Introduction

Submarine groundwater discharge (SGD) refers to the natural process of groundwater flow into the ocean through coastal margins. Hence, the understanding and quantifying of SGD is essential, especially for regions with long coastlines, as significant volumes of groundwater that could potentially be harnessed to meet the present-day demand for drinking and irrigation gets lost unnoticed1. SGD can also act as a carrier for various pollutants such as heavy metals, micronutrients, and pesticides from the land to the sea2,3. Different human activities have contributed to increase the nutrients in coastal waters in recent years4. Contaminants from the sewage, agricultural runoff, fertilizers, pesticides and industrial effluents often pollute the coastal aquifers, making the SGD a pathway to reach the ocean. This can lead to harmful algal blooms, red tides, low oxygen zones, and loss of marine biodiversity5. Moreover, the interactions between seawater and freshwater within the coastal aquifers significantly impact the coastal ecosystems6. The dynamics of groundwater discharge in coastal areas are highly complex, being strongly influenced by the tidal cycles, which can shift the location, volume, composition, and even the direction of SGD multiple times a day7. Hence, the accurate assessment and monitoring of SGD are vital for the sustainable management and protection of coastal environments.

Several methodologies have successfully estimated the SGD along coastal zones that include: Seepage meters, Hydrologic models, Remote sensing, Isotope tracing, Water balance approaches, Hydrogeochemistry and Resistivity survey8,9,10. Among all these, the isotope tracing serves as an important technique for identification and quantification of SGD. The radon isotopes have proven to be highly effective for detecting and quantifying SGD11,12,13. Groundwater tends to have significantly higher concentrations of radon (222Rn) compared to surface waters, often by 2 to 4 orders of magnitude. This difference makes it possible to detect even small amounts of radon in coastal waters despite the dilution1,14. The naturally occurring radionuclide 222Rn, with a half-life of 3.8 days, is widely utilized as a tracer in SGD studies15,16. Radon is generated within aquifers through the decay of 226Ra in mineral matrices, while its presence in surface waters is minimal due to a lack of production sources. This stark contrast in concentration levels enables the identification of SGD zones and the estimation of discharge rates using radon measurements17. The radon mass balance approach originally developed and subsequently refined by the authors Burnett and Dulaiova and is widely used to quantify submarine groundwater discharge (SGD) fluxes14,15,18. The input of nutrient into the ocean via SGD can be estimated by multiplying this flux with average nutrient concentrations in the groundwater19. Hence, the understanding of SGD and the associated nutrient transport are vital for protecting and managing the coastal ecosystems20. Although the occurrence of SGD along the southwest coast of India at Kanyakumari has been previously reported, the present study offered the first detailed quantitative evaluation of SGD using the combined methods of radon and nutrient mass balance. The relationship between radon concentration, electrical conductivity and nutrient levels helped to demarcate the areas of major discharge and highlights their influence on coastal nutrient dynamics and water quality.

Study area

Kanyakumari district lies in the southernmost part of the Indian subcontinent at the coast of Indian Ocean, between latitude 8˚10’ and 8˚15’N and longitude 77˚10’ and 77˚20’E (Fig. 1). This area features a diverse landscape of rocky shores and sandy beaches and serves as the meeting point of the Bay of Bengal and the Arabian Sea, making it a distinctive geological formation. The major rivers flowing in the region are Pazhayar, Valliyar, and Tamiraparani21. The estuarine ecosystems in areas like Thengapattinam and Manakudi provide crucial habitats for mangroves, mudflats, and diverse avian species, contributing to the region’s ecological richness.

Fig. 1
Fig. 1
Full size image

Location map of Kanyakumari coastal zone in southernmost India and the sampling points of groundwater and porewater for the submarine groundwater discharge estimation.

Rainfall and climate

Kanyakumari has a tropical coastal climate with moderate to high humidity throughout the year. It receives rain from both the Southwest (June-September) and Northeast monsoons (October-December), making it one of wettest districts of the Tamil Nadu state22. The average annual rainfall ranges from 1,000 to 1,400 mm. Summers (March-May) are pleasant, with temperatures ranging from 25 to 35 °C, and winters (December-February) are mild, with temperatures ranging from 20 to 30 °C. The region’s coastline influences its climate, ensuring both the constant temperature and frequent sea breeze.

Geology and hydrogeology

Lithology of the area primarily consists of charnockites, khondalite, migmatite, laterites, sandstones, variegated clay, and river alluvium23,24. The region’s aquifer system is composed of unconsolidated to semi-consolidated formations and fractured crystalline rocks. In hard-rock terrains, the thickness and nature of weathered zone vary both laterally and with depth, significantly influencing the groundwater recharge. Groundwater in alluvial formations exists under water table conditions and generally occurs in an unconfined state.

Materials and methods

222Rn in groundwater and porewater

Fifty-eight water samples (29 groundwater and 29 porewater) were collected during pre-monsoon (May 2023) and post-monsoon (January 2024) seasons based on the tidal fluctuations (Table 1). In-situ parameters like pH, EC (Electrical conductivity), TDS (Total Dissolved Solids), Temperature and Salinity were measured from the field itself using portable water quality analyzer. Samples for radon and nutrients (nitrate, phosphate and silicates) analysis were transported to the laboratories immediately after the sampling. Nutrient analysis was carried out using spectrophotometric methods and Radon analysis was done using RAD7 instrument within 8 h of sampling. Groundwater from inland wells and porewater near the shoreline were collected using 250 ml glass bottles. Porewater was collected from the intertidal zone at a depth of 1.5 m using push point sampler and the groundwater was collected from the wells of the nearby houses.

Table 1 Tidal schedule for sample collection at Kanyakumari Coast of southernmost India (Tide-forecast 2024)25.
Full size table

RAD7 (Durridge Company Inc) is a widely used instrument for measuring 222Rn in water samples and it helps to track freshwater inflows into coastal areas26. A sealed water sample (250 ml) was collected and placed in an air-tight bubbling chamber, where radon gas was transferred from water to air. The radon-enriched air was then circulated through the RAD7’s electrostatic sensor, which detected radon decay by identifying the charged polonium isotopes. The device displayed radon levels in Bq/m³ (or pCi/L), which then got converted to water concentrations using a partition coefficient. Higher radon levels in coastal waters indicated SGD presence, helping estimate the groundwater flux.

RAD7 calibration and measurement principle

The calibration involved comparison with a secondary standard radon chamber, achieving a reproducibility better than ± 2% and an overall accuracy of ± 5% under standard temperature and humidity conditions. Annual recalibration is recommended by the manufacturer to ensure the accuracy. RAD7 operates on an electrostatic detection principle, and its internal algorithm converts alpha counts to radon activity concentration (Bq m− 3) using the calibration factor, and automatically applies corrections for temperature, humidity and background counts.

In Normal mode, the working range is 0.1–400,000 Bq m− 3, with a minimum detectable concentration (MDL) of approximately 0.4 Bq m− 3 (for a 1-hour cycle at 5% relative humidity). In WAT-250 mode, which involves equilibrating dissolved radon between water and a closed air loop using an aeration bottle of 250 mL, the system detects concentrations as low as 0.37 Bq L− 1 depending on counting time and loop volume. The overall measurement uncertainty combining calibration error and counting statistics was estimated at ± 10% for all reported radon concentrations.

Radon-based mass balance approach

Radon mass balance approach is a scientific tool that measures the movement and concentration of radon (more especially, 222Rn) within a system4,27,28. This method basically applies the principle of mass conservation to track the flow of radon through an environment. It is frequently used to estimate groundwater discharge rates in coastal areas by measuring radon levels in both surface and groundwater29. SGD estimation protocol for the radon mass balance model was detailed in Table 2 and in Eq. (1) as follows14:

$${\mathbf{J}}_{{{\mathbf{benthic}}}} + {\mathbf{\lambda I}}_{{{\mathbf{Ra}}}} - {\text{ }}{\mathbf{J}}_{{{\mathbf{atm}}}} - {\text{ }}{\mathbf{\lambda I}}_{{{\mathbf{Rn}}}} \pm {\text{ }}{\mathbf{J}}_{{{\mathbf{hor}}}} = {\text{ }}{\mathbf{0}}$$
(1)

where

Jbenthic = Total radon input from both advective and diffusive processes in the water column.

λ = Decay constant of ²²²Rn.

IRa = Radium inventories.

IRn = ²²²Rn inventories.

Jatm = Atmospheric radon flux.

Jhor = Horizontal mixing of ²²²Rn.

Table 2 Measurement protocol for radon mass balance model of this study.
Full size table

The balance between radioactive decay, atmospheric loss, and mixing is used to calculate net radon flux14. The amount of radon in sediment porewater is estimated since groundwater carries radon from underground sources11. The calculated radon flux from SGD can then be converted into a water flux by dividing by the radon concentration in groundwater. It is also assumed that no river runoff affects the study area, ensuring that the primary source of radon is groundwater rather than riverine input. This assumption is supported by field evidence that the Kanyakumari coast lacks perennial river inflow and industrial or sewage discharge points. Sediment porewater and nearshore groundwater exhibited radon concentrations of one to two orders of magnitude higher than the surface seawater, confirming groundwater as the dominant source. The sandy, permeable coastal sediments and gentle hydraulic gradients in this area further facilitate the radon transport through the submarine groundwater discharge.

Additionally, the radon loss is considered to occur mainly due to radioactive decay and atmospheric evasion, neglecting other minor processes that could contribute to its removal. Minor removal mechanisms such as adsorption onto suspended particulates, biological uptake and molecular diffusion across stratified layers were neglected. These processes typically contribute less than 5% to total radon loss in shallow, well-mixed nearshore systems and are therefore considered insignificant relative to atmospheric evasion and radioactive decay. Another key assumption is that the water column is well-mixed, meaning there is no significant stratification in radon distribution. This ensures that the measured radon concentrations are representative of the entire water column rather than being confined to specific layers. These assumptions help streamline the calculation and improve the accuracy of SGD estimates.

Dissolved inorganic nutrients

SGD serves as an important pathway of dissolved inorganic nutrients to the coastal waters12. Dissolved Inorganic Nitrogen (DIN), Dissolved Inorganic Phosphorus (DIP), and Dissolved Silicate (DSi) represent the inorganic forms of nitrogen, phosphorus and silica respectively, which are often considered key nutrients that are released from land into the ocean, sometimes contributing significantly to the coastal nutrient levels and impacting the marine ecosystems. The SGD-mediated inflow of nutrients can significantly impact coastal ecosystems and the water quality, altering levels of dissolved and gaseous metabolites, including ammonium, methane, and hydrogen sulfide13,32. Multiplying this average nutrient concentration by calculated SGD rates results in the nutrient fluxes33.

Measurement of DIN, DIP and DSi parameters

All nutrient concentrations were measured using a double-beam UV–VIS spectrophotometer (Model 2203, Systronics). Calibration was performed using certified 1000 mg L⁻¹ stock standards with three working standards of 1, 2, and 3 mg L⁻¹ prepared for each analyte. The absorbance–concentration relationship showed good linearity (R2> 0.99). Each sample was analyzed in triplicate, and the mean value was reported. The analytical precision, based on replicate measurements, was within ± 5% for DIN, and DIP, and within ± 7% for DSi.

Nitrite in water reacted with sulphanilamide in an acidic medium, forming a diazo compound. This compound subsequently reacted with N-(1-naphthyl)-ethylene diamine dihydrochloride (NEDA), resulting in the formation of a bright pinkish-red azo dye. The absorbance of both the standard solution and the sample was measured at 540 nm following the same procedure22. Nitrate in water was quantitatively reduced to nitrite by passing the sample through a copper-cadmium (Cu-Cd) reduction column. The resulting nitrite reacted with sulphanilamide, forming a diazonium salt, which then combines with NEDA to produce a pink-colored dye, measured by absorbance at 540 nm34.

Orthophosphate levels were measured using the ascorbic acid method by the reaction with ammonium molybdate to produce molybdophosphoric acid35. This compound was then reduced to form a blue-colored complex indicating the amount of phosphate with the depth of the blue color. The absorbance of this color was recorded at a wavelength of 880 nm. Similarly, the silicate estimation involved preparing molybdate reagent, metol-sulfate solution, oxalic acid solution, and 50% sulfuric acid, subsequently mixed with the reducing reagent. Absorbance measured at 810 nm, with a blank solution as a reference, indicated the concentration of dissolved silicate. The procedure is repeated for 20 ml samples to determine silicate concentration.

Results and discussion

Physicochemical characteristics

Water samples in pre-monsoon showed pH of 5.6–7.4, with a mean of 6.5 ± 0.4, whereas the post-monsoon samples had pH between 5.4 and 7.53, with a mean of 6.7 ± 0.3 (Fig. 2). Values close to 7 in both groundwater and porewater for both the seasons indicated nearly neutral conditions. EC of pre-monsoon groundwater (256–3770 µS/cm) and porewater (51.5 to 52.4 mS/cm) and the post-monsoon groundwater (329 to 1701 µS/cm) and porewater (36.2 to 38mS/cm) indicated SGD in all the post-monsoon and most of the pre-monsoon sites where groundwater showed EC < 3000 µS/cm36. High EC in groundwater from Chinnathurai could be an indication of seawater intrusion. Enayum Edapadu beach and Mandaikadu beach showed low EC values during both the seasons, which suggested the presence of SGD. EC of porewater remained lower in post-monsoon compared to the pre-monsoon. Highest EC in the pre-monsoon and lowest EC in post-monsoon reflected enhanced evaporation and high salt concentration in the dry pre-monsoon season and dilution of the groundwater from precipitation in the post-monsoon37. TDS varied from 128 to 1880 mg/L in groundwater and 25,700 to 28,800 mg/L in porewater during the pre-monsoon. In post-monsoon, It was characterized by 83–672 mg/L in groundwater and 20,500–26,300 mg/L in porewater. The higher salinity of groundwater in the areas with more electrical conductivity indicated seawater intrusion.

Radon concentration in groundwater and porewater

Radon (222Rn) concentration varied between 32.02 and 66.96 Bq/L in groundwater and between 11.68 and 19.77 Bq/L in porewater of pre-monsoon. In post-monsoon samples, the groundwater and porewater had radon of 25.6–189.56 Bq/L and 18.9–41.5 Bq/L, respectively. Some of the groundwater and all the porewater of post-monsoon had more radon. Maximum radon at Muttam Beach during both the seasons possibly reflected radioactive decay of parent 226Ra nuclide from the aquifer matrix31. More radon and higher EC indicated saline SGD whereas the higher radon and lower EC demarcated areas of fresh SGD7. The spatial distribution of both in sites such as Muttam exhibited elevated levels of radon but comparatively low EC, signifying fresh groundwater discharge into the sea. In contrast, the areas with high EC and low radon values are mainly influenced by seawater intrusion or surface evaporation (Fig. 3). These variations resulted from a combination of environmental and hydrogeological factors. Seasonal fluctuations have modified the groundwater recharge and discharge dynamics, thereby influencing the radon activity and nutrient concentrations38,39. Tidal movements have led to intermittent mixing of seawater and groundwater, affecting the salinity and electrical conductivity along the shoreline. The interaction between groundwater and seawater has regulated the transfer of dissolved substances with the fresh groundwater input generally raising both radon8,40,41 and nutrient levels, and seawater intrusion enhancing EC. Furthermore, the local geological characteristics, including lithology, aquifer structure, and permeability differences, determined the extent and distribution of submarine groundwater discharge along the coastal zone of Kanyakumari.

Dissolved nutrients concentration and distribution

Tables 3 and 4 shows the measured physical parameters, nutrients and radon in groundwater and porewater samples during pre-monsoon and post-monsoon seasons. Nitrate and phosphate in groundwater mainly originate from the natural breakdown of organic matter, the dissolution of minerals, and human-related activities such as the use of fertilizers, application of manure, and discharge of wastewater. Elevated silica levels are likely linked to the weathering of silicate-rich sediments. The patterns of nutrient also differ by location, depending on factors like oxygen availability and biological processes such as nitrification and denitrification within the aquifer. Over-extraction of groundwater, rainfall patterns, and tidal movements also influence the nutrient fluxes6. DIN values between 3.04–33.17µmoles/L in groundwater and between 1.22–4.26µmoles/L in porewater of pre-monsoon were higher compared to the post-monsoon (0.37–4.51 µmoles/L; groundwater and 0.13–3.31 µmoles/L; porewater, Fig. 4). Maximum values of DIN in groundwater and pore water were observed at Enayum Edapadu beach and Muttam. DIP of 0.42–5.96 µmoles/L in groundwater in pre-monsoon was higher compared to the post-monsoon (0.23–3.54 µmoles/L). Some pre-monsoon porewater (0.40 to 1.29 µmoles/L), however contained less DIP compared to the post-monsoon (0.13 to 3.31 µmoles/L). High DIP was observed at Chinnathurai and Mandaikadu beach. DSi of groundwater (32.99–314 µmoles/L) and porewater (28.13 to 35.86µmoles/L) in pre-monsoon was more than the post-monsoon (17.47–38.98 µmoles/L; groundwater and 27.04–246.55 µmoles/L; porewater). Maximum values of DSi at Muttam beach in both the seasons showed indication of SGD (Fig. 4). This discharge of DIN, DIP, and DSi into seawater can lead to nutrient buildup, and problems like algal blooms and eutrophication7,42.

Table 3 Measured physical parameters, nutrients and radon in groundwater and pore water samples during pre- monsoon season at Kanyakumari Coast of southernmost India.
Full size table
Table 4 Measured physical parameters, nutrients and radon in groundwater and pore water samples during post-monsoon season at Kanyakumari Coast of southernmost India.
Full size table
Fig. 2
Fig. 2
Full size image

Variation of physicochemical parameters during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

Fig. 3
Fig. 3
Full size image

Radon and EC of the study area during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

Fig. 4
Fig. 4
Full size image

Spatial distribution of DSi, DIN and DIP during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

The box plot analysis (Figs. 5 and 6) showed seasonal and tidal differences in nutrient behavior. In pre-monsoon, the concentrations of DSi and DIN were generally higher and more variable, especially at low tide, suggesting active benthic and groundwater contributions when freshwater dilution was minimal. DIP also showed larger fluctuations at low tide, likely due to nutrient release from the surface sediments.

Fig. 5
Fig. 5
Full size image

Variation of nutrients (DSi, DIP and DIN) during pre-monsoon season at Kanyakumari coast of southernmost India.

Fig. 6
Fig. 6
Full size image

Variation of nutrients (DSi, DIP and DIN) during post-monsoon season at Kanyakumari coast of southernmost India.

In contrast, the post-monsoon nutrient concentrations were lower and more consistent between tides, indicating that monsoonal runoff and enhanced mixing reduced the relative influence of SGD and sedimentary inputs. Overall, these results demonstrated a clear transition from nutrient enrichment driven by subsurface inputs before the monsoon to a more diluted and well-mixed system after the monsoon season.

Spatial and seasonal variations of radon, nutrients (DIN, DIP, DSi) and salinity

Scatter plots compared the radon (222Rn) concentrations with salinity during the pre-monsoon and post-monsoon periods, and interpreted the SGD patterns. During pre-monsoon, the salinity ranged widely, with one sample (LG1) from Chinnathurai beach showing an exceptionally high salinity (~ 2000 mg/L). This sample with extremely high salinity and relatively low radon implied mixing with seawater. Higher radon levels with lower salinity suggested stronger SGD from freshwater sources. During post-monsoon, the salinity remained more confined (mostly < 700 mg/L), suggesting increased dilution by the freshwater or precipitation. Many samples exhibited increased radon levels, with some exceeding 100 Bq/L (notably HG8, reaching ~ 200 Bq/L) (Fig. 7.). Overall higher Radon concentrations in post-monsoon indicated greater SGD due to enhanced groundwater flow. The monsoon cycle significantly impacted SGD dynamics, with post-monsoon conditions favoring more groundwater discharge.

Salinity versus nutrient concentration provided insights about SGD and its role in coastal nutrient dynamics. During pre-monsoon, the high DSi (> 200 µmol/L) at lower salinity levels (< 500 mg/L) suggested a strong terrestrial groundwater influence. Maximum DSi levels with low salinity in samples from Muttam demarcated this location as a probable fresh SGD zone. Higher DIN values at lower salinity levels suggested fresh groundwater as a primary source. During post- monsoon, the DSi concentrations remained significantly lower (< 50 µmol/L), suggesting dilution by rainfall-infiltrated groundwater or changes in groundwater flow paths. DIP concentrations, however, remained consistently low during both the seasons, suggesting limited phosphorus input from SGD (Fig. 8).

Radon based estimation of SGD fluxes

Seasonal variation in SGD flux provided insights into groundwater-seawater interactions and the impact of monsoonal recharge on SGD dynamics6. SGD flux during pre-monsoon and post-monsoon varied between 0.01 and 0.54 m3m−2d−1 and 0.04 to 0.98 m3m−2d−1, respectively (Fig. 9). Enayum Edapadu beach and Mandaikadu beach showed the maximum flux rate in both the seasons, indicating localized high groundwater discharge zones. Rainfall infiltration recharged the aquifer and increased hydraulic gradients, leading to higher post-monsoon SGD flux43.

Table 2 lists all radon mass-balance components with propagated 1σ uncertainties for each sample (HG = High Tide Groundwater, HP = High Tide Porewater, LG = Low Tide Groundwater, LP = Low Tide Porewater). Inventory depth (H) was taken as 5 m with ± 20% uncertainty. The decay constant of 222Rn (λ) was 0.181 d− 1 with negligible uncertainty. Atmospheric evasion was computed using a gas transfer velocity (k) of 0.5 ± 10%. Horizontal eddy diffusivity (Kx) was assumed to be 10 ± 15%, with a mixing length of 5–8 m based on nearshore hydrodynamics along the Kanyakumari coast. Diffusive benthic flux (Jdiff) was approximated as 750 ± 10% Bq m− 2 d− 1. Radon measurement uncertainty (RAD7 calibration and counting statistics) was ± 10%, and SGD flux uncertainty ± 10%. Radium-supported inventory (IRa) was estimated as 10% of IRn with ± 10% uncertainty. All uncertainties were propagated using standard variance and multiplicative error propagation rules. The residual term represented the imbalance between total inputs and losses, reflecting unquantified flux components or model simplifications.

Fig. 7
Fig. 7
Full size image

Radon versus salinity plot for groundwater samples collected during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

Fig. 8
Fig. 8
Full size image

DIN, DIP, and DSi versus salinity for groundwater samples collected during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

Fig. 9
Fig. 9
Full size image

Observed SGD flux in groundwater during pre-monsoon and post-monsoon seasons at Kanyakumari coast of southernmost India.

Dissolved nutrients mass balance

SGD flux multiplied by the nutrient concentration provided the nutrient mass flux. DSi, DIP and DIN fluxes were higher along Mandaikadu beach in both seasons44. During pre-monsoon, the SGD associated dissolved nutrients were 2, 0.21, and 20.75 µmol L− 1 for DIN, DIP, and DSi, respectively. They were 0.79, 0.28 and 17.22 µmol L− 1, respectively, for post-monsoon seasons. More DIN in pre-monsoon period may increase algal blooms in coastal waters. DSi flux was found to be more in both seasons. High DSi suggested prolonged contact with silicate-rich minerals (e.g., quartz, feldspars) during the subsurface flow42.

Conclusions

Radon is a reliable tracer of SGD in saline environments with distinct enrichment in groundwater relative to the seawater. Here, we used both radon and dissolved nutrients as tracers for quantifying the SGD flux into the Arabian Sea along the western coastal stretch of Kanyakumari at southernmost India. Freshwater SGD was confined along the northern portion of the study area i.e. the Enayam beach. This spatial variation was controlled primarily by differences in coastal geomorphology, permeability and tidal forcing along the shoreline. The radon mass balance quantified SGD flux of 0.01–0.54 m³m⁻²d⁻¹ during the pre-monsoon and 0.03–1.02 m3m−2d−1 in the post-monsoon. This seasonal variation indicated influence of monsoonal recharge and tidal pumping, which enhanced the groundwater discharge during the post-monsoon. However, the pre-monsoon conditions favored higher nutrient accumulation reflected in the DIN, DIP, and DSi contents of 2, 0.21, and 20.75 µmol L−1, respectively. The post-monsoon conditions promoted dilution and hence reduced the dissolved nitrogen (0.79 µmol L−1) and silicate (17.22 µmol L−1) contents. Insight into these dynamics was important for effective control of coastal nutrient inputs and understanding SGD-driven nutrient fluxes for the sustainable coastal management, especially in a region where anthropogenic and natural processes jointly affect the nutrient dynamics.

Despite the robustness of the radon mass balance approach, certain limitations should be known. The method assumes quasi–steady-state conditions and well-mixed nearshore waters, which may not fully capture short-term variability driven by tidal pumping, wave action, or intermittent recharge events. Uncertainties associated with atmospheric evasion and horizontal mixing terms derived from empirical parameterizations can influence SGD estimates. In addition, the conversion of radon flux to water flux depends on representative groundwater end-member concentrations, which may vary spatially due to aquifer heterogeneity. Furthermore, while radon effectively constrains total SGD, it does not explicitly distinguish between fresh groundwater discharge and recirculated seawater without balancing tracers. Nevertheless, when applied with appropriate hydrological context and uncertainty assessment, the approach remains a powerful tool for quantifying SGD and associated nutrient fluxes in monsoon-influenced coastal systems.