Report Time

May 24, 2024 (Friday) 14:00 - 16:30

Report Content

Report 1: "Hybrid Physics-Data Near-Ground Elements Diagnosis for Two Schemes" (Presenter: Researcher Feng Jin)

Accurate prediction of ground meteorological elements is of great importance to many sectors. The physical schemes in Numerical Weather Prediction (NWP) and data-driven correction methods are limited due to the uncertainty of parameterization and lack of robustness. The report will introduce two physics information-data models: one is based on the traditional Fortran framework of near-ground multi-layer physics, but optimizes the forcing parameters according to the observed data; the other is based on the near-ground physical parameterization simulator and error migration learning corrector. Both schemes have achieved better forecasting results than numerical models. The latter can provide convenient, fully gridded accurate forecast results, and improve the forecasting performance of extreme values, with zero-sample learning capabilities.

Report 2: "The Look of the Next Generation Weather and Climate Forecasting Model Based on AI" (Presenter: Xia Jiangjiang, Associate Researcher)

The report will introduce the recent progress in the application of artificial intelligence technology in weather and climate forecasting, and explore the possible development trends of the next-generation weather and climate forecasting model based on AI.

Guest Profile

Feng Jin,a researcher at the Institute of Urban Meteorology, CMA, Beijing, has been selected as a Young Meteorological Talent by the China Meteorological Administration and selected for the Outstanding Engineer Growth Plan by the Beijing Association for Science and Technology. Visiting scholar at the Mesoscale and Microscale Meteorology Laboratory of the National Center for Atmospheric Research. He has undertaken projects from the National Natural Science Foundation of China (NSFC) Youth and General Projects as well as numerous national key research and development programs. Feng Jin has published over 30 papers, and his development plan has been adopted by the National Center for Environmental Prediction's next-generation regional rapid refresh forecast system. His research interests include AI meteorology and big data application technology, numerical model forecasting technology, air quality assessment, and prediction technology, among others.

Xia Jiangjiangis an associate researcher at the Institute of Atmospheric Physics, Chinese Academy of Sciences, and deputy director of the Research Center for Artificial Intelligence in Atmospheric Sciences. His research combines meteorological big data and numerical weather prediction models, leveraging machine learning and deep learning techniques, primarily focusing on three aspects: (1) developing post-processing methods for models, and reconstructing weather and climate information from high spatiotemporal resolution data sets; (2) developing weather and climate forecasting models, including short-term extrapolation forecasting models, medium- and short-term weather forecasting models, and seasonal-scale extreme weather and climate event forecasting models; and (3) applying deep learning interpretability techniques to the study of meteorological and climatological physical processes.