Response of a simple ACGM's annular mode patterns to thermal forcings

We examine the responses of a simple atmospheric general circulation model to thermal forcings to investigate whether annular mode patterns arise as a result of the forcing.

The setup of the model is the same as that used by Ring and Plumb (2007); it includes Newtonian relaxation toward a reference temperature profile. For each trial we alter the reference temperature profile, compile a climatology of the model under the new temperature profile, and compare to a control run of the model. Model runs which used hemisphere-scale alterations to reference temperature did not consistently produce the annular mode patterns. However, the patterns were found for trials where the thermal forcing was concentrated poleward of 45 degrees.

Through use of a zonally symmetric model, it is determined that the direct effects of the temperature perturbations do not produce the annular mode patterns. Instead an eddy feedback is necessary to capture properly both the shape and the strength of the patterns.

We further investigate these runs in tandem with the model runs using momentum forcing reported in Ring and Plumb (2007). For each trial, an effective torque is considered which includes the instantaneous response of the overturning terms, as well as the directly applied torque (if any). We then consider the projection of this effective forcing on the annular modes, as represented by the leading EOF of zonal-mean zonal wind in each hemisphere, and compare to the projection of the zonal wind response in each trial on the annular modes. We find a linear relationship between the two projections.