Australasian Joint Conference on Artificial Intelligence

AI 2016: AI 2016: Advances in Artificial Intelligence pp 71-87

Forecasting Monthly Rainfall in the Western Australian Wheat-Belt up to 18-Months in Advance Using Artificial Neural Networks

Conference paper

DOI: 10.1007/978-3-319-50127-7_6

Volume 9992 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Abbot J., Marohasy J. (2016) Forecasting Monthly Rainfall in the Western Australian Wheat-Belt up to 18-Months in Advance Using Artificial Neural Networks. In: Kang B., Bai Q. (eds) AI 2016: Advances in Artificial Intelligence. AI 2016. Lecture Notes in Computer Science, vol 9992. Springer, Cham

Abstract

Accurate medium-term rainfall forecasts are a significant constraint to dry land cropping. In Australia, official monthly forecasts for the Western Australian wheat-belt are currently based on output from the Bureau of Meteorology’s general circulation model, the Predictive Ocean Atmosphere Model for Australia (POAMA). These forecasts are provided in a two-category format (above or below median rainfall) up to three months in advance for large grid areas, and are not considered reliable. An alternative approach is presented here for the three locations of Narrogin, Merredin and Southern Cross using artificial neural networks (ANNs) to forecast monthly rainfall up to 18 months in advance. Skilful monthly rainfall forecasts can be achieved at all lead times measured in terms of root mean square error (RMSE) and mean absolute error (MAE). This approach is of practical benefit to wheat growers in this region, with potential application to other locations with long historical temperature and rainfall records.

Keywords

Artificial neural network Machine learning Monthly rainfall forecast Dry land cropping Climate indices 

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  1. 1.Climate Modelling LaboratoryNoosavilleAustralia
  2. 2.Department of EngineeringUniversity of TasmaniaHobartAustralia
  3. 3.Institute of Public AffairsMelbourneAustralia