Microsoft open-sources a tool aimed at improving long-range weather forecasts

The Verge reports that Microsoft’s open-source Probabilistic Bias Correction tool uses machine learning to correct systematic bias in existing weather models. Researchers say it could improve forecasts two to six weeks ahead while running on CPUs, potentially making it more accessible to agencies with limited computing resources. In the reported study, it improved accuracy for 98 percent of precipitation targets and 92 percent of temperature targets in a leading physics-based model.