Chinese researchers turn to AI for accurate weather forecasts

Chinese researchers from the Shanghai Academy of Artificial Intelligence for Science (SAIS) at Fudan University have announced the creation of a new artificial intelligence (AI) model to predict weather conditions.

The AI model is tailor-made for sub-seasonal forecasts, demonstrating high accuracy for predictions within a two-week window. The model, dubbed FuXI-Subseasonal, can make weather forecasts for longer periods, with lead researcher Qi Yuan noting proficiency in 30-day predictions.

Qi disclosed that the motivations for proceeding with a generative AI model for weather forecasts revolve around the need to spot climate disaster warnings early enough. Using the model, the researchers say Chinese authorities will be able to identify climate crises ahead of time and take necessary steps to mitigate the damage.

“Here, we introduce FuXi Subseasonal-to-Seasonal (FuXi-S2S), a machine learning based subseasonal forecasting model that provides global daily mean forecasts for up to 42 days, covering 5 upper-air atmospheric variables at 13 pressure levels and 11 surface variables,” read a paper.

When compared to traditional numerical weather predictions relying on supercomputers, FuXI-Subseasonal stands apart on several metrics and is capable of generating accurate predictions on limited data.

“Climate disaster warning is another important value of this FuXi-Subseasonal model,” said Qi in an interview with Global Times.

Developed in collaboration with China’s National Climate Center, the machine learning model is significantly faster than traditional weather prediction models, with the report pegging operational speed at a “thousand-fold increase.” A longer forecasting period and increased accuracy levels could see Chinese authorities lean on local AI models for climate disaster prevention rather than North American and European offerings.

Already, locally designed AI models have shown promise in predicting the patterns of climate disasters in Mainland China. In mid-2023, Chinese authorities confirmed that an early-stage AI model developed by SAIS accurately predicted the movement and intensity of typhoon Doksurik, providing sufficient data for aid and relief efforts.

The rise of AI-based weather prediction systems

While generative AI finds new use cases across several sectors, utility continues to rise in meteorology, with several firms rolling out their offerings and merging them with other emerging technologies.

NASA and IBM (NASDAQ: IBM) launched a foundational model specifically for severe weather patterns while Johns Hopkins teamed up with the National Oceanic and Atmospheric Administration (NOAA) to unveil their intelligent weather forecasting model.

Google (NASDAQ: GOOGL) has commercially released multiple weather prediction systems since mid-2023, but among other nations, India appears to be the biggest adopter of the technology as it attempts to mitigate a slew of environmental disasters.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: Transformative AI applications are coming