In the newly released report "United in Science 2024" released by WMO, there is an 86% chance that at least one of the next five years will exceed 2023 as the hottest year on record. At the same time, the report pointed out that in the next five years, there is an 80% chance that the global average near-surface temperature will temporarily exceed 1.5℃ above pre-industrial levels for at least one year in the next five years.
In order to cope with the climate crisis, it is particularly critical to create a fast, economical and accurate weather warning system. Artificial intelligence (AI) is bringing disruptive changes to the field of weather warning with its complex algorithms and powerful computing capabilities. As Saulo points out, “artificial intelligence has revolutionized the science of weather forecasting by making it ‘faster, cheaper and more accessible’”.
Image by United in Science 2024
This report points out that in the field of weather prediction, AI models have broken through the numerical weather prediction (NWP) model based on physical models, and have surpassed physical models in predicting certain weather variables and extreme or dangerous events (such as tropical cyclones). Research by scholars such as Keisler and Pathak has demonstrated the significant advantages of AI models in cyclone prediction.
By using the Fourier Forecast Neural Network, an emerging global data-driven weather forecast AI model, it has completely realized the accurate prediction of high-resolution and fast time-scale variables such as surface wind speed, precipitation and atmospheric water vapor. A week's forecast can be generated in 2 seconds, which is several orders of magnitude faster than IFS.
These capabilities were previously limited to large global forecast centers by computational burdens, but are now available to institutions without sufficient resources. The barriers to entry for running high-level forecast models have been significantly lowered, and lower costs enable scale. Smaller public and private entities can use AI to enter the field of weather forecasting, greatly changing the traditional pattern of the weather forecasting industry.
In recent years, with the rapid development of AI technology, many technology giants and research institutions such as Google, NVIDIA, and Huawei have made major breakthroughs in the field of weather forecasting and developed a series of eye-catching AI weather forecasting products. These products not only improve the accuracy and speed of weather forecasting, but also show unprecedented potential in critical areas such as extreme weather forecasting.
Among them, the Pangu-Weather model developed by Huawei Cloud has become one of the innovative achievements that have attracted global attention. The model was published in Nature magazine in July 2023. It was trained using 39 years of global reanalysis weather data. Its prediction accuracy is comparable to the world's top numerical weather forecast system IFS (Numerical Forecast System of the European Center for Medium-Range Weather Forecasts). However, the prediction speed is more than 10,000 times faster than IFS at the same spatial resolution. This breakthrough demonstrates the huge advantages of AI forecasting models in terms of efficiency and cost.
In November of the same year, Google DeepMind launched another breakthrough AI weather forecast model-GraphCast. The model can predict hundreds of weather variables for the next 10 days in one minute at a global 0.25° resolution. Compared with traditional weather forecasting methods, GraphCast not only significantly improves forecasting efficiency, but also performs well in forecasting extreme weather events.
Microsoft’s AI for Earth project is also leveraging machine learning and big data to improve climate predictions and weather forecasts. Although it did not directly develop a specialized AI weather forecast model, the related research supported by the project has played an important role in improving the accuracy of weather forecasts and helping the world better cope with the challenges posed by climate change.
Overall, these cutting-edge AI weather forecast models have greatly improved the speed, accuracy and coverage of weather forecasts through the training of large-scale historical data and the application of advanced deep learning algorithms. AI technology is revolutionizing traditional weather forecasting in a more accurate and faster way, playing an increasingly important and critical role in addressing the global challenges of extreme weather events and climate change.