On April 9, 2021, the national key R&D plan "Global Change and Response" key special project - "Big Data and deep learning method Innovation Earth System Model development and Application Research" project kick-off meeting and "Development of Common Earth System Model Efficient parallel Computing Framework" project kick-off meeting was held in Wuxi. This conference is hosted by Tsinghua University and Nanjing University of Information Science and Technology, and organized by Wuxi Jiufang Technology Co., LTD.

The conference was divided into two launch sessions, attended by more than 50 industry experts.

The project of "Developing a General Efficient Parallel Computing Framework for Earth System Models" was led by Tsinghua University with the participation of National Meteorological Center, the First Institute of Oceanography, Wuhan University and East China Normal University. The project of "Big Data and Deep Learning Method Innovation of Earth System Model Development and Application Research" was led by Nanjing University of Information Science and Technology, and Tsinghua University, Tongji University and Chengdu University of Information Science and Technology participated in the meeting. The chief scientists and project leaders of each project made detailed reports to the expert group on the project research objectives and contents, project organization and implementation plan and technical route respectively. And listened to the expert's professional advice and guidance, finally, the review group of experts unanimously agreed to pass the implementation plan, and suggested the project team to further optimize the implementation process, strengthen the organization and management of follow-up work.

Huang Xiaomeng, the project leader of "Developing a universal Efficient parallel computing Framework for Earth System Models", said that through the research of the universal highly extensible parallel computing framework, efficient migration and optimization of models and other key science and technology, significantly improve the efficiency of model development, migration and optimization, and promote the innovation of earth system models. To promote the scientific and technological progress in the field of earth system model and simulation forecast, at the same time, improve the accuracy and real-time performance of model forecast, effectively improve the ability of meteorological disaster prevention and reduction, reduce disaster losses, and provide security for national economic production activities.

Luo Jingjia, the project leader of "Big Data and Deep Learning Method Innovation Earth System Model Development and Application Research", said that by integrating big data of global change and artificial intelligence technology, this project will develop a more accurate season-interannual forecasting system combining dynamic and artificial intelligence, and improve the forecasting skills of important climate phenomena, which will greatly improve the level of climate forecasting in China. We will comprehensively reduce the impact of complex climate phenomena such as ENSO, typhoons and the East Asian monsoon on human beings.