The Occurrence and Predictability of Extreme Events over the Southwest Pacific with Particular Reference to ENSO

Abstract

Extreme events in the Southwest Pacific (SWP) are widespread and largely centre on tropical cyclones, droughts and floods. These events are directly influenced by the state of the El Niño Southern Oscillation (ENSO). The impacts of climate extremes are dramatic, particularly on Small Island Developing States (SIDS) of the SWP region. They cause destruction and widespread damage to human, socio, economic and development well being of local communities, examples being tropical cyclones Ofa, Val and more recently Heta, Olaf and Percy that devastated Samoa, Niue, Tonga and the Cook Islands and the severe impacts of the 1997–1998 El Niño event on Papua New Guinea and Fiji. Although warming in the SWP region is decreasing, extremes of low temperature, the trend of more El Niño episodes in the 20th century, based on observations, has increased both dry periods and extreme rainfalls in significant parts of the region, especially near the Date Line. The phase of ENSO also significantly affects temperature and rainfall anomalies and extreme across the region, with sharp gradients between contrasting areas of above and below average departures. ENSO also changes tropical cyclone track, density and occurrence, with El Niño events increasing the frequency and displacing tropical cyclones more towards the central South Pacific, and La Niña episodes decreasing the frequency and confining tracks to the west of the Date Line.

Seasonal to Inter-annual (SI) prediction has become a major research area and application tool that offers much promise for early warning and disaster preparedness to extreme events. In the SIDS of the SWP region, the Island Climate Update (ICU), a multinational virtual type regional climate prediction tool, with success hit rates (percentage of prediction being correct after comparing it with what actually happened) greater than 70 percent being achieved in some parts of the region since it was launched in September 1999, provides an extremely useful regional mechanism for preparing and predicting seasonal rainfall departures and other extreme events associated with ENSO for SIDS. The ICU could be used as a model in other regions.