Geochemical tracer data (i.e., ²²²Rn and four naturally occurring Ra isotopes), electromagnetic (EM) seepage meter results, and high-resolution, stationary electrical resistivity images were used to examine the bi-directional (i.e., submarine groundwater discharge and recharge) exchange of a coastal aquifer with seawater. Our study site for these experiments was Lynch Cove, the terminus of Hood Canal, WA, where fjord-like conditions dramatically limit water column circulation that can lead to recurring summer-time hypoxic events. In such a system a precise nutrient budget may be particularly sensitive to groundwater-derived nutrient loading. Shore-perpendicular time-series subsurface resistivity profiles show clear, decimeter-scale tidal modulation of the coastal aquifer in response to large, regional hydraulic gradients, hydrologically transmissive glacial terrain, and large (4-5 m) tidal amplitudes. A 5-day ²²²Rn time-series shows a strong inverse covariance between ²²²Rn activities (0.5−29 dpm L^-1) and water level fluctuations, and provides compelling evidence for tidally modulated exchange of groundwater across the sediment/water interface. Mean Rn-derived submarine groundwater discharge (SGD) rates of 85 ± 84 cm d^-1 agree closely in the timing and magnitude with EM seepage meter results that showed discharge during low tide and recharge during high tide events. To evaluate the importance of fresh versus saline SGD, Rn-derived SGD rates (as a proxy of total SGD) were compared to excess 226Ra-derived SGD rates (as a proxy for the saline contribution of SGD). The calculated SGD rates, which include a significant (>80%) component of recycled seawater, are used to estimate associated nutrient (NH⁴⁺, Si, PO₄^3-, NO₃ + NO₂, TDN) loads to Lynch Cove. The dissolved inorganic nitrogen (DIN = NH₄ + NO₂ + NO₃) SGD loading estimate of 5.9 × 10⁴ mol d^-1 is 1−2 orders of magnitude larger than similar estimates derived from atmospheric deposition and surface water runoff, respectively.