- Chinese scientists have developed the “FuXi-Subseasonal” AI model for improved sub-seasonal weather forecasts.
- The model offers up to a thousand times faster operational speed and superior accuracy compared to existing international models.
- The forecasting period has been extended from 30 to 36 days, allowing earlier warnings for extreme weather.
- The model is part of China’s expanding efforts in AI-driven weather prediction technologies.
- The Fengwu model from Shanghai Artificial Intelligence Laboratory previously outperformed European and American models in predicting Typhoon Doksuri.
- Fengwu uses AI to analyze atmospheric data, providing high-precision forecasts with minimal computational resources.
Main AI News:
Chinese scientists have unveiled an advanced artificial intelligence (AI) model designed to enhance sub-seasonal weather forecasting, offering a promising solution for mitigating climate disaster costs, as reported by state media on Thursday. The new model, named “FuXi-Subseasonal,” has been developed collaboratively by the Shanghai Academy of Artificial Intelligence for Science (SAIS), Fudan University, and China’s National Climate Center. According to Global Times, this model boasts a remarkable increase in operational speed—up to a thousand times faster—and surpasses existing international models in both forecasting accuracy and extended forecasting periods.
Research team leader Qi Yuan highlighted the model’s significant contribution to climate disaster warnings. The FuXi-Subseasonal model extends the prediction window for extreme weather events from 30 days to 36 days, providing crucial additional time for response and mitigation strategies. This model exemplifies China’s growing capability in AI-driven extreme weather prediction.
Previously, in July of the previous year, the Fengwu model from the Shanghai Artificial Intelligence Laboratory outperformed its European and American counterparts in forecasting the trajectory of Typhoon Doksuri. Fengwu, which emphasizes forecasting precision, relies on atmospheric reanalysis data to produce accurate predictions. As explained by scientist Bai Lei, the model utilizes AI to analyze meteorological elements like wind speed, temperature, and humidity, achieving high-precision forecasts with minimal computational resources. Unlike traditional physical models that require supercomputers, Fengwu can deliver global weather forecasts for the next 10 days in just 30 seconds using a single graphics processing unit.
Conclusion:
The introduction of the FuXi-Subseasonal model represents a significant advancement in AI-driven weather forecasting, showcasing China’s growing prowess in this field. The model’s increased speed and extended forecasting period not only enhance accuracy but also offer valuable additional time for disaster response and mitigation. This development could reshape the global market by setting new benchmarks for weather prediction and potentially influencing international standards and practices in climate disaster management.
