The ocean's memory of the atmosphere: residence-time distribution and water-mass ventilation (formerly paper 5.3)

Measurements of transient tracers such as tritium, CFCs and C-14 are useful for estimating water-mass formation rates because they yield information about the time since fluid elements had last contact with the atmosphere. However, relating the budget of an invading tracer to water-mass formation rates is complicated by the fact that the eddy diffusivity and tracer gradients are only poorly known. To interpret tracer measurements in terms of water-mass formation rates one must use models of the ventilation process that make assumptions (often implicit) about the importance of eddy-diffusive fluxes. Because the assumptions made in different studies are not necessarily consistent with each other or with the real ocean, estimates of water-mass formation based on different tracers can yield different results.

To overcome some of the difficulties associated with the interpretation of tracer data, we develop a diagnostic of the ventilation rate into and out of the ocean interior that is both tracer- and model-independent. To this end, we introduce the joint distribution, R, of residence time and surface-ventilation location. This allows us to partition the ventilation flux, φ, according to where fluid elements enter and exit the ocean interior and the corresponding residence time in the interior. For overlapping regions of ventilation into and out of the interior both φ and R are singular at zero residence time. The singularity is due to eddy diffusion which dominates transport for short residence times. While R is always properly normalized, the residence-time integrated ventilation flux is infinite when entry and exit regions coincide. The singular nature of surface-to-surface transport demands caution when interpreting ventilation rates obtained from simple box models.

We illustrate the rich detail R and φ provide about the way the ocean communicates with the surface by computing R and φ with an off-line time-averaged global OGCM. Our diagnostics yield information consistent with current knowledge of the global circulation while quantifying surface-to-surface transport in detail.