Submarine groundwater discharge (SGD) is an important source of dissolved elements to the ocean, yet little is known regarding the chemical reactions that control their flux from coastal aquifers. The net flux of elements derived from SGD is dependent on biogeochemical reactions in the groundwater-seawater mixing zone, termed the subterranean estuary (STE). Here, we present results on STE trace metal cycling in two contrasting coastal aquifers, Waquoit Bay, MA, USA (unconfined, sandy-type) and Flic en Flac, Mauritius (volcanic rock-type). The elements were chosen based on their susceptibility to redox (Fe, Mn, U) and ion exchange (Sr, Ba, Ra) driven reactions. In the Waquoit Bay STE, SGD-driven recirculation of seawater through reducing permeable sediments makes this environment a net sink for uranium from coastal seawater. This finding is supported by surface water concentrations of U in the bay, which were significantly depleted in U compared with adjacent coastal waters. In contrast, large-scale release of barium and radium occurs in the mid to high salinity zone of the subterranean estuary, leading to a net input of these elements to the coastal ocean. Surprisingly, we observed non-conservative release of Sr in the STE, though the magnitude of the release was significantly less than that of its alkaline earth neighbor Ba. In the Mauritius STE, significantly less deviation from conservative mixing was observed for the aforementioned elements. Recent studies suggest that volcanic islands, with their higher than average rainfall and relief, may be disproportionately large sources of SGD to the oceans. Our results suggest that, while such islands may have enhanced SGD rates, they are not likely to represent similarly large sinks (U) and sources (Sr, Ba, Ra) of trace metals to the oceans. In addition to salt-induced desorption of Ba/Ra from aquifer sediments, reductive dissolution of Mn oxides and weathering processes may explain the distribution of these elements and others in coastal groundwater. Given the non-conservative removal of U observed in these and other STEs around the world, we calculate the global removal of U due to SGD at 20 x 10 super(6) mol U per year, which is the same order of magnitude as other major U sinks for the ocean. These results suggest a need to revisit and reevaluate the oceanic budgets for elements that are likely influenced by SGD-associated processes.