Development and future prospects of the Regional Integrated Earth System Model (RIEMS)

East Asia faces increasing environmental risks under climate change and intensifying human activities. The region is dominated by the East Asian monsoon, which is driven by land–sea thermal contrast. Regional conditions are further shaped by strong interactions among climate, air pollution, ecosystem dynamics, and human activities. Understanding these cross-sphere processes and their associated risks requires high-resolution models that explicitly represent key mechanisms. Regional Earth system models address this need by providing higher resolution and more detailed process representations than global models. This review summarizes the development of the Regional Integrated Earth System Model (RIEMS), led by the Institute of Atmospheric Physics, Chinese Academy of Sciences, in collaboration with Nanjing University and other institutions. We describe RIEMS’s evolution from a regional climate model to a comprehensive Earth system model, emphasizing key advances and the evaluation of its latest version, RIEMS 3.0.The development of RIEMS has progressed through three major stages. Early versions (RIEMS 1.0 and 2.0) established the foundation by integrating land surface processes, ocean components, and atmospheric chemistry with regional atmospheric dynamics. RIEMS 2.0 represented a major step forward by adopting a non-hydrostatic framework (Mesoscale Model 5 version 3; MM5v3) and incorporating spectral nudging, which effectively reduced large-scale circulation drift in long-term integrations. It also integrated the Atmosphere–Vegetation Interaction Model (AVIM) and online aerosol chemistry, enabling robust simulations of vegetation–climate feedbacks and aerosol–monsoon interactions.RIEMS 3.0 marks a shift toward a fully coupled “Atmosphere–Ocean–Land–Human” system. A key advance is the implementation of non-flux-adjusted coupling between the atmospheric component (Weather Research and Forecasting model version 4; WRF v4) and the ocean component (LASG/IAP Climate Ocean Model; LICOM-np) through the Ocean Atmosphere Sea Ice Soil version 3 (OASIS3) coupler. This configuration ensures rigorous conservation of energy and mass across the air–sea interface, substantially improving the representation of critical regional features such as the Western Pacific Subtropical High and tropical cyclone precipitation. To explicitly represent human influences, RIEMS 3.0 incorporates terrestrial carbon and nitrogen (CN) biogeochemical cycles into its land surface schemes (NoahMP-CN and AVIM-CN), enabling dynamic assessment of ecosystem responses to fertilization and nitrogen deposition. In addition, it includes an advanced urban canopy model to represent anthropogenic heat and impervious-surface effects, together with a multi-source satellite data assimilation system for initializing land surface states.Long-term experiments (1991–2014) demonstrate that RIEMS 3.0 effectively reduces systematic biases. For example, it lowers the root-mean-square error of 2-m air temperature over eastern China to approximately 1.0 K, outperforming both standalone WRF simulations and the CMIP6 multi-model mean. In the MICS-Asia III intercomparison, RIEMS 3.0 shows leading performance among participating models in simulating PM2.5 concentrations during severe pollution events in the Beijing–Tianjin–Hebei region.Looking ahead, development of the next-generation RIEMS 4.0 is underway to support national carbon neutrality strategies. Future work will focus on three strategic directions: (1) extending terrestrial biogeochemistry to include phosphorus (P) cycling, enabling complete C–N–P interactions to refine estimates of ecological carbon sinks; (2) building a seamless “land–river–ocean” continuum to simulate the transport of water, nutrients, and pollutants from terrestrial sources to coastal oceans, thereby clarifying mechanisms underlying coastal hypoxia and acidification; and (3) strengthening bidirectional coupling between atmospheric chemistry and physics to better resolve aerosol–cloud–radiation interactions. RIEMS 4.0 also aims to leverage artificial intelligence—through machine-learning parameterizations and differentiable modeling—to improve computational efficiency and predictive skill, providing a robust scientific basis for climate adaptation and sustainable development in East Asia.