Natural hazard prediction

Watchdog Says Machine Learning Could Transform Weather Forecasting

The Government Accountability Office has suggested using artificial intelligence to enhance weather forecasting and mitigate billions of dollars in damage annually.

In a paper published on Dec. 14, the watchdog explained that machine learning-powered algorithms could automate processes and rapidly extract insights from large datasets to give authorities more time to prepare for extreme weather events, FedScoop reported Friday.

The study involved assessments of ML models, interviews with different stakeholders and reviews of existing reports and studies.

GAO found that routine forecasting operations already use some ML models to improve storm warning times. Other potential uses of ML are either close to operational or require more years of testing.

According to GAO, ML could allow existing data to be maximized and reduce prediction uncertainty.

Data limitations could hinder ML model training and reduce accuracy in some areas, the watchdog noted. GAO also flagged bias concerns, distrust of the technology and development and operational costs as obstacles to adoption.

To remedy such challenges, the watchdog recommended improvements to data management, more education and training opportunities and working to overcome personnel and resource shortage issues.