Jeong et al. (2026) Evaluating North American winter extreme precipitation variability and its drivers in CMIP6 models

Identification

• Journal: Weather and Climate Extremes
• Year: 2026
• Date: 2026-08-23
• Authors: Dae Il Jeong, Bin Yu, Alex J. Cannon
• DOI: 10.1016/j.wace.2026.100948

Research Groups

• Climate Modelling Research Division, Environment and Climate Change Canada, Toronto, Ontario, M3H 5T4, Canada
• Climate Modelling Research Division, Environment and Climate Change Canada, Victoria, British Columbia, V8W 2Y2, Canada

Short Summary

This study evaluates the ability of 32 CMIP6 global climate models (GCMs) to capture winter extreme precipitation variability over North America. The analysis focuses on monthly maximum 5-day precipitation (Rx5day) and its drivers during extended winters (November–March, 1961–2010). The results demonstrate that the multi-model ensemble provides a more robust representation of extreme precipitation than individual models.

Objective

• Investigate the ability of CMIP6 GCMs to simulate winter extreme precipitation variability over North America.
• Examine the relationships between Rx5day variability and large-scale climate drivers, including atmospheric circulation patterns and moisture fluxes.

Study Configuration

• Spatial Scale: Regional (North America)
• Temporal Scale: Monthly (November–March, 1961–2010)

Methodology and Data

• Models used: CMIP6 GCMs (32 models), including two initial-condition ensembles from CanESM5 and MPI-ESM1-2-LR.
• Data sources: ERA5 and JRA-55 reanalysis datasets, GPCC Full Data Daily dataset.

Main Results

• The multi-model ensemble captures the spatial patterns of climatological Rx5day statistics over North America with high accuracy (spatial pattern correlations > 0.82).
• Three distinct modes of monthly Rx5day variability are identified using common EOF analysis, explaining 12.3%, 10.4%, and 8.0% of the total variability.
• The amplitude ratios of most GCMs fall within the 'about right' range (0.8–1.2), indicating good agreement with the common PCs.

Contributions

• This study provides a comprehensive evaluation of CMIP6 GCMs in simulating winter extreme precipitation variability over North America.
• The results highlight the importance of internal climate variability in contributing to uncertainty in Rx5day variability.
• The multi-model ensemble is shown to provide a more robust representation of extreme precipitation than individual models.

Funding

• This research was funded by Environment and Climate Change Canada.

Citation

@article{Jeong2026Evaluating,
  author = {Jeong, Dae Il and Yu, Bin and Cannon, Alex J.},
  title = {Evaluating North American winter extreme precipitation variability and its drivers in CMIP6 models},
  journal = {Weather and Climate Extremes},
  year = {2026},
  doi = {10.1016/j.wace.2026.100948},
  url = {https://doi.org/10.1016/j.wace.2026.100948}
}