Climate Variability and Radiocarbon in the CM2Mc Earth System Model

The distribution of radiocarbon (C) in the ocean and atmosphere has fluctuated on time scales ranging from seasons to millennia. It is thought that these fluctuations partly reflect variability in the climate system, offering a rich potential source of information to help understand mechanisms of past climate change. Here, a long simulation with a new, coupled model is used to explore the mechanisms that redistribute C within the earth system on interannual to centennial time scales. The model, the Geophysical Fluid Dynamics Laboratory Climate Model version 2 (GFDL CM2) with Modular Ocean Model version 4p1(MOM4p1) at coarseresolution (CM2Mc), is a lower-resolution version of the Geophysical Fluid Dynamics Laboratory’s CM2M model, uses no flux adjustments, and is run here with a simple prognostic ocean biogeochemistry model including C. The atmospheric C and radiative boundary conditions are held constant so that the oceanic distribution of C is only a function of internal climate variability. The simulation displays previously described relationships between tropical sea surface C and the model equivalents of the El Niño–Southern Oscillation and Indonesian Throughflow. Sea surface C variability also arises from fluctuations in the circulations of the subarctic Pacific and Southern Ocean, including North Pacific decadal variability and episodic ventilation events in the Weddell Sea that are reminiscent of the Weddell Polynya of 1974–76. Interannual variability in the air–sea balance of C is dominated by exchange within the belt of intense ‘‘Southern Westerly’’ winds, rather than at the convective locations where the surface C is most variable. Despite significant interannual variability, the simulated impact on air–sea exchange is an order of magnitude smaller than the recorded atmospheric C variability of the past millennium. This result partly reflects the importance of variability in the production rate of C in determining atmospheric C but may also reflect an underestimate of natural climate variability, particularly in the Southern Westerly winds.