Researchers from CSIRO and the University of Melbourne found that the Pacific Ocean’s initial climate conditions significantly influence the accuracy of long-range climate forecasts.

The findings could help communities prepare earlier for droughts, floods and other extreme weather events.

Researchers analysed more than 7,500 historical climate forecasts from models developed in nine countries. Predictions beginning during La Niña conditions remained reliable up to a year ahead, while those starting from neutral conditions performed less accurately.

Scientists also identified potential for machine learning and statistical models to improve forecasting under less predictable conditions. Further research is needed before the approach can be introduced into operational forecasting systems.

Set The Portugal News as your preferred source for news on Google

GoogleAdd as a preferred source on Google