China has launched an upgraded version of its marine numerical forecasting model, MaCOM 2.0, featuring higher forecasting resolution, broader application scenarios and enhanced technological self-reliance.
The new model was unveiled on Sunday at the 11th Global Mazu Culture Forum in Putian, Fujian province. Developed by the Ministry of Natural Resources' National Marine Environmental Forecasting Center, MaCOM 2.0 is designed to support deep-sea and polar exploration, as well as maritime safety.
The upgrade includes significant improvements in global variable-resolution numerical simulation, an updated physical dynamics framework, expanded applications in sea ice and marine ecology, as well as enhanced large-scale parallel computing capabilities.
The model's name is inspired by Mazu, a Chinese goddess believed to protect people at sea. Mazu culture is an important part of China's maritime heritage and an iconic symbol of its maritime civilization.
Marine numerical models provide essential support for ocean development, ecological conservation, scientific research and forecasting, and are of strategic importance comparable to that of computer chips. However, China's marine forecasting and disaster prevention efforts have long relied heavily on models developed by the United States and the European Union. The lack of independently developed core models has become a major bottleneck in advancing the country's ocean forecasting capabilities.
Zhang Yu, an associate research fellow at the National Marine Environmental Forecasting Center, said the ministry has intensified efforts to develop the MaCOM series since 2018 to address this gap. Two models were introduced in 2021 and 2023 and have since been put into operational use.
"Compared with similar international models, MaCOM features an advanced physical dynamics framework and is the world's first marine forecasting model to support GPU-accelerated computing," Zhang said.
Since it was made an opensource model in 2024, MaCOM has been deployed by marine forecasting agencies in coastal provincial-level regions across China. It has also supported technology transfers and provided technical training to meteorological departments in countries including Indonesia and Bangladesh.
Building on its two predecessors, MaCOM 2.0 is designed to address strategic priorities such as strengthening China's maritime capabilities, safeguarding maritime passages and preventing marine disasters. The upgraded system aims to take the country's marine numerical forecasting capabilities beyond independent development and control toward providing stronger strategic support, greater accuracy and mobile deployment.
A key feature is its variable-resolution ocean current model, which is designed to provide high-resolution environmental information and security support around strategically important islands, reefs and major sea lanes. It also delivers more accurate, higher-resolution global ocean forecasts to meet growing demand for improved prediction capabilities worldwide.
In addition, GPU-based, or graphics processing unit, lightweight technology enables the development of a mobile global ocean forecasting system, allowing oceangoing vessels to conduct localized forecasts on board.
Tests and evaluations have demonstrated significant improvements in the model's ability to capture small — and meso-scale ocean processes. Its forecasts of sea temperature, salinity and current velocity have reached the level of advanced international models.
Its predictions of significant wave height are also comparable in accuracy to those produced by global wave models operated by the European Centre for Medium-Range Weather Forecasts and the US' National Oceanic and Atmospheric Administration.
Leveraging MaCOM's lightweight design and low energy consumption, China's polar research icebreaker Xuelong 2 has tested a global ocean current and wave forecasting system based on the model. The system provides mobile, high-resolution forecasts at sea, tailored to the specific objectives of scientific expeditions in the Arctic and Antarctic.
The forecasting center said that during the 15th Five-Year Plan period (2026-30), it will explore the integration of MaCOM with artificial intelligence. Priorities will include expanding the model's physical framework and application scenarios, strengthening model verification and evaluation, and improving the assimilation and analysis of data from China's ocean-monitoring satellites. These efforts aim to establish a comprehensive, fully self-reliant operational capability for ocean forecasting.
Lin Xinyun in Putian, Fujian, contributed to this story.
Contact the writers at limenghan@chinadaily.com.cn
