Zi et al. (2026) East Asian spring precipitation and its dry trend revealed by CMIP6 high-resolution coupled models

Identification

• Journal: Climate Dynamics
• Year: 2026
• Date: 2026-03-30
• Authors: Peng Zi, Jiandong Li, Rui Yang, Yimin Liu, Bian He, Qing Bao, Zihan Yang, T. Li
• DOI: 10.1007/s00382-026-08151-x

Research Groups

• Yunnan Key Laboratory of Meteorological Disasters and Climate Resources in the Greater Mekong Subregion, Yunnan University, Kunming, China
• State Key Laboratory of Earth System Numerical Modeling and Application, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
• Key Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources, CAS, Beijing, China

Short Summary

This study evaluates CMIP6 HighResMIP coupled models' ability to simulate East Asian spring precipitation climatology and its drying trend (1980–2014), finding that high-resolution models generally improve mean precipitation patterns but show varied and often limited success in reproducing the observed drying trend, with only a few models demonstrating significant improvement due to better representation of regional circulation drivers.

Objective

• To investigate whether high-resolution coupled models simulate East Asian spring precipitation climatology and its drying trend more accurately than low-resolution models.
• To determine if these models can reproduce regional circulation systems and remote large-scale forcings that govern East Asian spring precipitation and its changes.

Study Configuration

• Spatial Scale: East Asia (specifically the main rainfall belt: 110°E–135°E, 20°N–36°N). Global for model simulations and data. Model horizontal resolutions: standard (low) and high (≤ 50 km). Observational data resolutions: 0.1° (MSWEP), 0.625° × 0.5° (MERRA2). All data interpolated to a common 0.25° grid for analysis.
• Temporal Scale: Analysis period: 1980–2014. Spring season defined as February-April (FMA). Model historical coupled experiments: 1950–2014.

Methodology and Data

• Models used: Seven pairs of CMIP6 HighResMIP Tier 2 historical coupled models: CMCC-CM2, CNRM-CM6-1, EC-Earth3P, ECMWF-IFS, FGOALS-f3, HadGEM3-GC31, and MPI-ESM1-2.
• Data sources:
  • Precipitation: Multi-Source Weighted-Ensemble Precipitation (MSWEP) dataset (high-resolution observational, 3-hourly, 0.1°).
  • Atmospheric circulation: NASA Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA2) reanalysis (0.625° × 0.5°).
  • Atlantic Multidecadal Oscillation (AMO) index.
  • Methods: Skill score, Vector Similarity Coefficient (VSC) for vector fields, Relative-weight analysis (RWA), Bilinear interpolation, Mann-Kendall (MK) test for trend detection.

Main Results

• East Asian spring precipitation (FMA, 1980–2014) exhibits a significant drying trend, observed at -0.43 mm day⁻¹ decade⁻¹ (MSWEP).
• This drying trend is primarily driven by the combined effects of an intensified Mongolia High (observed trend of 0.56 decade⁻¹) and a weakened Western North Pacific Anticyclone (observed trend of -0.21 decade⁻¹), which together contribute 63% of the variance.
• Remote forcings, such as Pacific K-type warming (weakening WNPA) and North Atlantic warming (strengthening Mongolia High, AMO index trend of 0.09 °C decade⁻¹), modulate these regional circulation systems.
• High-resolution models generally outperform their low-resolution counterparts in simulating the spatial pattern and intensity of East Asian spring mean precipitation, with the high-resolution multi-model ensemble (MME) skill score increasing from 0.78 to 0.84 (pattern correlation from 0.76 to 0.86). This improvement is linked to better representations of low-level winds and moisture transport, and a reduction in the underestimation of extreme precipitation events (≥ 20 mm day⁻¹).
• However, most high-resolution models struggle to reproduce the observed long-term drying trend. Only two models (FGOALS-f3-H and EC-Earth3P-HR) show improved performance for this trend.
• FGOALS-f3-H exhibits the best skill, increasing its regional mean drying trend from -0.10 mm day⁻¹ decade⁻¹ (low-resolution) to -0.33 mm day⁻¹ decade⁻¹, which is closer to the observed -0.43 mm day⁻¹ decade⁻¹. This improvement is attributed to more realistic representations of the weakening Western North Pacific Anticyclone, strengthening Mongolia High, and weakening vertical velocity over East Asia.
• Conversely, five out of the seven high-resolution models show degraded performance or amplified biases in reproducing the precipitation trend, sometimes even simulating a wetting trend.
• The limited improvement in trend simulation is linked to insufficient representation of changes in key atmospheric circulation systems and their teleconnections, as well as potential limitations in physical parameterizations and the representation of internal climate variability.

Contributions

• Provides the first comprehensive evaluation of CMIP6 HighResMIP coupled models for East Asian spring precipitation climatology and its drying trend.
• Quantifies the impact of increased horizontal resolution on simulating both the mean state and long-term trends of East Asian spring precipitation.
• Identifies specific high-resolution models that show significant improvements in simulating the drying trend and elucidates the underlying physical mechanisms (e.g., improved representation of Western North Pacific Anticyclone, Mongolia High, and vertical velocity).
• Highlights that resolution enhancement alone is often insufficient for improving trend simulations, emphasizing the critical need for simultaneous improvements in physical parameterizations governing precipitation-related interannual variability.
• Utilizes high-resolution observational data (MSWEP) for a more accurate evaluation of high-resolution models.

Funding

• National Key Research and Development Program of China (2022YFF0802003)
• National Science Foundation of China (42288101, 42275026)
• Natural Science Foundation of Yunnan Province (202301AV070001, 202302AN360006)
• National large Scientific and Technological Infrastructure “Earth System Numerical Simulation Facility” (https://cstr.cn/31134.02.EL)

Citation

@article{Zi2026East,
  author = {Zi, Peng and Li, Jiandong and Yang, Rui and Liu, Yimin and He, Bian and Bao, Qing and Yang, Zihan and Li, T.},
  title = {East Asian spring precipitation and its dry trend revealed by CMIP6 high-resolution coupled models},
  journal = {Climate Dynamics},
  year = {2026},
  doi = {10.1007/s00382-026-08151-x},
  url = {https://doi.org/10.1007/s00382-026-08151-x}
}