391
latitudes. He showed that if the time scales characteristic for temperature and salinity anomalies are different, multiple steady states as well as
self-sustained oscillations are possible.
In more complex, 2- and 3-dimensional models mechanisms generating natural variability are more difficult to understand, and the range of mechanisms and time scales is quite broad. At present, we do not have a unifying
theory explaining natural variability on the century time scale but we are
in the stage of collecting evidence for such fluctuations both from the observational and the modeling side. It is hoped that over the coming years
the representation of simple atmospheres used in driving ocean models becomes more realistic and that with this improvement, a more consistent
picture will emerge. For now, a list of the type and time scales of variability found in numerical models (ocean, atmosphere, coupled) helps us
in discussing physically plausible mechanisms of variability.
Table 1 gives a summary of self-sustained oscillations found in a number
of ocean, atmosphere and coupled models. We focus only on the most
robust cycles in these models. The decadal-to-century time scale does in
most cases include the ocean circulation, in particular the thermohaline
part of it. Mechanisms are connected with mainly with the thermohaline
but also with the wind-driven circulation as well as with the hydrological
cycle. Note that long-term variability is also found in atmospheric GeMs
suggesting interesting possibilities of interaction between the atmosphere
and the ocean also on interdecadal time scales.
latitudes. He showed that if the time scales characteristic for temperature and salinity anomalies are different, multiple steady states as well as
self-sustained oscillations are possible.
In more complex, 2- and 3-dimensional models mechanisms generating natural variability are more difficult to understand, and the range of mechanisms and time scales is quite broad. At present, we do not have a unifying
theory explaining natural variability on the century time scale but we are
in the stage of collecting evidence for such fluctuations both from the observational and the modeling side. It is hoped that over the coming years
the representation of simple atmospheres used in driving ocean models becomes more realistic and that with this improvement, a more consistent
picture will emerge. For now, a list of the type and time scales of variability found in numerical models (ocean, atmosphere, coupled) helps us
in discussing physically plausible mechanisms of variability.
Table 1 gives a summary of self-sustained oscillations found in a number
of ocean, atmosphere and coupled models. We focus only on the most
robust cycles in these models. The decadal-to-century time scale does in
most cases include the ocean circulation, in particular the thermohaline
part of it. Mechanisms are connected with mainly with the thermohaline
but also with the wind-driven circulation as well as with the hydrological
cycle. Note that long-term variability is also found in atmospheric GeMs
suggesting interesting possibilities of interaction between the atmosphere
and the ocean also on interdecadal time scales.