84th AMS Annual Meeting

Thursday, 15 January 2004
The global nonlinear impact of SST changes in the central tropical Pacific
Hall 4AB
Prashant D. Sardeshmukh, NOAA/CIRES/CDC, Boulder, CO; and G. P. Compo
An atmospheric general circulation model (AGCM) experiment with anomalous SST forcing only in the Nino-4 area of the central tropical Pacific captures all the major elements of the global response obtained in multi-decadal AGCM integrations with prescribed observed global SST variations. This area lies at the western edge of the region of warmest SST anomalies during El Nino, but is dominant in generating the global anomalies associated with El Nino. In particular, the Nino-4 experiment suggests that rainfall anomalies over the west Pacific, Africa, central America and Amazonia observed during El Nino may be largely forced by remote central tropical Pacific rather than by local SST anomalies. The nonlinearity of the global response to SST forcing in this important area is then investigated by comparing large ensembles of AGCM integrations with prescribed positive and negative 1K, 3K, and 5K SST anomalies in Nino-4. To a first approximation, the response remains similar in pattern but varies nonlinearly in amplitude as the forcing is varied across this range. Plotted as a function of the Nino-4 SST, the precipitation response is J-shaped: it saturates at strong negative forcing but is remarkably linear from weak negative to strong positive forcing. The circulation response, on the other hand, is S-shaped: it saturates at both strong negative and strong positive forcing, but at different amplitudes. The negative saturation is closely but not excusively linked to the negative saturation of the precipitation response. The positive saturation cannot be so linked; it has a purely dynamical origin. This asymmetry of the S-response saturation levels is evident in most dynamical variables examined here, including geopotential heights in regions of the North Atlantic/Arctic oscillations.

These results have important implications for the improved modeling of the mean tropical climate, tropical variability, and tropical-extratropical interactions from monthly to centennial scales. They also have a bearing on assessments of global climate change associated with anthropogenic forcing. The asymmetric saturation of the circulation response for positive and negative Nino-4 forcing is especially relevant in this regard. Because of it, one may expect the mean global climate to be affected not just by a mean change in Nino-4 SST, but also by a change in its variability associated with a change in El Nino variability.

Supplementary URL: