Use of a simple model for studying oceanic tracer distributions and the global carbon cycle

We have studied, based on work of Shaffer and Sarmiento (1992), a model for simulating the transport of CO2 and tracers in the ocean (HILDA, for High-Latitude Exchange/Interior Diffusion-Advection Model) that combines features of box models and of the box-diffusion model. It is latitudinally divided into two zones; in the low latitudes, transport into the deep ocean occurs by eddy diffusion, while the high-latitude zone consists of two boxes (surface and deep ocean). We compare different ways of calibration and find that in order to reproduce the distributions of natural 14C as well as of bomb-produced 14C, the vertical eddy diffusivity K must decrease with depth. The concept of eddy diffusion is discussed by calculating apparent eddy diffusivities from 3-D model tracer simulations. The depth dependence of K is qualitatively confirmed by these calculations, reflecting the fact that the water circulation is more vigorous near the surface than at depth. We find that eddy diffusivities derived from 14C are not appropriate for representing the large-scale vertical temperature distribution, because the latitudinal distribution of temperature differs in a systematic way from that of 14C and also of anthropogenic CO2. Oceanic uptake of anthropogenic CO2, biospheric CO2 emissions and isotopic perturbations are calculated, based on the observed atmospheric CO2concentration history. The results indicate an oceanic uptake of 1.9 Gt C yr-1 in 1980 and a near-zero net contribution from the biota in the past several decades. The HILDA model is compared with other models, and we find that its response to atmospheric CO2 perturbations is rather similar to that of a 3-D ocean carbon cycle model of Sarmiento et al. (1992).