Submarine groundwater discharge: An important source of new inorganic nitrogen to coral reef ecosystems

Using radium (Ra) isotopes and nutrient analyses, we found that submarine groundwater discharge (SGD) is an important source of ‘‘new’’ nutrients, particularly nitrogen, to coral reefs around the world. Nitrogen input estimates associated with SGD range from 3 to 800 mmol h21 per meter of shoreline. The use of Ra isotopes allows us to quantify the inorganic nitrogen input from this source of nutrients. Increasing coastal population and land use practices may enhance anthropogenic nutrient loading from submarine groundwater contributing to reef degradation. The relationship between nutrient dynamics and productivity in coral reef systems has received considerable attention. The impetus for this is the contrast between the high productivity and biomass of these systems and the typically clear, nutrient-poor surface waters bathing them (Marsh 1977; D’Eliaet al. 1981). In addition to efficient recycling of nutrients (Dubinsky 1990), coral reefs derive new nutrients by effective acquisition of both particulate and dissolved sources from seawater impinging on the reefs; vigorous water circulation and flow (Larend and Atkinson 1997); nitrogen (N) fixation (Weibeet al. 1975); and dust deposition (Hingaet al. 1991). Terrestrially derived input from submarine groundwater discharge (SGD) has been indicated as an important source of nutrients to coastal systems in general (Corbett et al. 1999; Umezawa et al. 2002; Garrison et al. 2003) and to coral reefs in particular (Valiela et al. 1990; Paerl 1997). However, it has been notoriously difficult to track non–point source groundwater as it moves into coastal seas and to describe the interactions between fresh groundwater and seawater at the land–sea interface (Burnett et al. 2002). Accordingly, direct measurements of SGD to coastal reef systems have not been extensive, and the role of SGD as a source of nutrients to coral reef ecosystems has not been extensively and quantitatively determined. Discharge of groundwater into the sea is widespread; it occurs anywhere that an aquifer is connected hydraulically with the sea through permeable sediments or rocks and where the aquifer head is above sea level. Submarine groundwater flows into the coast at the interface between freshwater and seawater (the mixing zone) where the unconfined aquifer outcrops at the beach (Reay et al. 1992). Toward the seaward edge of the mixing zone, water is brackish as a result of intrusion of salt water through permeable aquifer mixing as well as wave and tidal pumping (Li et al. 1999). The chemistry of the water in the mixing zone is altered such that it is chemically different than either the terrestrial freshwater or seawater components (Church 1996). This area has been referred to as the subterranean estuary (Moore 2003). Accordingly, the term SGD, as used in this article, does not refer to freshwater (meteoric) input but rather includes the freshwater-recirculated seawater mixture that is discharging at the coastline (Buddemeier 1996). Over the last few decades, Moore and collaborators have pioneered the use of the quartet of naturally occurring radium (Ra) isotopes as tracers for ocean mixing and saline submarine groundwater input to coastal systems (Krest and Harvey 2003). The divalent cation Ra isotopes are bound to soil particles and rocks in freshwater. They readily desorb via ion exchange in the presence of solutions of higher ionic strength (Webster et al. 1994; Yang et al. 2002). Accordingly, in coastal aquifers, where seawater with high ionic strength mixes and interacts with freshwater and aquifer rocks, waters enriched in Ra are observed (Moore 2003). Open seawater, on the other hand, has very low or constant Ra activities. Therefore, excess Ra (over the open seawater activities) indicates a coastal source that, in many cases, is due to SGD. Ra isotopes are excellent tracers for the study of SGD and mixing in coastal systems because of the distinct difference in activities between the end-member sources (e.g., open ocean and terrestrial saline waters) and because they behave conservatively after leaving the aquifer (accounting for radioactive decay). In addition, the use of Ra isotopes has advantages over other techniques used for quantifying SGD related fluxes, since it allows for temporal and spatial integration over the mean-life of the radionuclides (Moore 2003), and the different timescales of decay are useful as mixing tracers. Indeed, Ra isotopes have been extensively used to determine the discharge of freshwater, nutrients (Cable et al. 1996; Krest et al. 2000; Kelly and Moran 2002), and other dissolved constituents to the coastal ocean (Shaw et al. 1998). However, SGD-associated nutrient supply to coral reef systems has not been fully evaluated using Ra isotopes. Recently, concern that fringing reefs are degrading through human effects, particularly as a result of increases in terrestrial-derived inputs of nutrients and sediments, has been raised (Wilkinson 1999). Therefore, evaluating and, particularly, quantifying the contribution of SGD-associated nutrient input to fringing reef systems assume critical importance. Methods—To determine if submarine groundwater is discharging at the beach and reaching the reefs, water samples were taken along several transects from the water line to some distance offshore (typically within 100 m from the shoreline) at representative fringing reef sites around the world (Fig. 1). Other sites (Mexico, Heron Island in Australia) were also examined, and preliminary results are consistent with the data presented here; however, these data are not included because only a limited number of samples were collected. Only one representative transect is shown in the