Izzaddin et al. (2025) How well do RCMs simulate Portugal’s climate?

Identification

• Journal: Theoretical and Applied Climatology
• Year: 2025
• Date: 2025-10-01
• Authors: Aras Izzaddin, Marwah Yaseen, Vito Iacobellis
• DOI: 10.1007/s00704-025-05718-2

Research Groups

• Department of Civil, Environmental, Land, Building Engineering and Chemistry, Polytechnic University of Bari, Italy

Short Summary

This study evaluates the performance of EURO-CORDEX regional climate models (RCMs) in simulating daily temperature and precipitation over Portugal (1971-2004), finding that RCMs reliably capture mean temperature indices but struggle with extreme precipitation events.

Objective

• To evaluate the performance of EURO-CORDEX regional climate models (RCMs) in simulating daily precipitation, daily minimum temperature, and daily maximum temperature, including nine mean and extreme climate indices, over Portugal.

Study Configuration

• Spatial Scale: Portugal, covering approximately 92,000 km², from 37°N to 42°N latitude and 6°W to 9.5°W longitude, at 0.11° (approximately 12.2 km) resolution.
• Temporal Scale: Daily variables for the period 1971 to 2004.

Methodology and Data

• Models used: EURO-CORDEX regional climate model (RCM) simulations (historical experiment r1i1p1 runs) driven by multiple global climate models (GCMs) from the Coupled Model Intercomparison Project Phase 5 (CMIP5).
• Data sources: Two high-resolution observational datasets: E-OBS (version 28.0, 0.1° resolution) and Iberia01 (0.11° resolution).

Main Results

• Model performance varied significantly by variable and climate index.
• Maximum temperature mean indices showed the highest accuracy, with total errors generally below 30%.
• Minimum temperature mean indices followed, with total errors under 40%.
• Precipitation indices exhibited the greatest uncertainties, with total errors exceeding 50% for extreme precipitation indices (e.g., annual maximum one-day precipitation (Rx1day) and consecutive dry days (CDD)).
• Mean climate indices were generally better captured than extreme indices; over 85% of model simulations achieved total errors below 50% for mean temperature indices.
• For annual total precipitation (PRCP), 85% of models exhibited a wet bias and 87% overestimated variability relative to Iberia01.
• For Rx1day and CDD, no models achieved total errors below 50%, with many showing errors exceeding 100% for Rx1day.
• Most models exhibited a cold bias for mean maximum temperature (TXmean).
• Spatial analysis revealed localized strengths and weaknesses across Portugal, with temperature indices performing slightly better at lower elevations.
• The Aras diagram proved an effective tool for comprehensive model performance evaluation, integrating bias, variability, correlation, and total error.

Contributions

• Provides a comprehensive evaluation of EURO-CORDEX RCMs over Portugal using a novel diagnostic tool (Aras diagram), integrating bias, variability, correlation, and total error into a single framework.
• Identifies the most reliable EURO-CORDEX RCMs for climate impact studies and adaptation planning in Portugal.
• Offers valuable feedback for future regional climate model development by highlighting specific model biases and limitations, particularly for extreme precipitation events.
• Enhances the robustness of model evaluation by using two independent high-resolution observational datasets (Iberia01 and E-OBS), reducing observational uncertainty.

Funding

• Open access funding provided by Politecnico di Bari within the CRUI-CARE Agreement.

Citation

@article{Izzaddin2025How,
  author = {Izzaddin, Aras and Yaseen, Marwah and Iacobellis, Vito},
  title = {How well do RCMs simulate Portugal’s climate?},
  journal = {Theoretical and Applied Climatology},
  year = {2025},
  doi = {10.1007/s00704-025-05718-2},
  url = {https://doi.org/10.1007/s00704-025-05718-2}
}