Extreme rainfall has devastating effects on human health and our economy. Climate change is influencing all extreme rain events. The 1°C temperature rise that has already occurred has led to an increase of about 7% in the amount of water vapour in the atmosphere, which in turn leads to more intense rainfall.

There are known problems with Australian extreme rain forecasts, particularly in the tropics. We need to understand the processes by which extreme rain events occur in order to improve our forecast models, and also understand how these events may change in a changing climate in order to prepare society for the future.

 In this work we use satellite observations of rain over Australia to identify extreme rainfall days in the tropics, sub-tropics and midlatitudes. Using global reanalysis data, we investigate the relationship of the satellite-identified extreme rain days to the large-scale dynamical features of the atmosphere. By using a clustering approach, we classify the extreme rain days into synoptic types based on the near-surface circulation associated with each day. Each synoptic type reveals information about the lifting and moisture transport associated with the extreme rainfall event. This provides insights into possible deficits in current forecast models and key areas for model improvement.