191
salt (fresh) anomaly is at low (high) latitudes, the circulation slows down
and the flux boundary condition becomes more important for determining
salinity, thereby enhancing the anomaly. When the polar (low latitude)
ocean has the salty (fresh) anomaly, overturning increases, and the flux
boundary condition, which is now acting to reduce the anomaly, becomes
less effective. This mechanism has also been noted by Mikolajewicz and
Maier-Reimer (1990) in their model response to stochastic forcing. Simple
model interpretations of loop oscillations have been carried out (Welander,
1986; Winton and Sarachik, 1993), which usually determine the period of
loop oscillation as the time it takes the anomaly to travel around the mean
circulation loop. Note that the temperature variability involved in loop
oscillations are small since the salinity anomaly is the dominant factor in
advection feedback, hence variability in the surface heat flux is also small.
6 Coupled Atmosphere-Ocean Modes
The practical problems of coupled atmosphere-ocean modeling make it difficult to simulate long runs of coupled models. Low resolution atmospheric
models (R15-approximately 50 x 80) have now been used to examine variability in the climate system represented by these models. There have
recently been a number of relatively long simulations using such low resolution coupled models and this sections will be based on those results.
In order to have a truly coupled atmosphere-ocean mode, the oceanic
SST has to induce atmospheric heat and momentum fluxes that reinforce
the SST. The problem is that most higher latitude modes require higher
latitude SST to affect the atmosphere and this influence has been difficult
to detect. Lau and Nath (1994) and Graham et al. (1994) have indicated
that the tropical SST dominates the effects on the atmosphere over the
mid-latitude SST.
It should be realized that the definition of a coupled mode, a mode that
depends critically on the interaction of the atmosphere and ocean, is far
more difficult in mid-latitudes than our favorite example of ENSO. A great
deal of analytic work on ENSO has shown that coupled ENSO modes arise
solely from the interaction of the atmosphere and the ocean and would
not exist without this interaction (e.g. Hirst, 1986). In midlatitudes, we
have two possibilities that can masquerade as a coupled mode: the ocean
can have decadal variability which then, through its SST expression, drives
atmospheric motions on decadal scales, or, more likely, the atmosphere can
salt (fresh) anomaly is at low (high) latitudes, the circulation slows down
and the flux boundary condition becomes more important for determining
salinity, thereby enhancing the anomaly. When the polar (low latitude)
ocean has the salty (fresh) anomaly, overturning increases, and the flux
boundary condition, which is now acting to reduce the anomaly, becomes
less effective. This mechanism has also been noted by Mikolajewicz and
Maier-Reimer (1990) in their model response to stochastic forcing. Simple
model interpretations of loop oscillations have been carried out (Welander,
1986; Winton and Sarachik, 1993), which usually determine the period of
loop oscillation as the time it takes the anomaly to travel around the mean
circulation loop. Note that the temperature variability involved in loop
oscillations are small since the salinity anomaly is the dominant factor in
advection feedback, hence variability in the surface heat flux is also small.
6 Coupled Atmosphere-Ocean Modes
The practical problems of coupled atmosphere-ocean modeling make it difficult to simulate long runs of coupled models. Low resolution atmospheric
models (R15-approximately 50 x 80) have now been used to examine variability in the climate system represented by these models. There have
recently been a number of relatively long simulations using such low resolution coupled models and this sections will be based on those results.
In order to have a truly coupled atmosphere-ocean mode, the oceanic
SST has to induce atmospheric heat and momentum fluxes that reinforce
the SST. The problem is that most higher latitude modes require higher
latitude SST to affect the atmosphere and this influence has been difficult
to detect. Lau and Nath (1994) and Graham et al. (1994) have indicated
that the tropical SST dominates the effects on the atmosphere over the
mid-latitude SST.
It should be realized that the definition of a coupled mode, a mode that
depends critically on the interaction of the atmosphere and ocean, is far
more difficult in mid-latitudes than our favorite example of ENSO. A great
deal of analytic work on ENSO has shown that coupled ENSO modes arise
solely from the interaction of the atmosphere and the ocean and would
not exist without this interaction (e.g. Hirst, 1986). In midlatitudes, we
have two possibilities that can masquerade as a coupled mode: the ocean
can have decadal variability which then, through its SST expression, drives
atmospheric motions on decadal scales, or, more likely, the atmosphere can