 |
| Titel |
Evaluating coupled ecosystem ocean circulation models: can we trust the concept of preformed nutrients? |
| VerfasserIn |
Wolfgang Koeve, Heiner Dietze |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250056551
|
|
|
|
|
|
| Zusammenfassung |
| It is common practice to separate inorganic nutrient concentrations in the interior of the ocean
into a preformed and a remineralised pool by following an "oxygen saturation approach"
(based on additional measurements of oxygen, temperature and salinity). By definition, the
preformed pool is exported from the surface to the deep ocean (by subduction or mixing)
during water mass formation while the remineralised pool constitutes the accumulated signal
of organic material remineralised along the way from the surface to the sampling location.
The concept is increasingly popular as it helps, e.g., to disentangle physical and
biogeochemical deficiencies of coupled ecosystem-circulation models. However, the
concept is based on a number of assumptions which are not always met. Among them
are, (1) perfect oxygen saturation during water mass formation, (2) insignificant
subsurface warming, and (3) linear mixing of source waters. Only a limited number of
studies have evaluated this approach systematically, e.g. using biogeochemical ocean
models run into steady state. Based on results from a coupled ecosystem-circulation
model which has additional, explicit, representations of preformed/remineralized
nutrients we examine pitfalls of the "oxygen saturation approach". The volume
weighted distribution of total phosphate (preformed + remineralised) from our model
compares well with observed phosphate (World Ocean Atlas). At the same time
the distributions of preformed phosphate computed using the oxygen saturation
approach clearly differs between model and data, suggesting the need to compare
preformed and remineralised nutrients independently. However, comparing preformed
nutrients from our explicit tracer and those computed via the oxygen saturation
approach tells that the interpretation of preformed nutrients computed from data or
model output may be ambiguous. We discuss implications for model-observation
intercomparisons. |
|
|
|
|
|
|