Coral skeletal Sr/Ca is a widely applied proxy indicator for tropical sea surface temperature (SST) because the elemental ratio in coral aragonite is influenced by both SST and seawater Sr/Ca. Application of the methodology assumes that seawater Sr/Ca is a constant in coral reef environments, and the ratio can be used to solve for paleotemperatures based upon an established coral Sr/Ca – SST relationship. This study tests that assumption by documenting seawater Sr/Ca variability in the Florida Keys and in the U.S. and British Virgin Islands using continuous osmotic pumps. Samples are analyzed for Sr/Ca using a novel method via Inductively Coupled Plasma – Atomic Emission Spectrometry. While mean seawater Sr/Ca did not vary significantly between sites, all sites exhibited significant annual variability (~0.5 – 0.1 mmol/mol), with the greatest variability observed in locations most impacted by freshwater discharge. These findings correspond to large temperature offsets (>2 degrees Celsius) in standard coral Sr/Ca-based SST reconstructions.