models have played a particularly important role in the development of oceano-
graphy and, in particular, the role of the ocean in climate.
The basic tool of climate studies is the coupled model, where the various compo-
nents of the climate system—the atmosphere, ocean, cryosphere, biosphere, and
chemosphere—are simultaneously and consistently coupled. The ocean component
of such a coupled model exchanges its heat, fresh water, and momentum with the
atmosphere at the sea surface. The test of the successful coupling of the atmosphere
and ocean is the correct simulation of the time-varying sea surface temperature and
surface winds, both of which are relatively easy to measure: Directly by ship or
mooring, remotely by satellite, or by a combination of the two.
The development of off-line ocean-only models requires the heat, momentum,
and freshwater forcing from the atmosphere to be known. Since precipitation and
evaporation, in particular, are so poorly measured over the ocean, it is a continual
struggle to know whether errors in the ocean model are due to errors in the model
itself or errors in the forcing of the ocean by the atmosphere.
Ocean models themselves are relatively simple in concept: The known equations
of water and salt are discretized and time stepped into the future. The discretization
process requires a trade-off between fine resolution for accuracy and the need to
simulate over long periods of time, which, because of limited computer resources,
requires coarser resolution. While the equation of state of seawater relating density
to salt, temperature, and pressure cannot be written down simply, it has, over the
course of time, become known to high accuracy.
What makes ocean modeling difficult is the specification of those mixing
processes that unavoidably cannot be resolved by whatever resolution is chosen.
We are beginning to understand that enhanced small-scale mixing occurs near
bottom topography and near boundaries: Purposeful release experiments, where a
dye is released and then followed in time to see how the dye cloud evolves, has
revealed this to us. Larger scale mixing, where parcels are interchanged because of
the large-scale circulation (but still unresolved by the ocean models) itself is more
problematic, but recent advances in parameterizing these unresolved mixing effects
have shown promise.
6 THE FUTURE OF THE OCEAN IN CLIMATE
It is clear that the ocean is a crucial component of the climate system. Since so much
of what is not known about the past and future of the climate system depends on
active interactions with the ocean, it is clear that we have to learn more about
its essential processes. How to go about learning about the ocean is the difficult
question.
Direct measurements are very expensive, and satellites, while giving a global look
at the entire ocean, see only its surface. Designs are currently underway for a Global
Ocean Observing System (GOOS), but the cost of implementing such a system in
toto is prohibitive, even if shared among the wealthier countries of the world.
132 THE OCEAN IN CLIMATE