Development and Prospects of Data Assimilation in Numerical Weather Prediction

For numerical weather prediction (NWP), data assimilation combines short-term forecasts and various atmospheric observations to achieve optimal initial conditions used for subsequent weather forecasts. With the rapid advancements in numerical models and observing systems, data assimilation has evolved significantly. Modern methods now account for uncertainties across different spatial and temporal scales, incorporate diverse observation error statistics, and enforce dynamical constraints and model balances. Meanwhile, observations from various platforms, such as ground-based, aircraft, and satellite, have been assimilated. These include data from polar-orbiting and geostationary satellites, radar-derived radial winds and reflectivity, Global Navigation Satellite System (GNSS) radio occultations, and more. To further utilizing the improved observing systems and data assimilation for high-impact weather predictions, target observation strategies have been developed to identify regions where additional observations yield the greatest prediction improvements. Based on the advancements of data assimilation, China’s operational systems have also made significant progress, establishing advanced operational data assimilation systems. Over the past decade, the forecast skill of 5-day global weather prediction has improved by approximately 15%. The article reviews a century of development in data assimilation, and discusses future directions, including advanced methods, operational system frameworks, integration of novel observations, and the synergy between data assimilation and artificial intelligence.