El-Bouhali et al. (2025) Performance Evaluation of IMERG Satellite-Based Precipitation Estimates Against Rain Gauge Records in the Sebou Watershed, Morocco

Identification

• Journal: Remote Sensing in Earth Systems Sciences
• Year: 2025
• Date: 2025-11-27
• Authors: Abdelaziz El-Bouhali, Khadija El Ouazani Ech-Chahdi, Mohammed Yazami Ztait, Mhamed Amyay, Mohamed El Mazi
• DOI: 10.1007/s41976-025-00262-z

Research Groups

• ESEAD Laboratory, Ibn Zohr University, FLASH, Ait Melloul, Morocco
• Department of Geography, Sidi Mohamed Ben Abdellah University, Fez, Morocco
• Department of Geography, Moulay Ismail University, Meknès, Morocco

Short Summary

This study evaluates the performance of IMERG-F V06 satellite-based precipitation estimates against ground-based rain gauge records in the Sebou watershed, Morocco, revealing that IMERG-F accuracy is strongly influenced by topography, climatic conditions, and rainfall intensity, with general overestimation in low altitudes and underestimation in high altitudes.

Objective

• To evaluate the performance of IMERG-F V06 precipitation estimates by comparing them with ground-based measurements at both monthly and seasonal timescales, including statistical and spatial analysis, within the Sebou watershed.

Study Configuration

• Spatial Scale: Sebou watershed, northern Morocco, spanning 40,000 km².
• Temporal Scale: 15-year period, from 2000 to 2014.

Methodology and Data

• Models used: Multiple Linear Regression (MLR) model for spatializing ground measurements.
• Data sources:
  • In-situ precipitation data from 30 climate stations provided by the Sebou Hydraulic Basin Agency (ABHS).
  • GPM IMERG-F V06 (Integrated Multi-Satellite Retrievals for GPM Final Run V06) monthly accumulated grids from NASA.

Main Results

• Across the entire Sebou watershed, IMERG-F demonstrates good accuracy in detecting precipitation, with a Correlation Coefficient (CC) of 0.86, a Relative Bias (RBIAS) of 9%, a Mean Absolute Error (MAE) of 19 mm, and a Root Mean Square Error (RMSE) of 33 mm.
• The accuracy of IMERG-F strongly depends on topography, climatic conditions, and rainfall intensity.
• Spatially, IMERG-F underestimates precipitation in high-altitude regions (e.g., Rif and Middle Atlas) during winter, autumn, and spring, while tending to overestimate precipitation at lower altitudes (e.g., Gharb and Saïs).
• Monthly and seasonal evaluations show strong consistency (CC > 0.8) between IMERG-F and in-situ measurements from January to May and September to December, and during autumn, winter, and spring. However, performance weakens significantly in summer (CC < 0.6).
• Errors (MAE and RMSE) are generally low in summer (below 10 mm) but increase significantly in winter, autumn, and spring, exceeding 50 mm for MAE and 60 mm for RMSE in high-precipitation areas.

Contributions

• Provides a comprehensive evaluation of IMERG-F V06 in the Sebou watershed, a region with diverse topography and climate in North Africa, where such assessments are limited.
• Highlights the strong influence of topography, climate, and rainfall intensity on IMERG-F accuracy, offering practical guidance for water resources management and hydro-climatic risk forecasting in the Moroccan context.
• Validates IMERG-F performance using a robust multiple linear regression model for spatial comparison, addressing limitations of simple interpolation methods.

Funding

• Not applicable.

Citation

@article{ElBouhali2025Performance,
  author = {El-Bouhali, Abdelaziz and Ech-Chahdi, Khadija El Ouazani and Ztait, Mohammed Yazami and Amyay, Mhamed and Mazi, Mohamed El},
  title = {Performance Evaluation of IMERG Satellite-Based Precipitation Estimates Against Rain Gauge Records in the Sebou Watershed, Morocco},
  journal = {Remote Sensing in Earth Systems Sciences},
  year = {2025},
  doi = {10.1007/s41976-025-00262-z},
  url = {https://doi.org/10.1007/s41976-025-00262-z}
}