A team of researchers at James Cook University, led by Dr Bithin Datta, have successfully used an AI-based tool to analyse existing meteorological and hydrological data related to the Ross River basin area, near Townsville in Queensland.
The results of the Artificial Neural Network’s (ANN) analysis found patterns that may correspond to an impending drought in the region. According to Dr Datta, ANNs are known for their exceptional pattern recognition capabilities which mimic the complex recognition and decision-making traits of the human brain.
“Once trained and validated, it is capable of establishing the relationship between apparently related variables and identifies less relevant variables. These are then used to make a prediction.”
This AI shows great promise for farmers, who may come to rely on advanced-AI predictions about the climate as extreme climate events worsen over time — something we’ve seen amply over the years in Australia. While the range of uses hasn’t been fully explored, the drought-predicting AI-based tool may also have applications in mining.
“The concept will remain the same if we select a catchment, or a region surrounding mines. I also understand that for mines, especially where dewatering is important, in some ways droughts may have a positive impact on groundwater levels.
“The preliminary results show our approach can also help predict groundwater levels, or even salinity in groundwater. That could be useful to mining as well. Water quality in both surface and groundwater systems is another complex issue. Our team — my Ph.D. students — are continuing to look into characterising sources of contamination and pathways of contaminant transport in mine sites, and it is another relevant issue.
“From a water supply point of view, any ability to forecast impending drought conditions with sufficient reliability will certainly help with immediate and longer-term planning for managing water demand and supply. Our efforts can certainly help in planning for water use” says Dr Datta.
Further training of this tool’s AI pattern recognition and decision-making processes will open possibilities for testing and use in several industries. Dr Datta said his research team planned to continue further development of the technology, with the intention of eventually rolling it out on a wider scale.
