Trends in rainfall and temperature extremes in the Pacific — ASN Events

Trends in rainfall and temperature extremes in the Pacific (7817)

Agata A Imielska 1 , Kirien Whan 2 , Simon McGree 3 , David Jones 3 , Lisa V Alexander 2
  1. Bureau of Meteorology, Surry Hills, NSW, Australia
  2. Climate Change Research Centre, UNSW, Sydney, NSW, Australia
  3. NCC, Bureau of Meteorology, Melbourne, Vic, Australia

Many people living in the Pacific Islands Countries and East Timor report their climate is changing. Despite the high vulnerability that these countries face, there is only very limited scientific information available to these countries about the nature and significance of climate trends. A recent report published by the Pacific Climate Change Science Program (PCCSP) found that collectively, the Pacific region indicates a climate in transition, driven by both natural and human influences, with large nearly monotonic warming, sea level rise and shift in rainfall patters.

Extreme climatic events can have large impacts, particularly on developing nations that have less under-developed infrastructure and lower capacity to react to high impact events.  As part of the Pacific-Australia Climate Change Science and Adaptation Program (PACCSAP) analysis of new homogenised datasets (including daily and monthly rainfall and temperature) has been carried out for countries in the Pacific to further explore climate variability in the western Pacific with a focus on climate extremes. Traditional analysis of trends in climate extremes utilise linear regression techniques and use percentile based definitions of extremes over large areas. This however introduces potential statistical biases in the sampling of data and potentially incorrect assumptions about its distribution. An alternative lies in the use of Extreme Value Analysis (EVA) techniques to define and analyse trends in climate extremes. This research focused on the use of EVA to examine western Pacific trends in climate extremes with results showing an increase in minimum and maximum temperature extremes with rainfall extremes increasing for higher order average recurrence intervals.