ENSO modelling at MPI

In this contribution the modelling activities of the climate research group at the Max—Planck—Institut fuer Meteorologie (MPI) concerning the El Nifio/Southern Oscillation (ENSO) phenomenon are briefly reviewed. The studies described encompass the investigation of the atmospheric response to observed sea surface temperature (SST) distributions, the oceanic response to observed wind stress and studies with a coupled ocean—atmosphere model investigating large scale air—sea interactions in the tropics. It is shown that the atmosphere model simulates realistically the Southern Oscillation during an extended range integration using near global SSTs for the period 1970 — 1985. In particular, low frequency changes of sea level pressure (SLP) as expressed by the Southern Oscillation Index (SOI) show a good correspondence with the observed changes. Furthermore, the spatial patterns of surface wind stress anomalies are simulated correctly, while the variance is somewhat underestimated by the atmosphere model. The equatorial oceanic general circulation model captures basic aspects of observed SST variability in the Equatorial Pacific, when forced with observed wind stress patterns. Associated sea level and current changes are consistent with the available measurements. In the next step the two models have been coupled together. Results are shown from three different experiments. In the first experiment the response of the coupled model to a westerly wind burst over the Western Pacific was investigated. It will be shown that the coupling of ocean and atmosphere is the most important factor for the persistence of SST anomalies in the Equatorial Pacific. The second experiment is an extended range integration with the coupled model over ten years. Low frequency variability is significantly underestimated in this run. The coupled model shows a climate drift consisting of a gradual cooling of the upper equatorial ocean. The coupled model was then used to study the possible role of anomalous snow cover over Eurasia for the initiation of ENSO events. By doubling the snow fall rate within the atmosphere model a weak El Nifio event was induced in the Tropical Pacific Ocean. Finally a simplified coupled ocean—atmosphere model was used for ENSO hindcast experiments. It will be shown that such a coupled system is quite successful in predicting the onset and the evolution of the 1982/1983 ENSO event