The best monthly rainfall forecasts come from computer programmes able to represent complex relationships between climate data while acquiring knowledge from many examples over time for better pattern detection, said researchers.
Dr John Abbot and Jennifer Marohasy from the Central Queensland University considered relationships between lagged values for temperature, atmospheric pressure and rainfall as well as climate data.
They compared results from their artificial neural networks (ANNs) analysis of the Inter-decadal Pacific Oscillation index against government-based seasonal rainfall-forecasting programmes.
"Forecasts using the ANN for sites in three geographically distinct regions within Queensland are shown to be superior compared to forecasts from the Predictive Ocean Atmosphere Model for Australia (POAMA), which is the general circulation model used to produce the official season rainfall forecasts," researchers said.
They said a major limitation of government forecasts is that they provide no information about the magnitude of the expected deviation from the median rainfall value within the defined forecast period.
Researchers said that for purposes including management of water infrastructure or scheduling mine operations, the distribution of rainfall within the three-month period is more important than an averaged seasonal value.
They said ANNs have been investigated for rainfall forecasting in many parts of the world. The study was published in the Atmospheric Research journal published by Elsevier.