Paquin et al. (2025) The Ouranos CRCM5-CMIP6 ensemble: A dynamically downscaled ensemble of CMIP6 simulations over North America

Identification

• Journal: Scientific Data
• Year: 2025
• Date: 2025-12-12
• Authors: Dominique Paquin, Christopher D. McCray, Charles B. Gauthier, Michel Giguère, Olivier Asselin, Pascal Bourgault, Marie-Pier Labonté, Dominic Matte
• DOI: 10.1038/s41597-025-06289-7

Research Groups

• Ouranos, Montréal, Québec, Canada
• Centre pour l’étude et la simulation du climat à l’échelle régionale (ESCER) at Université du Québec à Montréal (UQAM)
• Environment and Climate Change Canada (ECCC)

Short Summary

This paper presents Ouranos' contribution to the North American Coordinated Regional Climate Downscaling Experiment (NA-CORDEX) by providing a dynamically downscaled ensemble of CMIP6 simulations over North America using the Canadian Regional Climate Model (CRCM5). It describes the model configuration, validates its performance against reanalysis data, and details the generated climate projections under various future emission scenarios, making the high-resolution dataset publicly available.

Objective

• To generate and validate a dynamically downscaled ensemble of CMIP6 climate simulations over North America using the CRCM5 model, contributing to the NA-CORDEX project, and to provide comprehensive access to this dataset for climate change impact and adaptation studies.

Study Configuration

• Spatial Scale: North America, with a free analysis zone of 655x628 grid points at 0.11° horizontal grid spacing (approximately 12 km resolution).
• Temporal Scale: Historical simulations from 1950 to 2014; future climate simulations from 2015 to 2100. An evaluation simulation driven by ERA5 covers 1979-2020.

Methodology and Data

• Models used:
  • Regional Climate Model: Canadian Regional Climate Model version 5 (CRCM5), based on the Global Environmental Multiscale (GEM) model version 3.3.3.1.
  • Land Surface Scheme: Canadian Land Surface Scheme (CLASS) version 3.5.
  • Lake Model: Freshwater Lake (FLake) model.
  • Driving Global Climate Models (GCMs) from CMIP6: CanESM5, CNRM-ESM2-1, MPI-ESM-1-2-LR, NorESM2-MM.
• Data sources:
  • Reanalysis: ERA5 (European Centre for Medium-Range Weather Forecasts Reanalysis, 0.25° grid spacing).
  • Global Climate Model data: CMIP6 ScenarioMIP project (Earth System Grid Federation - ESGF).
  • Geophysical fields for CRCM5: ECOCLIMAP (soil texture, 1 km), Shangguan (bedrock depth, ≈30 m x 30 m), USGS-GLCC (land cover, 900 m), GTOPO30 (topography, 900 m).
  • Emissions Scenarios: Shared Socioeconomic Pathways (SSPs) including SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5.

Main Results

• Structural Biases (CRCM5 vs. ERA5): CRCM5 exhibits relatively small temperature biases, except for a strong wintertime cold bias over Greenland and parts of the Canadian/U.S. Plains, and a weaker summer warm bias (0.5-3 °C) over much of the United States and western Canada. Precipitation biases show an east-west gradient, with wet biases (10-50%, 0.5-2 mm/day) over the eastern continent and dry biases from the Rocky Mountains westward, strongest in summer.
• GCM-driven Simulation Biases: Biases in GCM-driven simulations are a combination of CRCM5 structural biases and inherited GCM biases. Most regions show a winter cold bias, sometimes exceeding 5 °C over the U.S. Northern Plains and Canadian Prairies. Wet biases are present over most regions in winter, except western Canada and parts of the northwestern U.S., with strong (>200%) wet biases off the Pacific coast of Mexico and California.
• Seasonal Cycle Representation: Seasonal cycles of temperature and precipitation are generally well represented across IPCC climate reference regions, with minor biases (e.g., 0-2 °C cold bias in spring, 1-4 °C warm bias in early winter for Northeastern North America).
• Future Climate Projections (SSP3-7.0, 2071-2100 vs. 1971-2000): Projected temperature changes are strongly correlated with the Equilibrium Climate Sensitivity (ECS) of the driving GCMs, with the highest warming (>10 °C in winter over Canada, Greenland, and Alaska) projected by the CanESM5-driven simulation (highest ECS). Precipitation changes show broad coherence in winter (decreasing over southwestern North America, increasing elsewhere, with >200% increases over the Arctic in some simulations), while summer changes are more uncertain.
• Global Warming Levels (GWLs): At a +2 °C global warming level (relative to 1850-1900), the Northeastern North America region projects a regional warming of 3.1-4.4 °C and a precipitation increase of 0.15-0.24 mm/day (relative to 1971-2000) among the nine ensemble members that reach this GWL. This approach reduces uncertainty related to climate sensitivity and scenario compared to fixed end-of-century comparisons.

Contributions

• Provides a comprehensive, high-resolution (0.11° / ~12 km) dynamically downscaled ensemble of CMIP6 climate simulations over North America, serving as a significant contribution to the NA-CORDEX project.
• Offers a valuable dataset for a wide range of climate change impact and adaptation studies, covering diverse emission scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) and a broad spectrum of driving GCM climate sensitivities (2.54 K to 5.64 K ECS).
• Details the model configuration, validation against reanalysis, and evaluation of GCM-driven simulations, ensuring transparency and usability for the scientific community.
• Makes the extensive dataset publicly accessible through a THREDDS Data Server, facilitating further research and analysis.

Funding

• Ouranos (generated and supplied CRCM5 data)
• Calcul Québec and the Digital Research Alliance of Canada (supercomputer operations on Beluga and Narval)
• Innovation, Science and Economic Development Canada (financial support for supercomputer operations)
• Ministère de l’Économie et de l’Innovation du Québec (financial support for supercomputer operations)

Citation

@article{Paquin2025Ouranos,
  author = {Paquin, Dominique and McCray, Christopher D. and Gauthier, Charles B. and Giguère, Michel and Asselin, Olivier and Bourgault, Pascal and Labonté, Marie-Pier and Matte, Dominic},
  title = {The Ouranos CRCM5-CMIP6 ensemble: A dynamically downscaled ensemble of CMIP6 simulations over North America},
  journal = {Scientific Data},
  year = {2025},
  doi = {10.1038/s41597-025-06289-7},
  url = {https://doi.org/10.1038/s41597-025-06289-7}
}