Zhang et al. (2025) Evaluation and statistical bias correction of ERA5-Land meteorological variables for a humid river basin in Southwest China

Identification

• Journal: Scientific Reports
• Year: 2025
• Date: 2025-11-20
• Authors: Lu Zhang, Zhiyu Yan, Kunlun Huang, Wei Zhang
• DOI: 10.1038/s41598-025-24942-4

Research Groups

• Institute of Science and Technology, China Three Gorges Corporation, Beijing, China
• National Engineering Research Center of Water Resources Efficient Utilization and Engineering Safety, Hohai University, Nanjing, China
• Global Energy Interconnection Group Co., Ltd, Beijing, China

Short Summary

This study evaluates the performance of the ERA5-Land reanalysis dataset for key meteorological variables in the Lower Jinsha River Basin, China, and develops a statistical bias correction procedure. The correction significantly reduces systematic biases and improves the accuracy of precipitation, wind speed, air temperature, and solar radiation, providing a more reliable dataset for clean energy planning in data-scarce regions.

Objective

• To comprehensively assess the accuracy of ERA5-Land estimates for multiple meteorological variables (precipitation, wind speed, air temperature, and solar radiation).
• To develop a statistical bias correction procedure for these variables to enhance estimate quality at daily and hourly scales.
• To investigate the spatial and temporal variations of multiple variables using the corrected ERA5-Land, providing insights for the sustainable development of clean energy bases.

Study Configuration

• Spatial Scale: Lower Jinsha River Basin (LJRB), Southwest China, spanning approximately 340,000 km².
• Temporal Scale: 1980–2019 (40 years). For solar radiation observations, the period was 1980–2016.

Methodology and Data

• Models used:
  • Statistical bias correction procedure combining month-specific regression fitting (Mode 1: modeling relationship between ERA5-Land values and discrepancies; Mode 2: modeling relationship between ERA5-Land estimates and ground observations).
  • Seven candidate function forms were considered for regression, with the cubic function generally showing superior performance.
  • Daily and hourly adjustments were applied, including scaling factors for precipitation and solar radiation, and both scaling and offsetting factors for air temperature.
• Data sources:
  • Ground gauge observations: Daily precipitation, wind speed, mean/maximum/minimum air temperature, and solar radiation from the China Meteorological Administration (CMA).
  • Reanalysis data: ERA5-Land product (approximately 9 km spatial resolution, hourly temporal resolution) from the European Centre for Medium-Range Weather Forecasts (ECMWF) / Copernicus Climate Data Store (CDS).

Main Results

• Raw ERA5-Land performance: Air temperature estimates showed the best agreement with ground observations (coefficient of determination (R²) > 0.87, percent bias (Pbias) < 18%, absolute errors < 4 °C), followed by solar radiation (R² = 0.63). Precipitation and wind speed exhibited larger uncertainties (R² < 0.31, Pbias up to 67.76% for precipitation and -23.36% for wind speed). Performance varied seasonally and with elevation.
• Bias Correction effectiveness: The statistical bias correction procedure largely eliminated systematic biases (Pbias reduced to approximately 0% for all variables). Absolute errors (Root Mean Square Error (RMSE) and Mean Absolute Error (MAE)) decreased by more than 10% for most variables. Temporal consistency (R²) improved moderately, with the most notable increases for wind speed (+29.5%) and solar radiation (+25.8%).
• Long-term meteorological variations (1980–2019) using corrected data:
  • Precipitation: Decreased across most of the basin, with an areal mean change rate of -2.4 mm/yr (p < 0.05).
  • Air temperature: Increased across the entire basin, with an areal mean change rate of 0.03 °C/yr (p < 0.05).
  • Solar radiation: Showed an upward trend, increasing by 1078 W/(m²·yr) across the basin (p < 0.05).
  • Wind speed: Exhibited increasing trends in northwestern areas and decreasing trends in southeastern areas.
  • Intra-annual patterns: Precipitation concentrated from June to September; wind speed higher from February to April; air temperature peaked in summer; solar radiation abundant during spring and summer.
  • Intra-day patterns: Precipitation primarily concentrated at night; wind speed and air temperature peaked between 13:00 and 18:00 local time; solar radiation reached maximum values between 11:00 and 13:00 local time.

Contributions

• Provides a comprehensive evaluation of ERA5-Land's accuracy for multiple meteorological variables crucial for clean energy applications in a complex, humid river basin in Southwest China.
• Introduces a practical statistical bias correction framework that combines month-specific regression fitting with daily and hourly adjustments, specifically designed for data-scarce regions.
• Offers a systematic understanding of ERA5-Land uncertainties and a methodological reference for enhancing reanalysis data applicability in similar geographical and climatic conditions, particularly for clean energy development.
• Demonstrates the improved representation of long-term climatic variations and spatiotemporal heterogeneity using the corrected dataset, which is vital for resource assessment and planning.

Funding

• National Natural Science Foundation of China (Grant No. 52409021)

Citation

@article{Zhang2025Evaluation,
  author = {Zhang, Lu and Yan, Zhiyu and Huang, Kunlun and Zhang, Wei},
  title = {Evaluation and statistical bias correction of ERA5-Land meteorological variables for a humid river basin in Southwest China},
  journal = {Scientific Reports},
  year = {2025},
  doi = {10.1038/s41598-025-24942-4},
  url = {https://doi.org/10.1038/s41598-025-24942-4}
}