The currently observed increase in atmospheric CO2 due anthropogenic emissions is substantially slowed by natural processes that incorporate CO2 into the terrestrial biota and the ocean. Year-to-year changes in the CO2 growth rate that exceed variations in the fossil fuel source indicate a significant variability in these global CO2 sinks. However, the enormous complexity of the terrestrial and oceanic biogeochemical systems that absorb atmospheric CO2 makes these sinks extremely difficult to understand and precisely quantify. Many techniques, including the interpretation of the relative changes in atmospheric CO2 and O2/N2, ocean modeling, and atmospheric data inversions, have been employed to estimate the mean and variability of global CO2 sinks. However, uncertainty remains large. The goal of this thesis is to improve understanding of global CO2 sinks by considering (1) the error in the atmospheric O2/N2 partitioning method due to the neglect of interannual variability in the air-sea fluxes of 02, and (2) the interannual variability of the ocean CO2 sink.
(cont.) A global, high-resolution ocean general circulation model is used to estimate the magnitude and understand the mechanisms of interannual variability in air-sea fluxes of both CO2 and 02. I find that the global variability in the fluxes of both gases are dominantly forced by large-scale physical processes governing upper ocean dynamics, particularly El Nifio / Southern Oscillation (ENSO) and, for 02, the North Atlantic Oscillation (NAO). Estimates of the extremes of CO2 and 02 flux variability for the period 1980-1998 are +/-0.5x1015 grams Carbon/yr (PgC/yr) and -70/+100x1012 mol/yr (Tmol/yr), respectively. Global 02 flux variability implies up to a 1.0 PgC/yr error in estimates of interannual variability in land and ocean CO2 sinks derived from atmospheric 02/N2 observations. This error is significant for estimates of annual sinks, but it is cumulatively negligible for estimates of mean sinks from October 1991 to April 1998. Increasing convergence of estimates of land.