Change point analysis of Southwest Western Australian winter rainfall and potential drivers — Australian Meteorological and Oceanographic Society

Change point analysis of Southwest Western Australian winter rainfall and potential drivers (#92)

Matthew Jones 1 , George Takacs 1
  1. University of Wollongong, Wollongong, NSW, Australia

Winter rainfall in southwest Western Australia (SWWA) has seen a notable decline since the 1970s. Multiple reasons for this change have been discussed such as changes in sea level pressure, a positive trend in the Southern Annular Mode (SAM) and a weaker African monsoon. The most significant contributor is thought to be land cover change, which was modeled by Pitman (2004) and shown empirically by Andrich and Imberger (2013). Change point analysis has been applied to monthly and seasonal rainfall data for stations in SWWA in order to identify if and when a significant change in the mean and variance of the data exists. Change points in winter rainfall are found in the mid 1960s for a number of stations, around the same time as the change found by Andrich and Imberger (2013). Applying the same methods to daily rainfall quantiles in winter reveals that the behavior of the 0.95 quantile varies across stations, with some indicating a change at the same time as the total rainfall and others indicating no change at all. The Indian Ocean Dipole (IOD) and SAM, two large scale climate drivers which impact the region, have also been investigated through change point analysis. While no significant changes were found for SAM, change points in the IOD during some winter months are found to have occurred in the 1960s – aligning with the change points noted in the winter rainfall. Just how closely this change in IOD is linked to rainfall in SWWA is questionable however, given IOD tends to have a larger effect in northern parts of the state. It is clear from these results that the rainfall decline in SWWA occurred quite suddenly. To fully understand the reasons behind this change, change point analysis will be applied to data for other potential causes.

#amos2020