Influence of mixing on CFC uptake and CFC ages in the North Pacific thermocline

[1] A diagnostic, isopycnal advection-diffusion model based on a climatological, geostrophic flow field is used to study the uptake of chlorofluorocarbons (CFCs) into the portion of the thermocline that outcrops in the open North Pacific (sigma(theta) less than or equal to 26.6 kg m(-3)). In addition to advection, isopycnal diffusion is required to match the CFC data collected during the World Ocean Circulation Experiment (WOCE) in the early 1990s. Using reduced outcrop saturations of 80 - 95% for isopycnals outcropping in the northwestern North Pacific (sigma(theta) greater than or equal to 25.4 kg m(-3)), together with an isopcynal interior diffusivity of 2000 m(2) s(-1) and enhanced diffusion (5000 m(2) s(-1)) in the Kuroshio Extension region, further improves the model-data agreement. Along-isopycnal diffusion is particularly important for isopycnals with shadow zones/pool regions in the western subtropical North Pacific that are isolated from direct advective ventilation. The isopycnal mixing causes an estimated increase in CFC-12 inventories on these isopycnals, compared to advection only, ranging from 10 - 20% (sigma(theta) = 25.6 kg m(-3)) to 50 - 130% (sigma(theta) = 26.6 kg m(-3)) over the subtropics in 1993. This contribution has important consequences for subduction rate estimates derived from CFC inventories and for the location of the subsurface CFC maxima. When tracer ages are derived from the modeled CFC distributions, time-evolving mixing biases become apparent that reflect the nonlinearities in the atmospheric CFC time histories. Comparison with model-calculated ideal ages suggests that during the time of WOCE ( similar to 1993), ventilation ages based on CFC-12 were biased young by as much as 16 - 24 years for pCFC-12 ages of 25 years, underestimating ideal ages by as much as 40 - 50%.