Development of prediction model for forecasting rainfall in Western Australia using lagged climate indices

The aim of the study was to develop a model to forecast autumn rainfall several months in advance for south-west division (SWD) of Western Australia (WA), by identifying and incorporating the relationship among major climate indices such as dipole mode index (DMI), southern oscillation index (SOI), ENSO Modoki index (EMI) and autumn rainfall. Eight rainfall stations from two regions of SWD were considered. Statistical analysis showed that DMI, SOI, Nino3.4, Nino3 and Nino4 have significant correlations with autumn rainfall for all these stations. On the other hand, EMI showed significant correlations for the stations in the north-coast region only. Meanwhile, DMI effect has been found stronger for all the stations compared to other climate indices. Several multiple regression analyses were conducted using lagged ENSO-DMI, lagged SOI-DMI and lagged EMI-DMI indices, and significant increase in the correlations between autumn rainfall and climate indices was observed. However, only statistically significant models were suggested.