As the Earth warms and weather events become more extreme, scientists are continuously improving and developing new tools for weather forecasting. According to the World Health Organization, an estimated 55 million people worldwide are affected by drought each year, with that number expected to grow as the climate continues to change.
A fourth-year student at the University of Waterloo in Canada is using artificial intelligence to enhance drought predictions. The student, Andrew Watford, says the goal was to bring together mathematics and machine learning to develop new methodologies and push the field forward to predict drought. While they’re not there yet, Watford and two of his supervising professors hope to create a tool to predict drought up to five years in advance.
Watford’s role in the project involved writing code to predict the normalized difference vegetation index in drought-prone regions in Kenya. The vegetation health data is then used with AI to forecast drought patterns.
The recently published paper examined the effectiveness of three different models for predicting droughts in Kenya. Two of the models they examined involved machine learning. They found that the models performed best in a particular farming zone since rain in the area followed a more significant and consistent yearly pattern.
Although the research is still in early stages, they hope it serves as a foundation for improving extreme weather prediction tools in the future. The ability to predict droughts earlier could have immense benefits such as enabling local governments to implement water management strategies, allowing farmers to select drought-resistant crops, and significantly enhancing natural disaster preparedness that could save lives.
