Doksuri is the strongest typhoon to hit China so far. “From July 21 to 27, Fengwu’s forecast of Doksuri’s path was off by an average of 38.7 km, while the corresponding figure for the European Center for Medium-Range Weather Forecasts was 54.1 km and the US National Weather Service was 55 km,” the Shanghai AI Lab told Yicai Global.
Doksuri, China's fifth typhoon this year, made landfall on July 28. More than 720,000 people in Fujian Province were affected, and direct economic losses totaled 52.3 million yuan ($7.3 million), according to figures released by the Fujian provincial flood control authority.
The Fengwu model was released by the Shanghai Artificial Intelligence Laboratory and the University of Science and Technology of China in April 2023.
Reducing the error by one kilometer in 24 hours can reduce direct economic losses by about 97 million yuan ($13 million), so accurate storm forecasting is crucial to reducing risks, said the Shanghai AI Lab.
In addition, Chinese researchers have developed an artificial intelligence (AI) model based on deep learning algorithms to forecast the development and morphology of El Nino phenomena in the central Pacific region.
In a recent study published in the journal Advances in Atmospheric, scientists said that the El Nino phenomenon in the central Pacific Ocean can have a profound impact on the global climate, so accurate forecasts will be important in preparation and risk reduction.
Based on convolutional neural network technology, researchers from the Institute of Atmospheric Physics (IAP) of the Chinese Academy of Sciences have developed a deep learning model to forecast sea surface temperature anomalies in the equatorial Pacific Ocean.
“This study demonstrates the potential of AI to improve forecasting of important weather events such as El Nino, which can have serious global impacts,” said Huang Ping, a scientist at IAP and author of the study.
According to the study, the AI model outperformed traditional dynamical models in terms of accuracy, especially in forecasting sea surface temperature anomalies in the western and central equatorial Pacific.
The study also shows that the hybrid model combining forecasts from both AI and dynamical models achieves even higher accuracy for El Nino events in the central and eastern Pacific.
Duc Dung (according to Yicai Global, rossaprimavera.ru)
Source
Comment (0)