It is uncertain whether coral reef ecosystems are oceanic sources or sinks of carbon dioxide (CO2). Understanding the complex interactions between biogeochemical and physical processes within reef ecosystems is important for determining the contribution of coral reefs to the global carbon cycle and the air-sea Bur of CO2. The influence of biogeochemical and physical processes on CO2 cycling was examined for 1 month at Hog Reef Flat, part of the rim reef of Bermuda. The fugacity of seawater CO2 (fCO(2)) was measured hourly by use of a Carbon Interface Ocean Atmosphere buoy, providing the longest time series of seawater CO2 data on a coral reef ecosystem. Seawater fCO(2) ranged from similar to 340 to 470 mu atm, with a diurnal variability ranging from similar to 20-60 mu atm. Air-sea CO2 was directed from ocean to atmosphere with a mean flux of 3.3 +/- 4.6 mmoles CO2 m(-2) d(-1). The reef data are compared with a seawater CO, time series collected at the U.S. Joint Global Ocean Flux Study Bermuda Atlantic Time-series study (BATS) site (31 degrees 50 'N, 64 degrees 10 'W) in the Sargasso Sea surrounding Bermuda. Sargasso Sea waters are the original source for Bermuda platform water, providing a context for understanding the biogeochemical modification of reef water at Hog Reef Flat. Seawater fCO(2) at Hog Reef Flat was elevated relative to the Sargasso Sea by similar to0 to similar to 120 mu atm, primarily as a result of calcium carbonate production. However, the ability of the reef to act either as a source or sink of CO2 to/from the atmosphere largely depended on the air-sea CO2 disequilibrium of offshore Sargasso Sea waters impinging on the reef site. This study also revealed that an assessment of the fate of CO2 on coral reefs is dependent on understanding the reef's physical regime and forcing. The dynamics of wind, tide, platform circulation, and fluxes of offshore or onshore waters;Ire necessary context for all coral reef sites.