Coastal regions are the most densely-populated areas in the world. It has been estimated that at least 1.2 billion people live within 100 km of the shoreline, representing an average population density nearly 3 times higher than the global average . Freshwater resources in coastal states and island nations are therefore under enormous stress, and their quantities and qualities are rapidly deteriorating. This problem is exacerbated by population growth, pollution, predicted climate change and political conflicts . Problems are especially felt in arid areas, such as Malta, where groundwater is the only source of freshwater and the periods of highest demand (e.g., agricultural and tourist seasons) coincide with the periods of lowest recharge from precipitation . Cape Town is the first major city in the modern era to face the threat of running out of drinking water, and other large cities like Jakarta, São Paolo and Beijing are likely to follow suit.
Offshore aquifers (OAs) have been proposed as an alternative source of freshwater to cover demand by domestic, agricultural and tourist industries in coastal regions . During the Last Glacial Maximum (19- 22,000 years ago), modern shelf areas were subaerially exposed, leading to the development of extensive water tables recharged by atmospheric precipitation (meteoric water), rivers, lakes and, in some areas, glacial meltwater . In view of the fact that sea level has been much lower than today for 80% of the Quaternary period (last 2.6 million years), and that meteoric groundwater systems migrate landwards more slowly than rising sea levels, remnants of meteoric groundwater occur extensively offshore . An OA has been defined by  “as a groundwater body with a minimum horizontal extent of 10 km, and a minimum concentration of total dissolved solids (TDS) less than 10 g/L, roughly 1⁄3 the salinity of seawater.” Two types of OAs can be distinguished (Fig. 1). The first type (active) entails a present-day, permeable connection of the OA with a terrestrial aquifer recharged by meteoric water [8, 9]. Such aquifers tend to be wedge-shaped, becoming thinner and more saline with increasing distance from the coast. However, onshore hydraulic heads are sometimes too low to drive water offshore  or a hydraulic connection between offshore and onshore aquifers may be absent . In such cases, offshore groundwater systems are associated with palaeo- groundwater (fossil) systems that have been emplaced by meteoric recharge during lowered sea level periods  and that are no longer recharged.
Cartoon depicting the differences between active (connected) and fossil (disconnected) offshore aquifers. The modern day active aquifers are recharged by precipitation (green arrows). Fossil aquifers are no longer fed by meteoric water and are subject to saltwater intrusion (red arrows).
Diagram by: Bradley Weymer/GEOMAR
Recent studies have estimated the volume of OAs to range between ~ 3 x 105 km3  and 4.5 x 106 km3 , with a more robust estimate of 5 x 105 km3 . The latter is two orders of magnitude greater than what has been extracted globally from continental aquifers since 1900 (4.5 x 103 km3). The salinity of OAs can range between freshwater and seawater values, and the salinity threshold defined by  corresponds to the upper limit of the salinity range used for the definition of brackish water for desalination . Since submarine groundwater can be exploited with technology from the oil and gas industry and onshore groundwater exploitation, and because the costs seem to be economically competitive with desalination , OAs have the potential to become an important resource that can relieve water scarcity and mitigate the adverse effects of groundwater depletion (e.g. land subsidence, saltwater intrusion) in densely populated coastal regions.
Detection and quantification of OAs
The characteristics of offshore groundwater systems remain poorly constrained, and there are many first-order questions, related to aquifer geometry and distribution, waiting to be addressed. This arises from a paucity of appropriate offshore data, in particular the limited coverage of sub-seafloor borehole data, as a result of the expense and technical challenges associated with acquiring such data. Direct observation of offshore groundwater reservoir structure and geochemical analyses of pore water data remain rare. Part of the problem is that methods used to map the extent, volume, controls, and connectivity of offshore/onshore freshwater aquifers are still in the experimental phase. This is compounded by the fact that most measurements and research efforts have focused on the nearshore zone. Conventional offshore groundwater aquifer and submarine groundwater discharge (SGD) methods rely on point-source data from boreholes, seepage meters, and chemical radionuclide tracer techniques that cannot provide continuous information of the groundwater system .
One promising geophysical method for locating and providing continuous information of offshore groundwater is controlled-source electromagnetics (CSEM). On land, electrical conductivity (σ) maps derived from CSEM measurements are routinely used to map aquifers and guide hydrological sampling [16- 18]. Offshore, CSEM techniques have also been shown to be relevant, since the σ of groundwater depends on sediment porosity and the amount and type of pore fluids. Previous studies have demonstrated that strong contrasts in σ occur between freshwater (low σ) and saltwater (high σ) [19-21]. To date, there are only a few studies that have used CSEM to map groundwater aquifers in the marine environment [18, 20, 22, 23].
Dynamics and evolution of OAs
The functioning of OAs depends both on the geometry of groundwater aquifers and the temporal variability of recharge and boundary conditions such as sea level change. We have limited understanding of the routes and mechanisms by which water is transferred between terrestrial and offshore aquifers, the rates at which it does so, and the key controlling factors. The dimensions of OAs and their hydrologic budgets still have to be quantified. These questions have so far been addressed using predictive mathematical models and ground-truthing by drilling. Despite representing a powerful tool to assess groundwater resources [24, 25], numerical models are generally based on limited well data, low resolution seismic data, and poorly defined aquifer properties. They also pay little attention to the land–sea connection, despite the fact that the best- studied offshore reservoirs have an onshore connection . So far no methodology exists that allows us to distinguish between active and fossil OAs, although this would be of utmost importance because the exploitation approaches for these two types of OAs are very different. Any OA exploitation program has to take into consideration all of the above to devise a sustainable pumping strategy, as well as how the groundwater system will respond to groundwater extraction and sea level rise associated with a changing climate.
1. C. Small et al., Journal of coastal research, 584-599 (2003). 2. B. Bates et al., Technical Paper of the Intergovernmental Panel on Climate Change, IPCC Secretariat, Geneva 210, (2008). 3. V. Post, Hydrogeology Journal 13, 120-123 (2005). 4. T. H. Bakken et al., Water resources management 26, 1015-1026 (2012). 5. W. Burnett et al., Science of the total Environment 367, 498-543 (2006). 6. R. L. Evans, Geophysics 72, WA105-WA116 (2007). 7. V. E. Post et al., Nature 504, 71 (2013). 8. J. F. Bratton, The Journal of Geology 118, 565-575 (2010). 9. R. H. Johnston, Journal of Hydrology 61, 239-249 (1983). 10. H. Kooi et al., Journal of Hydrology 246, 19-35 (2001). 11. M. Person et al., Geological Society of America Bulletin 115, 1324-1343 (2003). 12. P. Leahy et al., EOS (American Geophysical Union Transactions) 63, 322 (1982). 13. D. Cohen et al., Groundwater 48, 143-158 (2010). 14. J. F. Adkins et al., Science 298, 1769-1773 (2002). 15. L. F. Greenlee et al., Water research 43, 2317-2348 (2009). 16. K. Ernstson et al., Groundwater geophysics: a tool for hydrogeology. Springer, Berlin, 179-226 (2006). 17. A. Steuer et al., Journal of Applied Geophysics 67, 194-205 (2009). 18. H. Müller et al., Geo-Marine Letters 31, 123-140 (2011). 19. J. Bear et al., (Springer Science & Business Media, 1999), vol. 14. 20. F. Hoefel et al., Estuarine, Coastal and Shelf Science 52, 179-189 (2001). 21. W. Greenwood et al., Groundwater 44, 292-299 (2006). 22. A. Haroon et al., Geophysics 83, 1-61 (2017). 23. B. Siemon et al., Geophysics 80, WB21-WB34 (2015). 24. V. Freedman et al., Advances in Water Resources 25, 439-453 (2002). 25. C. Langevin et al., (Lewis Publishers Boca Raton, FL, 2004).