Sensitivity of CFC-11 uptake to physical initial conditions and interannually varying surface forcing in a global ocean model

Sensitivity of the oceanic chlorofluorocarbon CFC-11 uptake to physical initial conditions and surface dynamical forcing (heat and salt fluxes and wind stress) is investigated in a global ocean model used in climate studies. Two different initial conditions are used: a solution following a short integration starting with observed temperature and salinity and zero velocities, and the quasi-equilibrium solution of an independent integration. For surface dynamical forcing, recently developed normal-year and interannually varying (1958–2000) data sets are used. The model CFC-11 global and basin inventories, particularly in the normal-year forcing case, are below the observed mean estimates, but they remain within the observational error bars. Column inventory spatial distributions indicate nontrivial differences due to both initial condition and forcing changes, particularly in the northern North Atlantic and Southern Ocean. These differences are larger between forcing sensitivity experiments than between the initial condition cases. The comparisons along the A16N and SR3 WOCE sections also show differences between cases. However, comparisons with observations do not clearly favor a particular case, and model–observation differences remain much larger than model–model differences for all simulations. The choice of initial condition does not significantly change the CFC-11 distributions. Both because of locally large differences between normal-year and interannually varying simulations and because the dynamical and CFC-11 forcing calendars are synchronized, we favor using the more realistic interannually varying forcing in future simulations, given the availability of the forcing data sets.