Submarine groundwater discharge of nutrients and copper to an urban subestuary of Chesapeake Bay (Elizabeth River)

We investigated the role submarine groundwater discharge (SGD) plays in the delivery of nutrients and copper to the Elizabeth River (Virginia) estuary, a major subestuary of lower Chesapeake Bay. Using an approach based on radium isotopes, we concluded that two distinct sources of groundwater were equally impacting the estuary: a surface (marsh) aquifer and deep aquifer source each with a unique 228Ra/226Ra activity ratio. Considering each of these sources, we calculated an SGD flux of 1 3 106 m3 d21 (610%), which represented ;6% of the SGD flux for the entire Chesapeake Bay and ;5% of the James River, a major source of freshwater to lower Chesapeake Bay. SGD-derived dissolved inorganic nitrogen (DIN) and dissolved inorganic phosphorus (DIP) fluxes averaged 4.5 (64.6) and 0.16 (60.17) mmol m22 d21, respectively, and compared well with area-normalized fluxes to Chesapeake Bay. In contrast, SGD-derived Cu input of 730 (6390) kg yr21 was a relatively small source of Cu (;3%) to the Elizabeth estuary given that surface water inputs, such as antifouling paints associated with naval operations, are a major component of the Cu budget for this system. These findings were in general agreement with prior studies of SGD for this region. Submarine groundwater discharge (SGD) is often ignored when constructing geochemical budgets for elements in nearshore environments, mainly because the volume flux is difficult to estimate. However, many studies indicate that SGD may carry significant quantities of nutrients and trace metals to the ocean (Simmons 1992; Moore 1996; Krest et al. 2000; Montlucon and Sanudo-Wilhelmy 2001). In the case of nutrients, SGD has been the principal mechanism for eutrophication in many coastal embayments throughout the world (e.g., Valiela et al. 1990). Various approaches for quantifying SGD include water budgets, seepage meters, and natural tracers (see review by Burnett et al. 2001). The water budget approach, which attempts to balance aquifer inputs (precipitation) with losses (evapotranspiration), often results in a large uncertainty, since SGD is the difference between two very large numbers (e.g., Cambareri and Eichner 1998). Seepage meters, though useful for understanding local-scale patterns of SGD, are labor intensive and therefore require significant effort for quantifying SGD on large spatial scales (e.g., Michael et al. 1 Corresponding author ([email protected]). Acknowledgments We thank those who assisted in the field and laboratory aspects of this research, including John Andrews, Craig Herbold, Holly Michael, Steve Pike, and Linda Rasmussen. Mollie Wolcott (Virginia Port Authority), Kevin Cloe (Norfolk Naval Base), Dave Nelms (USGS), and Randy McFarland (USGS) provided access to local wells. Lary Ball and David Schneider of the WHOI ICP-MS facility assisted with the copper analyses, and Jinfeng Wu (MIT) kindly shared his Cu isotope spike. Matt Allen assisted with figure preparation. The manuscript was improved by discussions with Billy Moore and comments by Edward Sholkovitz and two anonymous reviewers. This work was funded by a grant to K.O.B. and M.A.C. from the Office of Naval Research (N00014-99-1-0038) and fellowships to M.A.C. from the G. Unger Vetlesen Foundation and Coastal Ocean Institute. This is WHOI contribution number 10988. 2003). Natural tracers, the most valuable of which are orders of magnitude more enriched in groundwater relative to the receiving water body, are useful because their excess concentration in the study area of interest represents an integrated measure of SGD (e.g., Cable et al. 1996; Moore 1996). There are two general approaches to defining submarine groundwater discharge (SGD): (1) the amount of freshwater that enters the coastal ocean from a hydraulically connected aquifer or (2) the advective flow of mixtures of fresh and brackish waters into the coastal zone (Burnett et al. 2001). Most SGD derives from inland precipitation that recharges aquifers, which then flows into the sea. Freshwater flowing down gradient from the water table may either discharge from a seepage face at the shore or flow directly into the sea. The hydraulic gradient that drives freshwater toward the sea along the interface also drives saltwater back to sea, creating a saltwater circulation cell (Li et al. 1999). Hence, SGD often consists of a substantial amount of cycled seawater, which can significantly alter the chemical composition of the discharging fluid (Moore 1999; Charette and Sholkovitz 2002; Testa et al. 2002). Therefore, studies aimed at understanding chemical fluxes to the coastal zone derived from SGD must consider both fresh and saline groundwater. Radium isotopes have proved to be useful tracers of total SGD in many environments on both small and large scales from salt marshes (e.g., Rama and Moore 1996) to the continental shelf (Moore 1996). In addition to being orders of magnitude more enriched in groundwater relative to seawater, there are four isotopes of Ra (224Ra, t1/2 5 3.66 d; 223Ra, t1/2 5 11.4 d; 228Ra, t1/2 5 5.75 yr; 226Ra, t1/2 5 1,600 yr) with a wide range of half-lives that make Ra useful for evaluating sources and rates of SGD, as well as nutrients and trace metals carried by this process. The goal of this work was to investigate the role of SGD in the delivery of nutrients and copper to an urban estuary using radium isotopes. We chose the Elizabeth River, Vir377 Groundwater discharge in an estuary Fig. 1. Location of sampling stations in the Elizabeth estuary and its four main branches. ginia, which is characterized by a heavily urbanized and industrialized watershed that includes the largest naval port in the world. As a tributary of lower Chesapeake Bay, the river itself consists of a main stem and four major branches (Fig. 1). The shorelines of the main stem, Southern Branch, and Eastern Branch consist of a single deep (dredged) channel with heavily developed shorelines (e.g., industry, shipyards). In contrast, the Western Branch and the Lafayette River are significantly less industrialized, as evidenced by their abundance of natural salt marshes. The Southern Branch continues south into the Great Dismal Swamp, which connects the Intracoastal Waterway of Virginia and North Carolina. Tidal flushing controls water residence times in the Elizabeth River and freshwater influx, which is restricted by canal locks in the upper river. In combination with the industrial activity, slow flushing of the estuary has led to severe contamination of sediments and water column from a variety of trace metals and organic compounds (Rule 1986; Bender et al. 1988; Mitra et al. 1999; Diaz et al. 2003). Eastern Virginia is characterized by a coastal plain–type geology, with abundant sources of groundwater typical of other coastal plain regions. For the Elizabeth estuary, sources of groundwater include the Columbia and Yorktown aquifers. The Columbia aquifer is shallow and unconfined, consisting of thin, discontinuous layers of sand and shell lenses. The hydraulic conductivity ranges from 28 to 1,200 cm d21 (Meng and Harsh 1988). The underlying Yorktown aquifer, confined by beds of silt and clay, consists of fine shelly sand, silty sand, and shell beds. It is breached in certain areas due to stream erosion and possibly dredging, which is common practice to maintain navigation channels for naval operations. Hydraulic conductivities are 2–5 orders of magnitude lower than in the Columbia aquifer, ranging from 0.01 to 0.63 cm d21 (Meng and Harsh 1988).