Evaluating the source and seasonality of submarine groundwater discharge using a radon-222 pore water transport model Christopher G. Smith a , Jaye E. Cable a, ⁎, Jonathan B. Martin b , Moutusi Roy b a Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803, United States b Department of Geological Sciences, University of Florida, Gainesville, FL 32611, United States ABSTRACT ARTICLE INFO Article history: Received 7 February 2008 Received in revised form 21 June 2008 Accepted 25 June 2008 Available online 11 July 2008 Editor: M.L. Delaney Keywords: radon radium submarine groundwater discharge subterranean estuary pore water model Indian River Lagoon non-local exchange Pore water radon ( 222 Rn) distributions from Indian River Lagoon, Florida, are characterized by three zones: a lower zone where pore water 222 Rn and sediment-bound radium ( 226 Ra) are in equilibrium and concentration gradients are vertical; a middle zone where 222 Rn is in excess of sediment-bound 226 Ra and concentration gradients are concave-downward; and an upper zone where 222 Rn concentration gradients are nearly vertical. These 222 Rn data are simulated in a one-dimensional numerical model including advection, diffusion, and non-local exchange to estimate magnitudes of submarine groundwater discharge components (fresh or marine). The numerical model estimates three parameters, fresh groundwater seepage velocity, irrigation intensity, and irrigation attenuation, using two Monte Carlo (MC) simulations that (1) ensure the minimization algorithm converges on a global minimum of the merit function and the parameter estimates are consistent within this global minimum, and (2) provide 90% confidence intervals on the parameter estimates using the measured 222 Rn activity variance. Model estimates of seepage velocities and discharge agree with previous estimates obtained from numerical groundwater flow models and seepage meter measurements and show the fresh water component decreases offshore and varies seasonally by a factor of nine or less. Comparison between the discharge estimates and precipitation patterns suggests a mean residence time in unsaturated and saturated zones on the order of 5 to 7 months. Irrigation rates generally decrease offshore for all sampling periods. The mean irrigation rate is approximately three times greater than the mean seepage velocity although the ranges of irrigation rates and seepage velocities are the same. Possible mechanisms for irrigation include density-driven convection, wave pumping, and bio-irrigation. Simulation of both advection and irrigation allows the separation of submarine groundwater discharge into fresh groundwater and (re)circulated lagoon water. © 2008 Elsevier B.V. All rights reserved. 1. Introduction Groundwater–aquifer systems connect recharge, storativity, trans- missivity, and discharge; understanding one or more of these hydro- geologic traits helps us better predict responses to perturbations (Bredehoeft, 2007). For example, groundwater discharge to marine coastal regions, typically referred to as submarine groundwater discharge (SGD), can contain both marine and fresh water components, of which the fresh component is significant as a geochemical sink from terrestrial aquifers and source to surface waters. The relative fractions of these two components are critical considering approximately 23% of the world's population now lives within 100 km of the coast, stressing regional freshwater resources. This problem will likely be exacerbated given the predictions for climate change consequences to the hydrologic cycle (IPCC, 2007). The marine component of SGD includes deep recirculating seawater and shallow pore water exchange across the sediment–water interface (Burnett et al., 2003). Each component has a different geochemistry (i.e., pH, Eh, and ionic strength), and where these waters mix prior to discharging, they form reactive zones known as the “subterranean estuary” (Moore, 1999). These dynamic interfaces affect the transport and transformation of dissolved constituents and impose source/sink limitations on the use of geochemical tracers to estimate SGD to coastal waters (Moore, 1999). Comparing SGD from coastal aquifers with disparate character- istics requires a method that is independent of climate and geological variability. Methods used include geochemical tracers (primarily radon and radium isotopes), water budgets, numerical models, and direct measurements (e.g., seepage meters), and except for the water budget approach, no universal method exists. When multiple methods are applied at the same site, discharge estimates may vary by an order of magnitude or more, because the most commonly applied methods measure different components of SGD (e.g. Taniguchi et al., 2002; Cable et al., 2004 and references therein). This problem has been studied systematically at several coasts over the past five years (Burnett et al., 2006; Mulligan and Charette, 2006; Martin et al., 2007). In these studies, SGD rates estimated from a radon ( 222 Rn) water column mass balance model were similar to those measured Earth and Planetary Science Letters 273 (2008) 312–322 ⁎ Corresponding author. Fax: +1 225 578 6326. E-mail address: jcable@lsu.edu (J.E. Cable). 0012-821X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2008.06.043 Contents lists available at ScienceDirect Earth and Planetary Science Letters journal homepage: www.elsevier.com/locate/epsl