2017; Qi et al., 2020), whereas DOC discharged from major Korean rivers predominantly contains relatively young DOC (–124 to 0.8‰; Lee et al., 2021). During the summer monsoon, slightly more depleted Δ¹⁴C-DOC values were observed in precipitation over Seoul, South Korea, ranging from –321 to –106‰ (Cha et al., 2020). Consistent with these findings, our summer precipitation value (–297‰) fell within this range. Overall, the Δ¹⁴C-DOC values from riverine and precipitation are higher than those observed in subsurface waters of the East/Japan Sea (Figure 2B), suggesting that groundwater may be a minor but potential source of the depleted DOC.

Despite the remarkably high seepage rate, the overall SGD flux (1.6 × 10¹⁰ m³ yr^–1, Kim et al., 2003) is relative low due to its limited spatial extent especially when compared to SGD inputs from the Yellow Sea (1–6.7×10¹¹ m³ yr^–1; Kim et al., 2005) and continental margins of Korea, Russia, and Japan, which have extensive coastlines (Kim et al., 2005; Jung and Yoon, 2025). Accordingly, this rapid groundwater turnover suggests younger DOC (Koh et al., 2005), yet our measurements indicate Jeju exports relatively older, refractory DOC (~6294 to 7431years) compared to surrounding seawater. DOC fluxes are typically estimated based on DOC concentrations and SGD fluxes (Burnett et al., 2006; Webb et al., 2019). The SGD fluxes into Hwasun Bay and Bangdu Bay were estimated to be 8.0 × 10⁸ m³ yr^–1 and 5.1 × 10⁷ m³ yr^–1, respectively (Kim et al., 2011). Previous studies indicate that on the eastern coast of Jeju (including Bangdu Bay) all groundwater discharge is attributed to recirculating seawater (Kim et al., 2003, 2011). Using the DOC concentration in saline groundwater of 71 μM, the estimated DOC discharge from Bangdu Bay SGD is approximately of 4.3× 10⁷ g yr^-1 (71 μM×5.1 × 10⁷ m³ yr^–1). In contrast, at Hwasun Bay, 25% of SGD is attributed to fresh SGD (Kim et al., 2011). The estimated fresh and saline SGD-derived DOC discharge from Hwasun Bay are 0.5–1.9× 10⁷ g yr^-1 and 2.2× 10⁸ g yr^-1, respectively. Although the DOC flux associated with fresh SGD is smaller than that from saline SGD, it may represent an important source of pre-aged DOC to the coastal ocean. Further evidence from Obama Bay on the western coast of Japan also highlight the role of groundwater inflow in supplying DOC to the East/Japan Sea (Cabral et al., 2023). In that region, average DOC concentrations in groundwater (57 ± 13 to 63 ± 61 μM, Cabral et al., 2023) were higher than those at our study sites. If similarly depleted in radiocarbon, this groundwater may represent one of the unknown sources of aged DOC to the subsurface waters of the East/Japan Sea.

3.2 Regional dynamics of brackish water DOM on Jeju Island

Four FDOM components (C1–C4) were identified using the PARAFAC model, based on comparisons with previous studies in the OpenFluor database (Supplementary Figure S2, Supplementary Tables S2, S3; Murphy et al., 2014). Among these, one component was protein-like (C3) and three were humic-like (C1, C2, and C4), as defined by Coble (2007). The optical characteristics of each FDOM component were interpreted based on the location of their fluorescence peaks and literature characterizations (Supplementary Table S2). C1 (Max_ex/em = 250/405 nm) and C2 (Max_ex/em = 250(370)/465 nm) were classified as terrestrial humic-like components. A strong correlation between C1 and C2 (R² = 0.987) indicates that they are likely to share similar sources and transformation processes. The fluorescence of C3 (Max_ex/em = 270/305 nm) aligned with that of the amino acid tyrosine, and corresponds to the B peak (Coble, 2007). The spectral signature of C4 (Max_ex/em = 285/350 nm) resembled the marine humic-like of M peak (Coble, 2007).

Analysis of spectral slope and absorption coefficients revealed two distinct subsets based on sampling location (Figure 3). In the Hwasun region, a₃₅₀ values increased from 0.009 m⁻¹ to 0.050 m⁻¹ due to seawater mixing in the subterranean estuary, while S_260–600 values decreased (Supplementary Table S2). This suggests the introduction of marine origin CDOM with higher molecular weight organic matter via seawater mixing (Figure 3). FGW from Hwasun exhibited the lowest DOC concentration (8 μM), the highest S_260–600 value (0.066 nm⁻¹), and the lowest a₃₅₀ value (0.009 m⁻¹) (Figure 3), indicating a reduction in molecular weight likely caused by microbial degradation during aquifer recharge and transit. The depleted Δ¹⁴C-DOC values further support potential biodegradation fractionation, aligning with the rapid turnover of the groundwater system on Jeju Island (Koh et al., 2005). Moreover, elevated higher humic-like FDOM and a high HIX (3.22) in FGW from Hwasun Bay suggest that highly decomposed organic matter significantly contributes to the DOC pool. FI values for FGW, SGW, and SW further indicate a dominant input from terrestrial organic matter.

Figure 3

Figure 3. Various parameters of water samples from Hwasun Bay and Bangdu Bay, Jeju Island. (A) DOC concentrations and Δ¹⁴C-DOC values. (B) Humic-like fluorescence dissolved organic matter (FDOM) and protein-like FDOM. (C) Fluorescence index (FI), biological index (BIX), and humification index (HIX). (D) Absorption at 350 nm (a₃₅₀) and spectral slope (S_260–600) values.

In contrast, absorption parameters in Bangdu Bay were markedly different. The a₃₅₀ values (1.896 ± 0.617 m⁻¹) were substantially higher than those in the Hwasun region (0.029 ± 0.021 m⁻¹), whereas S_260–600 values were lower (0.013 ± 0.001 nm⁻¹) than in the Hwasun area (0.048 ± 0.016 nm⁻¹). The Bangdu region also exhibited higher BIX (avg: 1.01 ± 0.08) and FI (avg: 1.62 ± 0.19) and lower HIX (avg: 1.89 ± 1.26), indicating greater contribution from recently produced, high molecular weight organic matter. Furthermore, the similarity in S_260–600 values across FGW, SGW, and SW (0.012–0.014 nm⁻¹) suggests well-mixed conditions within the coastal aquifers of Bangdu Bay.

The findings of this study confirm that, in Jeju Island’s coastal aquifers, the characteristics of organic matter in brackish groundwater vary significantly depending on the groundwater properties. The geological features of Bangdu Bay contribute to the presence of brackish groundwater in coastal artesian springs, indicating a strong seawater influence (Figure 4). Additionally, the region’s DOC levels may be impacted by algae (e.g., Ulva spp.) that accumulates along the coast of Jeju Island (Kwon et al., 2017). The organic matter from these algae likely enters the aquifer, resulting in similar molecular weights between groundwater and seawater. At Sinyang Beach in Bangdu Bay, where the highest a₃₅₀ value (2.580 m⁻¹) was recorded, the SGW had an ORP of −34.8 mV, suggesting severe organic pollution. Certain areas of Sinyang Beach (Bangdu region) experience seaweed accumulation and significant organic pollution in porewater. Elevated ammonium levels in SGW further suggest anoxic degradation of organic materials and anthropogenic influences (Supplementary Figure S3), such as nearby aquafarms (Samanta et al., 2019), contaminated groundwater inflow (Cho et al., 2019), and reduced circulation due to breakwaters (Choi et al., 2023). This is further supported by the Δ¹⁴C-DOC values of coastal seawater. At Hwasun, Δ¹⁴C-DOC values in seawater were slightly higher (~18‰) than in saline SGW, whereas at Bangdu, coastal seawater exhibited higher Δ¹⁴C-DOC values (~52‰) compared to saline SGW, possibly reflecting the influence of algal blooms (Figure 4).

Figure 4

Figure 4. A conceptual diagram illustrating the differences between Hwasun Bay and Bangdu Bay areas. In Bangdu Bay, submarine groundwater discharge is significantly influenced by seawater and seaweed blooms.

Seasonal variations in primary production and microbial activity likely account for the observed differences in coastal seawater between the two sites. Fresh SGD shows low DOC concentrations and aged Δ¹⁴C-DOC values but lowest a₃₅₀ value in Hwasun Bay, indicating a predominantly terrestrial source. In contrast, saline SGD reflects a strong marine influence, particularly in Bangdu Bay. Nutrient concentrations decrease with increasing salinity, whereas DOC concentrations increase (Supplementary Figure S3). The different patterns of nutrients and DOC with salinity likely reflect contrasting biogeochemical process: nutrients primarily result from terrestrial inputs, weathering, and the accumulation of remineralized organic matter, whereas DOC in saline SGD is predominantly of marine origin, especially in Bangdu Bay. The distinct chemical characteristics of brackish waters further highlight the contrasting hydrogeological settings of eastern and western Jeju Island. A larger proportion of recirculated seawater contributed to the SGD on the eastern side of Jeju Island. Due to the highly porous basaltic bedrock and coastal sediments, mixing between fresh groundwater and seawater occurs relatively rapidly. Our results suggest that in situ biogeochemical reactions within the subterranean estuary of the Jeju coast are not actively occurring, which likely explains the observed low DOC concentrations. These findings indicate that SGD is not only exports aged DOM and nutrients to the coastal ocean but is also modified by recirculated seawater, particularly along the eastern coast.

4 Conclusion and insights

Previous studies have shown that groundwater discharge plays a crucial role in supplying nutrients and DOM to coastal waters, thereby potentially stimulating biological production. This study demonstrates that groundwater not only serves as a potential source of pre-aged DOC to the adjacent ocean but is also modified by seawater intrusion, particularly in the eastern part of Jeju Island. High concentrations of humic-like FDOM and depleted Δ¹⁴C-DOC values in FGW suggest that the discharged DOC is highly degraded and may be preserved as it is transported to the open ocean. In contrast, the Bangdu area, which exhibits highly humic FDOM linked to recently produced DOC, indicates the influence of algal blooms. This seawater recirculation underscores the vulnerability of groundwater system to coastal influence and reinforces the importance of sustained monitoring efforts. However, given the diverse characteristics across Jeju Island, these findings may not fully represent the entire island system. Therefore, more extensive sampling of SGD across different depths and regions is required to better constrain the characteristics and dynamics of groundwater–seawater interactions.

Data availability statement

Data are all available in this paper: Fang, Ling; Kim, Jeonghyun; Kim, Minkyoung (2025), “Aged Dissolved Organic Carbon Supply through Submarine Groundwater Discharge in Jeju Island, South Korea”, Mendeley Data, V2, doi: 10.17632/zf7t7cryzv.2.

Author contributions

LF: Data curation, Methodology, Writing – original draft. HK: Data curation, Methodology, Writing – original draft. JK: Data curation, Funding acquisition, Writing – review & editing. PC: Methodology, Writing – review & editing. YR: Software, Writing – original draft. GK: Conceptualization, Writing – review & editing. MK: Conceptualization, Funding acquisition, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This research was supported by the Republic of Korea (MSIT) and the National Research Foundation of Korea (NRF- 20230206, FY2025), Korea Institute of Marine Science & Technology Promotion (KIMST) funded by the Ministry of Oceans and Fisheries (20220533), and Dongil Culture and Scholarship Foundation. MK was funded by the Research Program for the carbon cycle between oceans, land, and atmosphere of the NRF funded by the Ministry of Science and ICT (2022M3I6A1085990), NRF grant funded by the Korea government (MSIT) (2022R1C1C1002824), by the National Research Foundation of Korea (NRF) grant funded by the Korea government. (MSIT) (RS-2025-02263830), and by Global-Learning & Academic research institution for Master’s·PhD students, and Postdocs (LAMP) Program of NRF grant funded by the Ministry of Education (No. RS-2023-00301914). HK and JK were supported by NRF funded by the Korean government (NRF-2021R1C1C1004733) and by Regional Innovation Strategy (RIS) through NRF funded by MOE (2023RIS-009). This research was supported by the KIMST funded by the Ministry of Oceans and Fisheries (RS-2025-02307311) and by the Regional Innovation System & Education (RISE) program through the Jeju RISE center, funded by the Ministry of Education (MOE) and the Jeju Special Self-Governing Province, Republic of Korea (2025-RISE-17-001).

Acknowledgments

We would like to thank SangAh Ji and Jinjoo Lim in the Marine Biogeochemistry Laboratory of Jeju National University for helping with sampling campaigns.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmars.2025.1719001/full#supplementary-material

References

Abarike G. A., Brick S., Engelen B., and Niggemann J. (2024). Different dissolved organic matter sources sustain microbial life in a sandy beach subterranean estuary - an incubation study. Front. Mar. Sci. 11. doi: 10.3389/fmars.2024.1501781

Crossref Full Text | Google Scholar

Bai Y., He X., Pan D., Chen C. T. A., Kang Y., Chen X., et al. (2014). Summertime Changjiang River plume variation during 1998–2010. J. Geophysical Research: Oceans 119, 6238–6257. doi: 10.1002/2014JC009866

Crossref Full Text | Google Scholar

Bro R. and Kiers H. A. (2003). A new efficient method for determining the number of components in PARAFAC models. J. Chemom. 17, 274–286. doi: 10.1038/1831582a0

Crossref Full Text | Google Scholar

Broecker W. S., Olson E. A., and Bird J. (1959). Radiocarbon measurements on samples of known age. Nature 183, 1582–1584. doi: 10.1038/1831582a0

Crossref Full Text | Google Scholar

Burnett W. C., Aggarwal P. K., Aureli A., Bokuniewicz H., Cable J. E., Charette M. A., et al. (2006). Quantifying submarine groundwater discharge in the coastal zone via multiple methods. Sci. Total Environ. 367, 498–543. doi: 10.1016/j.scitotenv.2006.05.009

PubMed Abstract | Crossref Full Text | Google Scholar

Cabral A., Sugimoto R., Taniguchi M., Tait D., Nakajima T., Honda H., et al. (2023). Fresh and saline submarine groundwater discharge as sources of carbon and nutrients to the Japan Sea. Mar. Chem. 249, 1–11. doi: 10.1016/j.marchem.2023.104209

Crossref Full Text | Google Scholar

Cha J. Y., Lee S. C., Lee E. J., Go M., Dasari K. B., Yim Y. H., et al. (2020). High dissolved organic radiocarbon in precipitation during winter and its implication on the carbon cycle. Sci. Total Environ. 742, 1–7. doi: 10.1016/j.scitotenv.2020.140246

PubMed Abstract | Crossref Full Text | Google Scholar

Cho H. M., Kim G., and Shin K. H. (2019). Tracing nitrogen sources fueling coastal green tides off a volcanic island using radon and nitrogen isotopic tracers. Sci. Total Environ. 665, 913–919. doi: 10.1016/j.scitotenv.2019.02.212

PubMed Abstract | Crossref Full Text | Google Scholar

Choi Y., Yang K., Lee M. Y., Youn S. H., Son M., Park S. R., et al. (2023). Factors controlling massive green tide blooms on the coasts of Jeju Island, Korea. Mar. pollut. Bull. 186, 114446. doi: 10.1016/j.marpolbul.2022.114446

PubMed Abstract | Crossref Full Text | Google Scholar

Coble P. G. (2007). Marine optical biogeochemistry: The chemistry of ocean color. Chem. Rev. 107, 402–418. doi: 10.1021/cr050350+

PubMed Abstract | Crossref Full Text | Google Scholar

Cory R. M. and McKnight D. M. (2005). Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. Technol. 39, 8142–8149. doi: 10.1021/es0506962

PubMed Abstract | Crossref Full Text | Google Scholar

Downing J. A. and Striegl R. G. (2018). Size, age, renewal, and discharge of groundwater carbon. Inland Waters 8, 122–127. doi: 10.1080/20442041.2017.1412918

Crossref Full Text | Google Scholar

Druffel E. R., Bauer J. E., Williams P. M., Griffin S., and Wolgast D. (1996). Seasonal variability of particulate organic radiocarbon in the northeast Pacific Ocean. J. Geophysical Research: Oceans 101, 20543–20552. doi: 10.1029/96JC01850

Crossref Full Text | Google Scholar

Fenta M. C., Anteneh Z. L., Szanyi J., and Walker D. (2020). Hydrogeological framework of the volcanic aquifers and groundwater quality in Dangila Town and the surrounding area, Northwest Ethiopia. Groundwater Sustain. Dev. 11, 100408. doi: 10.1016/j.gsd.2020.100408

Crossref Full Text | Google Scholar

Ferguson G., McIntosh J. C., Warr O., Sherwood Lollar B., Ballentine C. J., Famiglietti J. S., et al. (2021). Crustal groundwater volumes greater than previously thought. Geophysical Res. Lett. 48, e2021GL093549. doi: 10.1029/2021GL093549

Crossref Full Text | Google Scholar

Fu Y.-C., Zhou W.-J., Du H., Cheng P., Zhao X.-L., Liu Q., et al. (2015). A preliminary study of small-mass radiocarbon sample measurement at Xi’an-AMS. Chin. Phys. C 39, 036202. doi: 10.48550/arXiv.1409.0633

Crossref Full Text | Google Scholar

Godwin H. (1962). Half-life of radiocarbon. Nature 195, 984. doi: 10.1038/195984a0

Crossref Full Text | Google Scholar

Han H., Na T., Cho H. M., Kim G., and Hwang J. (2022). Large fluxes of continental-shelf-borne dissolved organic carbon in the East China Sea and the Yellow Sea. Mar. Chem. 240, 104097.1–9. doi: 10.1007/s002540050223

Crossref Full Text | Google Scholar

Hahn J., Lee Y., Kim N., Hahn C., and Lee S. (1997). The groundwater resources and sustainable yield of Cheju volcanic island, Korea. Environ. Geology 33, 43–53. doi: 10.1007/s002540050223

Crossref Full Text | Google Scholar

Huguet A., Vacher L., Relexans S., Saubusse S., Froidefond J. M., and Parlanti E. (2009). Properties of fluorescent dissolved organic matter in the Gironde Estuary. Organic Geochemistry 40, 706–719. doi: 10.1016/j.orggeochem.2009.03.002

Crossref Full Text | Google Scholar

Hwang D. W., Lee Y. W., and Kim G. (2005). Large submarine groundwater discharge and benthic eutrophication in Bangdu Bay on volcanic Jeju Island, Korea. Limnology Oceanography 50, 1393–1403. doi: 10.4319/lo.2005.50.5.1393

Crossref Full Text | Google Scholar

Jung H. C. and Yoon Y. (2025). Climate change effects on submarine groundwater discharge and regional variations along the Korean Peninsula. Commun. Earth Environ. 6, 1–9. doi: 10.1038/s43247-025-02084-9

Crossref Full Text | Google Scholar

Kang S., Kim J.-H., Ryu J.-S., and Shin K.-H. (2024). Export of aged dissolved organic carbon to the Geum and Seomjin estuaries in South Korea. J. Hydrology 632, 130659. doi: 10.1016/j.jhydrol.2024.130659

Crossref Full Text | Google Scholar

Kim G., Kim J.-S., and Hwang D.-W. (2011). Submarine groundwater discharge from oceanic islands standing in oligotrophic oceans: Implications for global biological production and organic carbon fluxes. Limnology Oceanography 56, 673–682. doi: 10.4319/lo.2011.56.2.0673

Crossref Full Text | Google Scholar

Kim T.-H., Kwon E., Kim I., Lee S.-A., and Kim G. (2013). Dissolved organic matter in the subterranean estuary of a volcanic island, Jeju: Importance of dissolved organic nitrogen fluxes to the ocean. J. sea Res. 78, 18–24. doi: 10.1016/j.seares.2012.12.009

Crossref Full Text | Google Scholar

Kim G., Lee K. K., Park K. S., Hwang D. W., and Yang H. S. (2003). Large submarine groundwater discharge (SGD) from a volcanic island. Geophysical Res. Lett. 30, 1–4. doi: 10.1029/2003gl018378

Crossref Full Text | Google Scholar

Kim G., Ryu J. W., Yang H. S., and Yun S. T. (2005). Submarine groundwater discharge (SGD) into the Yellow Sea revealed by ²²⁸Ra and ²²⁶Ra isotopes: Implications for global silicate fluxes. Earth Planetary Sci. Lett. 237, 156–166. doi: 10.1016/j.epsl.2005.06.011

Crossref Full Text | Google Scholar

Kim J., Song B.-C., Lee M.-Y., and Kim T.-H. (2022a). Monthly variation in flux of inorganic nutrients from submarine groundwater discharge in a volcanic island: Significant nitrogen contamination in groundwater. Front. Mar. Sci. 9. doi: 10.3389/fmars.2022.835207

Crossref Full Text | Google Scholar

Kim J. and Kim G. (2017). Inputs of humic fluorescent dissolved organic matter via submarine groundwater discharge to coastal waters off a volcanic island (Jeju, Korea). Sci. Rep. 7. doi: 10.1038/s41598-017-08518-5

PubMed Abstract | Crossref Full Text | Google Scholar

Kim J., Song B.-C., and Kim T.-H. (2022b). Origin of dissolved organic carbon under phosphorus-limited coastal-bay conditions revealed by fluorescent dissolved organic matter. Front. Mar. Sci. 9. doi: 10.3389/fmars.2022.971550

Crossref Full Text | Google Scholar

Koh D. C., Chang H. W., Lee K. S., Ko K. S., Kim Y., and Park W. B. (2005). Hydrogeochemistry and environmental isotopes of ground water in Jeju volcanic island, Korea: implications for nitrate contamination. Hydrological Processes 19, 2225–2245. doi: 10.1002/hyp.5672

Crossref Full Text | Google Scholar

Kwon H. K., Kang H., Oh Y. H., Park S. R., and Kim G. (2017). Green tide development associated with submarine groundwater discharge in a coastal harbor, Jeju, Korea. Sci. Rep. 7, 1–9. doi: 10.1038/s41598-017-06711-0

PubMed Abstract | Crossref Full Text | Google Scholar

Kwon E., Park J., Park W. B., Kang B. R., and Woo N. C. (2021). Nitrate contamination of coastal groundwater: Sources and transport mechanisms along a volcanic aquifer. Sci. Total Environ. 768, 1–11. doi: 10.1016/j.scitotenv.2021.145204

PubMed Abstract | Crossref Full Text | Google Scholar

Lee D., Park J., Lee H. A., and Woo N. C. (2023). A numerical approach to evaluating groundwater vulnerability to seawater intrusion on Jeju volcanic island, South Korea. Sustainability 15, 1–15. doi: 10.3390/su15043081

Crossref Full Text | Google Scholar

Lee E. J., Shin Y., Yoo G. Y., Ko E. B., Butman D., Raymond P. A., et al. (2021). Loads and ages of carbon from the five largest rivers in South Korea under Asian monsoon climates. J. Hydrology 599, 1–10. doi: 10.1016/j.jhydrol.2021.126363

Crossref Full Text | Google Scholar

Luijendijk E., Gleeson T., and Moosdorf N. (2020). Fresh groundwater discharge insignificant for the world’s oceans but important for coastal ecosystems. Nat. Commun. 11, 1–12. doi: 10.1038/s41467-020-15064-8

PubMed Abstract | Crossref Full Text | Google Scholar

McDonough L. K., Andersen M. S., Behnke M. I., Rutlidge H., Oudone P., Meredith K., et al. (2022). A new conceptual framework for the transformation of groundwater dissolved organic matter. Nat. Commun. 13, 2153. doi: 10.1038/s41467-022-29711-9

PubMed Abstract | Crossref Full Text | Google Scholar

Moore W. S. (2010). The effect of submarine groundwater discharge on the ocean. Annu. Rev. Mar. Sci. 2, 59–88. doi: 10.1146/annurev-marine-120308-081019

PubMed Abstract | Crossref Full Text | Google Scholar

Moore W. S., Benitez-Nelson C., Schutte C., Moody A., Shiller A., Sibert R. J., et al. (2024). SGD-OD: investigating the potential oxygen demand of submarine groundwater discharge in coastal systems. Sci. Rep. 14, 1–11. doi: 10.1038/s41598-024-59229-7

PubMed Abstract | Crossref Full Text | Google Scholar

Murphy K. R., Bro R., and Stedmon C. A. (2014). Chemometric analysis of organic matter fluorescence. Aquat. organic matter fluorescence 261, 339–375. doi: 10.1017/CBO9781139045452.016

Crossref Full Text | Google Scholar

Oh S., Fang L., Kim J., Yang E. J., Hwang J., and Kim M. (2025). Sources and dynamics of dissolved organic carbon in Kongsfjorden: insights from radiocarbon analysis. Geophysical Res. Lett. 52, 1–10. doi: 10.1029/2025GL114917

Crossref Full Text | Google Scholar

Oh Y. H., Kim J., Kim G., and Kim T.-H. (2023). Groundwater as a source of phosphorus and silicate in an estuarine zone: results from continuous monitoring of nutrients and 222Rn. Front. Mar. Sci. 10. doi: 10.3389/fmars.2023.1162164

Crossref Full Text | Google Scholar

Ohno T. (2002). Fluorescence inner-filtering correction for determining the humification index of dissolved organic matter. Environ. Sci. Technol. 36, 742–746. doi: 10.1021/es0155276

PubMed Abstract | Crossref Full Text | Google Scholar

Qi Y. Z., Fu W. J., Tian J. W., Luo C. L., Shan S., Sun S. W., et al. (2020). Dissolved black carbon is not likely a significant refractory organic carbon pool in rivers and oceans. Nat. Commun. 11, 1–11. doi: 10.1038/s41467-020-18808-8

PubMed Abstract | Crossref Full Text | Google Scholar

Raymond P. A. and Bauer J. E. (2001). DOC cycling in a temperate estuary: a mass balance approach using natural 14C and 13C isotopes. Limnology Oceanography 46, 655–667. doi: 10.4319/lo.2001.46.3.0655

Crossref Full Text | Google Scholar

Ryu Y., Han H., Na T., Kim G., Druffel E. R., and Hwang J. (2023). Transport of aged dissolved organic carbon via the surface current revealed by radiocarbon. Geophysical Res. Lett. 50, e2023GL105296. doi: 10.1029/2023GL105296

Crossref Full Text | Google Scholar

Samanta P., Shin S., Jang S., Song Y. C., Oh S., and Kim J. K. (2019). Stable carbon and nitrogen isotopic characterization and tracing nutrient sources of blooms around Jeju coastal areas. Environ. pollut. 254, 1–10. doi: 10.1016/j.envpol.2019.113033

PubMed Abstract | Crossref Full Text | Google Scholar

Santos I. R., Chen X., Lecher A. L., Sawyer A. H., Moosdorf N., Rodellas V., et al. (2021). Submarine groundwater discharge impacts on coastal nutrient biogeochemistry. Nat. Rev. Earth Environ. 2, 307–323. doi: 10.1038/s43017-021-00152-0

Crossref Full Text | Google Scholar

Song J.-W., Kim J., and Kim T.-H. (2018). Distributions of dissolved organic matter in submarine groundwater discharge (SGD) in Jeju Island. Ocean Polar Res. 40, 77–85. doi: 10.4217/OPR.2018.40.2.077

Crossref Full Text | Google Scholar

Song Y. K., Kim J., Oh Y. H., Joung D., and Kim T. H. (2023). Factors controlling the distribution of dissolved organic carbon and nitrogen in the coastal waters off Jeju Island. Front. Mar. Sci. 10, 1250601. doi: 10.3389/fmars.2023.1250601

Crossref Full Text | Google Scholar

Stuiver M. and Polach H. A. (1977). Reporting of C-14 data - discussion. Radiocarbon 19, 355–363. doi: 10.1017/S0033822200003672

Crossref Full Text | Google Scholar

Taniguchi M., Dulai H., Burnett K. M., Santos I. R., Sugimoto R., Stieglitz T., et al. (2019). Submarine groundwater discharge: updates on its measurement techniques, geophysical drivers, magnitudes, and effects. Front. Environ. Sci. 7, 141. doi: 10.3389/fenvs.2019.00141

Crossref Full Text | Google Scholar

Thomas J. M., Hershey R. L., Fereday W., and Burr G. (2021). Using Carbon-14 of dissolved organic carbon to determine groundwater ages and travel times in aquifers with low organic carbon. Appl. Geochemistry 124, 104842. doi: 10.1016/j.apgeochem.2020.104842

Crossref Full Text | Google Scholar

Tomer A. S., Mckenzie T., Majtenyi-Hill C., Cabral A., Yau Y., Call M., et al. (2025). Groundwater releases CO2 to diverse global coastal ecosystems. Sci. Adv. 11, 1–10. doi: 10.1126/sciadv.adr3240

PubMed Abstract | Crossref Full Text | Google Scholar

Walker B. D., Griffin S., and Druffel E. R. M. (2017). Effect of acidified versus frozen storage on marine dissolved organic carbon concentration and isotopic composition. Radiocarbon 59, 843–857. doi: 10.1017/Rdc.2016.48

Crossref Full Text | Google Scholar

Wassenaar L., Aravena R., Hendry J., and Fritz P. (1991). Radiocarbon in dissolved organic carbon, a possible groundwater dating method: case studies from western Canada. Water Resour. Res. 27, 1975–1986. doi: 10.1029/91WR00504

Crossref Full Text | Google Scholar

Webb J. R., Santos I. R., Maher D. T., Tait D. R., Cyronak T., Sadat-Noori M., et al. (2019). Groundwater as a source of dissolved organic matter to coastal waters: Insights from radon and CDOM observations in 12 shallow coastal systems. Limnology Oceanography 64, 182–196. doi: 10.1002/lno.11028

Crossref Full Text | Google Scholar

Xue Y. J., Zou L., Ge T. T., and Wang X. C. (2017). Mobilization and export of millennial-aged organic carbon by the Yellow River. Limnology Oceanography 62, S95–S111. doi: 10.1002/lno.10579

Crossref Full Text | Google Scholar

Zhou W., Lu X., Wu Z., Zhao W., Huang C., Li L., et al. (2007). New results on Xi’an-AMS and sample preparation systems at Xi’an-AMS center. Nucl. Instruments Methods Phys. Res. Section B: Beam Interact. Materials Atoms 262, 135–142. doi: 10.1016/j.nimb.2007.04.221

Crossref Full Text | Google Scholar

Zhu T. Y., Zhao S. B., Xu B. C., Liu D. Y., Cardenas M. B., Yu H. M., et al. (2025). Large scale submarine groundwater discharge dominates nutrient inputs to China’s coast. Nat. Commun. 16, 1–9. doi: 10.1038/s41467-025-58103-y

PubMed Abstract | Crossref Full Text | Google Scholar