Show description

In the extratropical region there is some agreement between the 500 mb height anomalies which are modelled and observed. The notable result is that both differ in a major way from the Pan and Oort composite pattern indicating the different response in both model and atmosphere to the very strong El Nino forcing of 1982/83. ACKNOWLEDGEMENTS Mike Lazare set up the modified sea-surface temperatures and ran the simulations. The results could not have been obtained without his help. Lynda Smith produced the manuscript and Brian Taylor aided with the figures.

E N S 0 events are shaded and S dcnotes super-events (sec text for definition). 39 the five 50-year periods, the model yields a probability of occurrence of super-ENSO’s of about one in every 30-40 years. While the oscillator output looks similar to observation, we should not, however, belabour with the specific justification as to what exact parameter(s) in the observation to identify with the variable X ( t ) . The choice of the SO1 here is a pure matter of convenience. Instead, we should look upon eqn.

Stable states. 5 on the X-axis. It follows that the atmospheric system there is dominated by a single equilibrium state. The mean value of 265 W m-' suggests that the prevailing condition is either low-cloudiness or clear-sky (a value of less than 240 W m-' in the tropics is generally associated with deep convection). Occasional outbreak of deep convection does occur over this region as indicated by the presence of points with negative values and large tendencies although their occurrence is too infrequent to show up in the PDF.