Hu et al. (2025) How well does ERA5 reanalysis depict low-level winds associated with nocturnal rainfall in Sichuan Basin

Identification

• Journal: Atmospheric Research
• Year: 2025
• Date: 2025-10-30
• Authors: Xuelin Hu, Jian Li, Haoming Chen, Rucong Yu
• DOI: 10.1016/j.atmosres.2025.108600

Research Groups

• State Key Laboratory of Severe Weather Meteorological Science and Technology, Chinese Academy of Meteorological Sciences, China Meteorological Administration, Beijing, China
• Institute of Tibetan Plateau Meteorology, Chinese Academy of Meteorological Sciences, China Meteorological Administration, Beijing, China
• Research Center for Disastrous Weather over Hengduan Mountains & Low-Latitude Plateau, China Meteorological Administration, Kunming, China
• Jiangsu Collaborative Innovation Center for Climate Change

Short Summary

This study evaluates ERA5 reanalysis performance in depicting low-level winds associated with nocturnal rainfall in the complex terrain of the Sichuan Basin (SCB) using wind profiler radar observations. While ERA5 generally reproduces key features, it exhibits significant biases in the vertical structure of low-level winds, underestimating the height of maximum wind speed and failing to capture the observed increase in circulation with height.

Objective

• Evaluate the performance of ERA5 reanalysis in depicting low-level winds associated with nocturnal rainfall in the Sichuan Basin (SCB).

Study Configuration

• Spatial Scale: Sichuan Basin (SCB) and Yungui Plateau (YGP) in southwestern China. Specific locations include four wind profiler radar stations (Guiyang, Shapingba, Guangyuan, Pixian) and 101 rain gauge stations within the SCB.
• Temporal Scale: Summer months (June, July, August) from 2023 to 2024, using hourly data.

Methodology and Data

• Models used: ERA5 reanalysis (fifth-generation ECMWF reanalysis).
• Data sources:
  • Hourly wind profiles (horizontal wind speed and direction) from four L-band boundary layer wind profiler radars (WPR) at Guiyang, Shapingba, Guangyuan, and Pixian.
  • Hourly rain gauge records from 101 stations within the Sichuan Basin (SCB).
  • Hourly ERA5 reanalysis data, including horizontal winds on pressure levels and surface precipitation.

Main Results

• ERA5 generally reproduces the nocturnal flow enhancement over the Yungui Plateau (YGP) and its positive correlation with low-level vorticity in the Sichuan Basin (SCB), consistent with previous studies.
• However, ERA5 significantly underestimates the height of the maximum wind speed (hm) over the YGP (e.g., observed mean hm at Guiyang is 990 meters, ERA5 is less than half) while overestimating its magnitude.
• Observations show an "elevated peak" ("prominent nose") in vertical wind speed profiles, whereas ERA5 depicts a "stronger but lower-altitude maximum" ("flatter nose").
• ERA5 overestimates the lower-level intensity of circulation within the SCB and fails to reproduce the observed increase in circulation with height (e.g., observed circulation increases by approximately 5 x 10^6 square meters per second from 500 meters to 1500 meters, while ERA5 shows uniform circulation with height).
• ERA5 underestimates the occurrence of nocturnal low-level jets (NLLJs) and does not capture the observed decoupling of diurnal wind variation between the near-surface and higher layers.
• The average 200-1000 meter vertical wind shear is significantly underestimated by ERA5 (observed: 3.6 meters per second; ERA5: -0.4 meters per second).
• Despite these biases in low-level wind structure, ERA5 reproduces the diurnal cycle of precipitation reasonably well, suggesting a "right results for wrong reasons" phenomenon.

Contributions

• Provides a rigorous, observation-based evaluation of ERA5 reanalysis performance in depicting low-level winds over complex terrain (Sichuan Basin and Yungui Plateau), specifically for nocturnal rainfall events.
• Quantifies significant biases in ERA5's representation of the vertical structure of low-level winds, including the height and magnitude of maximum wind speed and vertical circulation profiles, which were previously identified using reanalysis/model outputs.
• Highlights that even if precipitation is reasonably reproduced, reanalysis data may misrepresent underlying atmospheric mechanisms due to structural wind differences, cautioning against uncritical use in complex terrain studies.
• Emphasizes the critical need for enhanced observational data to constrain uncertainties in reanalysis and improve their reliability for scientific understanding and practical applications like low-level aviation and wind energy assessment.

Funding

• National Natural Science Foundation of China (42225505, 42405010, U2142204)
• Jiangsu Collaborative Innovation Center for Climate Change
• Basic Research Fund of CAMS (2023Y002, 2023Z010)

Citation

@article{Hu2025How,
  author = {Hu, Xuelin and Li, Jian and Chen, Haoming and Yu, Rucong},
  title = {How well does ERA5 reanalysis depict low-level winds associated with nocturnal rainfall in Sichuan Basin},
  journal = {Atmospheric Research},
  year = {2025},
  doi = {10.1016/j.atmosres.2025.108600},
  url = {https://doi.org/10.1016/j.atmosres.2025.108600}
}