In a significant move towards technological independence, China is accelerating the development of a domestic weather dataset aimed at replacing Europe”s ERA5. This initiative is part of China”s broader strategy to establish itself as a leader in meteorology and climate science while reducing reliance on foreign data sources.
The ERA5 dataset, produced by the European Centre for Medium-Range Weather Forecasts, is renowned for its comprehensive climate reanalysis, which integrates millions of global observations to create a reliable historical account of weather patterns over the past 80 years. It has become the cornerstone for AI-driven weather forecasting, with various advanced models, including those from Chinese firms such as Huawei, utilizing its data.
However, concerns over national security and the potential vulnerabilities of relying on external data have prompted Chinese policymakers to rethink their approach. “Weather forecasting is national security,” stated Andreas Prein, a professor at ETH Zurich and expert in weather modeling, highlighting the strategic importance of self-sufficiency in meteorological data.
In response to these concerns, China has introduced the CMA-RA V1, a new global dataset launched by the China Meteorological Administration. This dataset is designed to support AI applications and is already being used to train local weather forecasting models. The higher spatial and temporal resolution of this new dataset offers advantages over ERA5, allowing for enhanced regional prediction capabilities.
As global researchers begin to explore the utility of China”s reanalysis data, there is a growing interest in its potential applications. Hui Su, a professor at the Hong Kong University of Science and Technology, is applying this data within her startup, Stellerus, to enhance AI model training and traditional forecasting methods.
Furthermore, the availability of Chinese weather data could stimulate the development of weather risk hedging markets, which are currently more established in regions like the US and Europe. David Whitehead, head of weather risk management at Vaisala, indicated that greater access to Chinese data would likely attract interest from international companies looking to leverage this resource for financial products.
China”s meteorological authority claims its reanalysis offers improved accuracy for wind speeds at 100 meters above ground level, a crucial metric for the country”s extensive wind power industry. Nonetheless, experts suggest that while the Chinese dataset presents promising features, it may not immediately displace the ERA5 as the global standard in the near term.
