Age Tracers in an Ocean GCM
Observations of transient tracers such as tritium and helium-3 Full-size image (<1 K) are frequently combined to construct “age-like” quantities generally interpreted to represent time elapsed since a fluid parcel was last at the surface. In a turbulent (“diffusive”) environment such as the ocean...
| Published in: | Deep Sea Research Part I: Oceanographic Research Papers |
|---|---|
| Main Authors: | , , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Elsevier
2001
|
| Subjects: | |
| Online Access: | https://oceanrep.geomar.de/id/eprint/4738/ https://oceanrep.geomar.de/id/eprint/4738/1/1-s2.0-S0967063700000947-main.pdf https://doi.org/10.1016/S0967-0637(00)00094-7 |
| Summary: | Observations of transient tracers such as tritium and helium-3 Full-size image (<1 K) are frequently combined to construct “age-like” quantities generally interpreted to represent time elapsed since a fluid parcel was last at the surface. In a turbulent (“diffusive”) environment such as the ocean, we must regard the fluid parcel as being composed of material fluid elements that have spent different lengths of time since their last contact with the surface. Hence, they are characterized by an age spectrum or distribution of transit times. In this study we explore the concepts of tracer-derived “ages” and the transit-time probability density function (PDF) with the aim of improving our understanding of their interpretation. Using an ocean general circulation model, we illustrate the effect of mixing on tracer-derived “ages” within the Atlantic Ocean. The mixing biases such ages towards younger values with respect to the ideal or mean age of a water parcel. In the North Atlantic, this bias is particularly pronounced in the thermocline because of large vertical gradients in tracer concentration, and in the deep ocean, where the penetration of recently ventilated water creates large gradients along the isopycnal surfaces. In contrast, the effect of mixing appears to be relatively small in the subtropical subduction region. Calculations of the transit-time PDF in the ocean model show, however, that the mean age can potentially be very large because of contributions from long transit-time pathways, in spite of the fact that such pathways make up a small fraction of the fluid parcel. These results illustrate the key idea that tracer-derived ages are weighted towards the leading part of the transit-time distribution, while the ideal age is more sensitive to its “tail”. These tracers are thus sensitive to and help constrain different time scales. We also find that the ideal age converges much more rapidly to the mean age compared with the first moment of the age spectrum, an important consideration in numerical studies. |
|---|