Forecasting of monthly rainfall in the Murray Darling Basin, Australia: Miles as a case study

Abstract

The Murray Darling Basin accounts for nearly 40% of the value of agricultural production in Australia, and 65% of the irrigated land. We use an artificial neural network (ANN), a form of machine learning, to show the potential for more reliable monthly rainfall forecasts with a lead time of 3 months, and the potential skill of the same model for 6, 9, 12 and 18 month lead-times for the township of Miles, in the northern Basin. The skill of these forecasts is contrasted with the skill of the Predictive Ocean Atmosphere Model for Australia (POAMA), a general circulation model used for operational forecasts by the Australian Bureau of Meteorology. Forecasts from the ANN are significantly more skilful for all lead times. The ANN's capacity to integrate information from the climate indices Niño 1.2, 3, 3.4 and 4, the Dipole Mode Index (DMI), and also a composite local maximum and minimum atmospheric temperature series, contributes on average approximately 60% of the skill of the ANN forecast.

Forecasting of Medium-term Rainfall Using Artificial Neural Networks: Case Studies from Eastern Australia

“…This chapter is a review of various studies undertaken since 2012 focused on this general area, with specific information on data and methodology in the published technical papers that are referenced [11][12][13][14][15][16][17][18][19][20][21]. However, in this first section, the method used in an early study [17] is provided in more detail, by way of background into how an ANN can be practically deployed to generate a rainfall forecast.…”

“…In contrast, medium-term rainfall forecasts issued to the public by the Australian Bureau of Meteorology are in the form of probabilities relative to the median seasonal rainfall, and do not differentiate between an anticipated rainfall slightly above the median and an extreme rainfall event, such as occurred in Queensland during the period December 2010 and January 2011 [14]. …”

Forecasting extreme monthly rainfall events in regions of Queensland, Australia using artificial neural networks

“…Two approaches have previously been investigated for neural network optimization of monthly rainfall forecasting in Queensland [14]. With the first approach, designated as "allmonth optimization", data for all 12 months of the year were included as input and optimised together, as in our previous studies [8][9][10]. With the second approach, designated as "single month optimisation", forecasts corresponding to each calendar month were performed individually [14].…”

“…Artificial neural networks (ANNs), a form of machine learning, provide an alternative technique for medium-term rainfall forecasting both in Australia [8][9][10], and in other regions of the world [11][12][13]. Our previously reported investigations applied neural networks to individual sites within Queensland with long rainfall records to generate forecasts of monthly rainfall with lead times up to 12 months [8][9][10].…”

“…Our previously reported investigations applied neural networks to individual sites within Queensland with long rainfall records to generate forecasts of monthly rainfall with lead times up to 12 months [8][9][10]. A recent study [14] focussing on the sites of Harrisville and Gatton in south-east Queensland clearly demonstrated that superior monthly rainfall forecasts can be generated using the ANN approach when compared to results generated by POAMA.…”