Chakri et al. (2025) Spatial Bias Correction of ERA5_Ag Reanalysis Precipitation Using Machine Learning Models in Semi-Arid Region of Morocco

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Atmosphere
• Year: 2025
• Date: 2025-10-26
• Authors: Achraf Chakri, Sana Abakarim, João Antunes Rodrigues, Nour‐Eddine Laftouhi, Hassan Ibouh, Lahcen Zouhri, Elena Zaitseva
• DOI: 10.3390/atmos16111234

Research Groups

[Information not provided in the text.]

Short Summary

This study aimed to correct ERA5_Ag reanalysis precipitation data using machine learning models and observational data in the Tensift basin, Morocco. It achieved significant improvements in precipitation accuracy, with R2 values between 0.80 and 0.90, and generated 42-year corrected raster maps for water resource management.

Objective

• To correct precipitation values from the ERA5_Ag reanalysis dataset using observational data from 20 meteorological stations in the Tensift basin, Morocco, to improve accuracy for water resource management.

Study Configuration

• Spatial Scale: Tensift basin, Morocco (regional basin scale, with 20 meteorological stations and generated raster maps).
• Temporal Scale: 42 years (for the generation of corrected precipitation raster maps).

Methodology and Data

• Models used: MLP (Multilayer Perceptron), XGBoost, CatBoost, LightGBM, Random Forest (all machine learning models).
• Data sources: ERA5_Ag reanalysis dataset, observational data from 20 meteorological stations.

Main Results

• Significant improvements were observed in the accuracy of precipitation estimates.
• R2 values for the corrected precipitation ranged between 0.80 and 0.90 in most stations.
• The best-performing machine learning model was used to generate spatially detailed raster maps of corrected precipitation over a 42-year period.

Contributions

• Provides a valuable, spatially detailed tool for water resource management through improved precipitation accuracy.
• Addresses the critical need for more accurate and informed decision-making regarding water scarcity in semi-arid regions like the Tensift basin.

Funding

[Information not provided in the text.]

Citation

@article{Chakri2025Spatial,
  author = {Chakri, Achraf and Abakarim, Sana and Rodrigues, João Antunes and Laftouhi, Nour‐Eddine and Ibouh, Hassan and Zouhri, Lahcen and Zaitseva, Elena},
  title = {Spatial Bias Correction of ERA5_Ag Reanalysis Precipitation Using Machine Learning Models in Semi-Arid Region of Morocco},
  journal = {Atmosphere},
  year = {2025},
  doi = {10.3390/atmos16111234},
  url = {https://doi.org/10.3390/atmos16111234}
}