Submarine groundwater discharge
Submarine groundwater discharge (SGD) can have a significant influence on the coastal environment. This article examines the underlying processes and the importance for coastal zone management. In order to assess the magnitude of SGD, different investigations based on models and on direct measurements have been carried out. Indirect indicators of SGD are also considered.
Process description and estimates
Submarine groundwater discharge (SGD) is defined as the discharging flow out of the aquifer across the sea floor . It is a process complementary to saltwater intrusion, which is defined as the invasion of seawater at the lower part of the aquifer due to the density difference between fresh groundwater and seawater, see Groundwater management in low-lying coastal zones. Both processes are shown schematically in Figures 1a and 1b.
If mixing between fresh groundwater and seawater wedge is weak, it can be assumed that the interface between the two areas is sharp (Figure 1a). In such cases, during the winter period, the submarine groundwater discharge consists of fresh groundwater, which flows to the sea above the saltwater wedge, and the interface is located more landward than in summer, which is due to the relatively low groundwater levels occurring in fall. During summer, the interface moves seaward due to the rise of the groundwater level occurring in spring, while, simultaneously with fresh groundwater discharge, seawater outflow takes place from the saltwater wedge. The seawater outflow can be greater than the fresh groundwater discharge .
In most cases, the interface between the freshwater and the saltwater is not sharp. Due to dispersion and molecular diffusion a freshwater-seawater mixing zone is built between the two areas. The mixing causes the salt water in the deeper part of the aquifer to become lighter, rise up and discharge back to the sea (Figure 1b). In this case, SGD consists of a mixture of fresh groundwater and recirculated seawater. The recirculation of seawater, in the nearshore part of the aquifer, is further enhanced by mechanisms acting from the marine side such as tidal pumping and wave set-up. The time variation of SGD exhibits components at small time scales (minutes, hours, days), which are due to the wave action and tides, as well as components of large time scale, which are due to the seasonal motion of the mixing zone and large-scale sea level variations.
Concerning the contribution of the fresh groundwater flow and the seawater recirculation to the total SGD, it is estimated that:
- The global total SGD flux, which includes recirculated seawater, could be three to four times larger than the global river flux, based on measured radium isotope concentrations in seawater. The terrestrially derived SGD was estimated by Burnett et al. (2003) in the range of 6-10% of the surface waters discharging into the ocean. However, more recent calculations by Luijendijk et al. (2020) show that the submarine groundwater discharge cannot exceed 5.5% of the river input into the world’s oceans, assuming that all groundwater recharge in coastal watersheds were to discharge directly into the oceans.
- Assuming that recirculation is only due to density effects and mixing, recirculated seawater can be up to 70% of the SGD ;
- Assuming that recirculation is also due to wave setup and tide, the recirculated sea water may constitute up to 96% of the SGD . However, Luijendijk et al. estimate that the fresh SGD is even a much smaller fraction, 0.06% (0.0003%–0.2%), of the global total SGD flux, which thus consists almost entirely of recirculated seawater. According to these authors, the contribution of fresh groundwater discharge to the oceans is globally in the order of ~0.6% (0.004%–1.3%) of the total freshwater input and ~2% (0.003%–7.7%) of the solute input for carbon, nitrogen, silica and strontium.
Submarine groundwater discharge (SGD) can take place not only near the shore but also at great offshore distances. The latter occurs when confined aquifers, underneath the continental shelf, have their outcrop on the sea bottom and at a large distance from the shore. SGD is usually distributed over large areas. Concentrated outflows at the sea bottom occur in cases where the outcrops of alluvial aquifers are of limited area. Such outflow locations on the sea bottom sometimes form pockmarks in the fine-grained sediments of the sea floor, see Fig. 2. Particularly strong, concentrated, submarine groundwater discharges occur in case of karstic aquifers discharging into the sea (karstic submarine springs).
The study of Luijendijk et al., which is based on numerical modeling, reveals a high spatial variability of the coastal discharge of fresh groundwater and nutrients. For an estimated 26% (0.4%–39%) of the world’s estuaries, 17% (0.3%–31%) of the salt marshes and 14% (0.1–26%) of the coral reefs, the flux of terrestrial groundwater exceeds 25% of the river flux and poses a risk for pollution and eutrophication. Coastal groundwater discharge hotspots cover 9% (0.02–30)% of the global coastline and are predominantly located in areas with a steep coastal topography due to glacio-isostatic rebound, active tectonics or volcanic activity and areas consisting of permeable unconsolidated sediments, carbonates or volcanic rocks. The distribution of hotspots is consistent with documented sites of high fresh groundwater discharge globally that are predominantly located in North America, Europe, and East Asia.
Importance for coastal zone management
Submarine groundwater discharge has been recognized in the last decades as a mechanism for transporting land derived pollutants to the sea . Nutrients, organics, metals and pathogens, which are dissolved in terrestrial groundwater, can induce chemical and biological effects in the nearshore sea area, where groundwater discharges. These effects are enhanced by the recirculated seawater which is enriched with pollutants as it reacts with the aquifer sediment. The pollutant concentration in the nearshore water and its impact on the chemistry and biology of this area depends not only on the pollutant fluxes but also on the intensity of mixing and the exchange with the open ocean. Thus, in order to determine whether SGD in an area of interest is of actual or probable importance, coastal zone managers need to estimate:
- the magnitude of submarine groundwater discharge and
- the intensity of mixing in the nearshore sea area and the exchange between the nearshore sea and the open ocean.
Estimation of the magnitude of SGD
Complementing several ongoing studies addressing the problem of submarine groundwater discharge estimation, an initiative was launched by the International Atomic Energy Agency (IAEA) and UNESCO in 2000 to assess the importance of SGD and the effectiveness of prediction methodologies for coastal zone management. The results of the projects undertaken in the framework of this initiative were presented in SCOR-LOICZ (2004) as well as in Burnett et al. (2006 ). A comprehensive report concerning current knowledge about SGD was presented by Gallardo and Marui (2006 ).
The methods to estimate SGD can be distinguished in:
- (a) model investigations and
- (b) experimental investigations.
These investigations involve hydrogeological methods based on the water balance approach and numerical models.
In the water balance approach, the estimation of SGD is based on the water balance equation at the basin scale. For extended periods (i.e. years), over which the change of water storage can be neglected, precipitation, evapotranspiration and surface discharge must be accurately known in order to reliably estimate SGD. The water balance approach only provides the fresh groundwater component of SGD as the total volume of groundwater discharging to the sea during the selected time period. Thus, the results of the water balance approach are not comparable with locally measured values or with results of numerical models that provide the distribution of SGD over the discharge area. An example of the water balance approach is the work of Sekulic and Vertacnik (1996 ), who estimated SGD in the Adriatic Sea.
Numerical models used in submarine groundwater discharge studies differ in the degree of complexity. Groundwater models neglecting density effects, as for instance MODFLOW , have been widely used. They provide only the freshwater component of SGD and can appropriately simulate the flow in the part of the coastal aquifer upstream of the saltwater intrusion area. SGD in such simulations is the groundwater flux at the seaward model boundary, assuming that there is no groundwater extraction in the area downstream of this boundary up to the shore. The collection of the data needed for the model calibration requires a considerable effort and it is time consuming, particularly for the parameters varying in time such as water levels and fluxes. However, in areas in which such data are available, these numerical models represent an efficient tool for the estimation of the freshwater part of SGD.
More recently, numerical models considering density effects have been applied in SGD studies (an extensive review of such models can be found in Bear et al. 1999 ). These models include the nearshore terrestrial part of the aquifer as well as the marine part below the sea floor, where density effects are relevant. They provide both the fresh groundwater discharge and the discharge of the seawater recirculated in the aquifer. Their application is feasible only if subsoil data are available. However, particularly for the part of the aquifer below the sea floor, the collection of such data is difficult. Therefore applications of such models, as those presented by Smith and Turner (2001 ), Kaleris (2002 ), Langevin (2003 ) and Destouni and Prieto (2003 ), are limited.
Luijendijk et al. (2020) simulated submarine and terrestrial groundwater discharge in coastal groundwater systems using a numerical model of coupled density-driven groundwater flow and solute transport that solved the fluid flow and solute transport equations and the equations of state for fluid density and viscosity in a two-dimensional cross section of the subsurface. The model simulated the flow of fresh (meteoric) groundwater, the mixing and recirculation of seawater at the fresh-salt water interface and the onshore and offshore discharge of groundwater. Assumed were a constant linear slope and a fully saturated standard aquifer of 100 m thickness, which is roughly equal to the thickness where globally the majority of young groundwater and active groundwater circulates. For the global model, 40,082 coastal watersheds were selected using global watershed and coastline datasets. Permeability of each coastal watershed was extracted from a global dataset of near-surface permeability. Comparison between modeled and measured average hydraulic gradients in 336 coastal watersheds showed that the model provides a realistic estimate of coastal groundwater flow. The model outcome for the global fresh submarine groundwater discharge is lower than estimates of earlier studies. The authors explain why the fresh submarine groundwater flow in previous studies is often overestimated; this is the case, for example, when a large part of the overall groundwater recharge is not discharged by SGD but by perennial streams.
The most important experimental methods for investigations of submarine groundwater discharge are seepage measuring techniques and the use of geochemical tracers such as radium isotopes and radon, which are both enriched in groundwater relative to seawater.
An overview of different types of seepage meters used to measure SGD at the bottom of the sea, as well as conclusions concerning their applicability in the field, have been presented by Taniguchi et al. (2003 ). Seepage meters provide local SGD values because of the variability of the hydraulic properties of the sea bottom material and the relatively small area (usually <<1m2), over which SGD is measured. Taking into consideration the variability of SGD in time, it can be concluded that in order to obtain representative values of SGD for a shore, which can be a few kilometres long, a large number of measurements at different locations and at different times must be performed. The results of seepage meter applications and comparisons of various methods have been presented by Burnett et al. (2006 ). Another measuring tool is described in Submarine groundwater discharge in coastal regions: development of a new sampling technique.
Radon is produced by the decay of radioactive isotopes in the sediments. It is therefore strongly enriched both in the freshwater part as well as in the recirculated part of submarine groundwater discharge relative to coastal marine waters . Thus, freshwater and recirculated parts of SGD can hardly be distinguished through radon measurements. To what extent a nearshore water sample used to measure radon activity, provides representative values for a larger part of the nearshore water body, depends on the degree of mixing of SGD with the surface water.
Radium is absorbed onto the surface of the geologic materials and is immobile in fresh groundwater. When saltwater intrudes the aquifer and replaces freshwater in the pores, radium is desorbed from sediment and becomes mobile . Thus, radium is transferred to the nearshore seawater primarily by the recirculated seawater actively involved in saltwater intrusion . Since this water discharges into the sea mixed with the fresh groundwater, it is not possible, as in case of radon, to distinguish the portions of fresh groundwater and seawater in SGD.
Kwon et al.  determined the submarine groundwater discharge from a global compilation of 228Ra observations, by using an inverse model. They found that the SGD integrated over the Atlantic and Indo-Pacific Oceans between 60°S and 70°N is (12 ± 3).104 km3/yr, which is 3 to 4 times greater than the freshwater fluxes into the oceans by rivers. Although this discharge consists mainly of seawater flowing back to the ocean, it is enriched during its temporary invasion of the coastal aquifer with trace elements, rare earth elements, nutrients, and dissolved inorganic and organic carbon through biogeochemical reactions. This suggests that SGD is an important pathway for dissolved terrestrial materials to the coastal waters.
An overview of the technologies developed for the measurement of radon and radium activities, advantages and disadvantages of the methods as well as applications for the estimation of submarine groundwater discharge, have been described by Burnett et al. (2006 ).
Indirect indicators of SGD
Application of the methods described above involves considerable financial, labor and/or computational effort. Thus, before such methods are applied in coastal zone management studies, it is important to determine if the magnitude of submarine groundwater discharge is relevant. Results of investigations presented in the literature can hardly help in such assessments as they vary widely. High values (above 100 cm/day) as well as low values (below 5 cm/day) have been observed . A way to determine the importance of SGD in an area, is to use indirect indicators , for example:
- the groundwater escaping from the sea bottom might be coloured by tiny gas bubbles;
- surrounding sediment might be stained red by the oxidation of iron;
- there are cold-water anomalies in the open water during the summer and warm-water anomalies in winter as well as salinity anomalies;
- the levels of radium, radon, methane, hydrogen sulphide or carbon dioxide might be elevated.
Mixing in the nearshore sea
The concentration of pollutants, transferred to the nearshore sea area by submarine groundwater discharge, depends on the intensity of mixing in this area, which is due to hydrodynamic circulation, transport and dispersion. The estimation of the pollutant concentration in the nearshore sea using numerical models, which simulate the forementioned processes, is computationally intensive and requires data, which in many cases are not readily available. A more convenient approach is to use cell models (well-mixed reservoir models).
An application of a cell model is given by Uchiyama et al. (2000 ), who investigated the nutrient concentration in the nearshore sea due to SGD. They found that the long-term pollutant concentration depends on the relationship between the pollutant fluxes due to SGD and the exchange flux between the nearshore sea and the open ocean. For the conditions of the investigated case, these authors found that the volumetric exchange flux between the nearshore sea and the open ocean is three orders of magnitude larger than the SGD flux. Thus, the long term pollutant concentrations, resulting under the assumption that the pollution of the open ocean water is negligible, were found to be of minor importance for the nearshore sea environment.
For estimations of the exchange of water between the nearshore sea (surf zone) and the offshore zone, a relatively simple model presented by Inman et al. (1971 ) can be used. The data required are the wave and beach characteristics (breaker height, breaker angle, beach slope in the surf zone, beach slope offshore). For open beaches, typical wave and beach characteristics, it is probable that the water exchange between the surf zone and the offshore strongly dilutes pollutants transported in the surf zone by submarine groundwater discharge, if SGD is of the same order of magnitude as in existing studies, which are reported in Burnett et al. (2006 ) and Gallardo & Marui (2006 ). However, in semi-closed bays or in nearshore seas with weak wave action and circulation, it can be shown using cell models, that the long term pollutant concentration in the nearshore environment due to the impact of SGD is not negligible. In such cases, measures to reduce the pollutant load of groundwater are required.
- Submarine groundwater discharge in coastal regions: development of a new sampling technique
- Groundwater management in low-lying coastal zones
- Burnett, W.C., et al. (2006). Quantifying submarine groundwater discharge in the coastal zone via multiple methods. Sciences of the Total Environment, 367 498-543.
- Michael H.A., Mulligan A.E. and Harvey C.F. (2005). Seasonal oscillations in water exchange between aquifers and the coastal ocean. Nature 436 1145-1148.
- Kwon, E. Y., Kim, G,. Primeau, F., Moore, W.S., Cho, H.-M., DeVries, T., Sarmiento, J. L., Charette, M. A. and Cho, Y.-K. 2014. Global estimate of submarine groundwater discharge based on an observationally constrained radium isotope model. Geophys. Res. Lett. 41: 8438–8444, doi:10.1002/2014GL061574
- Burnett, W.C., et al. (2003). Groundwater and pore water inputs to the coastal zone. Biogeochemistry 66 3-33.
- Luijendijk, E., Gleeson, T. and Moosdorf, N. 2020. Fresh groundwater discharge insignificant for the world’s oceans but important for coastal ecosystems. Nature Commun. 11, 1260. https://doi.org/10.1038/s41467-020-15064-8
- Smith A.J. (2004). Mixed convection and density-dependent seawater circulation in coastal aquifers. Water Resource Res., 40 W08309. doi:10.1029/2003WR002977.
- Kaleris V. (2006). Submarine groundwater discharge: Effects of hydrogeology and of nearshore surface water bodies. J. Hydrology, 325 96-117.
- Li L. et al. (1999). Submarine groundwater discharge and associated chemical input to a coastal sea. Water Resour. Res. 35(11) 3253-3259.
- Moore W.S. (1996). Large groundwater inputs to coastal waters revealed by 226Ra enrichment. Nature, 380 612-614. Cite error: Invalid
<ref>tag; name "Moore" defined multiple times with different content
- Gallardo A.H. and Marui A. (2006). Submarine groundwater discharge: an outlook of recent advances and current knowledge. Geo-Mar Lett 26' 102-113. Cite error: Invalid
<ref>tag; name "Gallardo" defined multiple times with different content
- Sekulic, B. and Vertacnik, A. (1996). Balance of Average Annual Fresh Water Inflow into the Adriatic Sea. Water Resources Development, 12(1) 89-97.
- McDonalt M.G. and Harbaugh A.W. (1988). A modular three-dimensional finite-differences groundwater flow model. US Geol Surv Open-File Report 83-875, 528p.
- Bear J. et al. (Eds), (1999). Seawater intrusion in coastal aquifers – concepts methods and practices. Kluwer Academic Publishers, Dordrecht.
- Smith A.J. and Turner J.V. (2001). Density-dependent surface water-groundwater interaction and nutrient discharge in the Swan-Canning Estuary. Hydrological Proc, 15 2595-2616.
- Kaleris V. et al. (2002). Modelling submarine groundwater discharge: an example from the western Baltic Sea. J. Hydrology, 265 76-99.
- Langevin G.D. (2003). Simulation of Submarine Ground Water Discharge to a Marine Estuary: Biscayne Bay, Fl. Ground Water, 41(6) 758-771.
- Destouni G. and Prieto C. (2003). On the possibility for generic modelling of submarine groundwater discharge. Biogeochemistry 66 171-186.
- Taniguchi M. et al. (2003). Spatial and temporal distribution of submarine groundwater discharge rates obtained fom various types of seepage meters at a site in the northeastern Gulf of Mexico. Biogeochemistry 66 35-53.
- Oberdorfer J.A. (2003). Hydrogeologic modeling of submarine groundwater discharge: comparison to other quantitative methods. Biogeochemistry 66 159-169.
- SCOR-LOIZ (2004). Submarine groundwater discharge management implications, measurements and effects. IHP-VI Series on Groundwater 5. IOC Manuals and Guides 44. United Nations Educational, Scientific and Cultural Organization, Paris.
- Uchiyama Y., Nadaoka K., Rölke P., Adachi K. and Yagi H. (2000). Submarine groundwater discharge into the sea and associated nutrient transport in a sandy beach. Water Resources Res. 36(6) 1467-1479.
- Inman D.L., Tai R.J. and Nordstrom C.E. (1971). Mixing in the surf zone. J. Geophys. Res., 76 3493-3514.
Please note that others may also have edited the contents of this article.
Article reviewed by