Wang et al. (2026) Evaluating WRF simulated temperature uncertainties across Northern Hemisphere climate zones with different land surface models and land cover datasets

Identification

• Journal: Atmospheric Research
• Year: 2026
• Date: 2026-03-10
• Authors: Bin Wang, Peng Wang, Jiandong Wang, Yuan Wang, Qi Ying, Zilu Zhang, Yuzhi Jin, Jinbo Wang, Meng Gao, Chao Liu, Xin Huang, Sa Wang
• DOI: 10.1016/j.atmosres.2026.108897

Research Groups

• State Key Laboratory of Climate System Prediction and Risk Management, Nanjing University of Information Science and Technology, China
• Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, School of Atmospheric Physics, Nanjing University of Information Science and Technology, China
• Department of Atmospheric and Oceanic Sciences, Fudan University, China
• Division of Environment and Sustainability, Hong Kong University of Science and Technology, Hong Kong SAR
• Joint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, China
• National Observation and Research Station for Atmospheric Processes and Environmental Change in Yangtze River Delta, China
• Department of Geography, Hong Kong Baptist University, Hong Kong SAR
• State Key Laboratory of Regional Environment and Sustainability, Beijing, China
• School of Environment, Tsinghua University, China

Short Summary

This study systematically evaluated the temperature simulation performance of four land surface models (LSMs) and three land cover (LC) datasets within the WRF model across the Northern Hemisphere. The CGLC-Noah combination demonstrated the best overall performance, though significant temperature underestimations were found in tropical and polar regions.

Objective

• To systematically evaluate the temperature simulation performance of different land surface models (LSMs) and land cover (LC) datasets within the WRF model across various climate zones in the Northern Hemisphere, including during extreme weather events.

Study Configuration

• Spatial Scale: Northern Hemisphere, covering tropical, temperate, and polar climate zones.
• Temporal Scale: Not explicitly stated for the simulation period in the provided text, but land cover datasets from 2010 and 2019 were used.

Methodology and Data

• Models used: WRF (Weather Research and Forecasting) model, CLM (Community Land Model), Noah (Noah Multi-Physics Land Surface Model), PX (Pleim-Xiu Land Surface Model), RUC (Rapid Update Cycle Land Surface Model).
• Data sources: MODIS-2010 (Land Cover Dataset), MODIS-2019 (Land Cover Dataset), CGLC-MODIS-LCZ (Global Land Cover Land Cover Zones Dataset).

Main Results

• The CGLC-Noah combination demonstrated the best overall temperature simulation performance, achieving optimal results at 44% of stations across the Northern Hemisphere.
• The MD19-PX combination was the second best, performing optimally at 19% of stations, with strong performance concentrated in cold, high-latitude regions of North America and Eastern Europe.
• Temperatures were underestimated in tropical regions with an average mean bias (MB) of -1.42 °C and in polar regions with an average MB of -0.66 °C.
• Temperate regions exhibited relatively small MB, ranging from -1 °C to +1 °C.
• During extreme weather events, all simulations showed optimal performance (normalized root mean square error (NRMSE) > 0.8) for both extreme high and low temperatures in tropical regions.
• The lowest performance during extreme weather events was observed in polar regions, with NRMSE typically below 0.2.

Contributions

• Provides a systematic and comprehensive evaluation of the combined impact of multiple land surface models and land cover datasets on WRF-simulated temperatures across diverse Northern Hemisphere climate zones.
• Identifies optimal LSM-LC dataset combinations for different regions, such as CGLC-Noah for overall performance and MD19-PX for cold, high-latitude areas.
• Quantifies regional biases in temperature simulations, highlighting significant underestimations in tropical and polar regions.
• Assesses model performance during extreme high and low temperature events, revealing regional strengths and weaknesses.
• Offers crucial insights for improving the accuracy and reliability of climate models, which is vital for developing effective climate change adaptation and mitigation strategies.

Funding

Not explicitly stated in the provided text.

Citation

@article{Wang2026Evaluating,
  author = {Wang, Bin and Wang, Peng and Wang, Jiandong and Wang, Yuan and Ying, Qi and Zhang, Zilu and Jin, Yuzhi and Wang, Jinbo and Wang, Jinbo and Gao, Meng and Liu, Chao and Huang, Xin and Wang, Sa},
  title = {Evaluating WRF simulated temperature uncertainties across Northern Hemisphere climate zones with different land surface models and land cover datasets},
  journal = {Atmospheric Research},
  year = {2026},
  doi = {10.1016/j.atmosres.2026.108897},
  url = {https://doi.org/10.1016/j.atmosres.2026.108897}
}