Garouani et al. (2025) Earth observation open data in Google Earth Engine for water resource management in the Saïss Plain, Morocco

Identification

• Journal: Mediterranean Geoscience Reviews
• Year: 2025
• Date: 2025-11-18
• Authors: Manal El Garouani, Mohamed Ali El-Omairi, Maryame El-Yazidi, Abderrahim Lahrach, Hassane Jarar Oulidi
• DOI: 10.1007/s42990-025-00204-3

Research Groups

• FST-Fez, Sidi Mohamed Ben Abdallah University, Fez, Morocco
• ENSA, Sidi Mohamed Ben Abdallah University, Fez, Morocco
• Hassania School for Public Works Engineering, Casablanca, Morocco

Short Summary

This study utilizes Google Earth Engine (GEE) and open Earth observation data to analyze water balance and monitor drought conditions in Morocco's Saïss Plain, revealing seasonal water deficits in summer/autumn and identifying vulnerable areas prone to recurring water scarcity.

Objective

• To investigate the current state of water resources in the Saïss Plain, northern Morocco, using new geospatial technologies, GEE, and open data.
• To provide a convenient, freely available tool for water managers and users to evaluate water budget components.
• To use open data for hydrological research, explore GEE for water balance analysis, and assess existing water resources to support improved management strategies.

Study Configuration

• Spatial Scale: Regional scale, focusing on the Saïss Plain in northwest Morocco (approximately 2200 km², 100 km long by 30 km wide). Data resolutions range from 250 meters (MODIS EVI/NDVI) to 9 kilometers (ERA5-Land).
• Temporal Scale: Analysis period primarily from 2000 to 2023, with specific components varying (e.g., SVI dynamics from 2000–2022, water balance components from 2013–2023). Monthly and annual variations are studied.

Methodology and Data

• Models used:
  • Water Balance Equation: P + SM = ET + R + ΔS (Precipitation + Soil Moisture = Evapotranspiration + Runoff + Change in Storage)
  • Evapotranspiration (ET): Penman–Monteith equation (MOD16 algorithm)
  • Surface Runoff: Curve Number (CN) method
  • Drought Monitoring: Standardized Vegetation Index (SVI) based on NDVI/EVI anomaly concept
• Data sources:
  • Platform: Google Earth Engine (GEE)
  • Land Use/Land Cover (LULC): MCD12Q1.006 MODIS Land Cover
  • Precipitation (P): Climate Hazards Group InfraRed Precipitation with Station Data (CHIRPS)
  • Soil Data: OpenLandMap
  • Soil Moisture (SM): TerraClimate, ERA5-Land dataset
  • Evapotranspiration (ET): MODIS (MOD16) products
  • Vegetation Index (EVI/NDVI for SVI): MODIS dataset (16-day NDVI composites at 250-meter resolution)
  • Groundwater levels: Local piezometer data from ABHS (Agence du Bassin Hydraulique de Sebou) for validation and context.
  • Field measurements: Used for validation of GEE-derived precipitation and evapotranspiration.

Main Results

• The water balance in the Saïss Plain is positive during winter and spring, and negative during summer and autumn.
• Areas with excess water are located in the center of the study zone, contributing to aquifer recharge, while deficit zones are found in the north and northeast.
• Drought monitoring using the Standardized Vegetation Index (SVI) shows lowest values (indicating water scarcity) typically from May to November, with recurring droughts every 2 to 3 years.
• The northwest and southeast regions of the plain are particularly vulnerable to water scarcity due to high population density and reduced vegetation cover, contrasting with the relatively stable southern and northeastern parts.
• GEE-derived precipitation estimates show a high correlation (0.91) with measured values, confirming the reliability of satellite-based methods.
• Soil moisture ranges from 0 to 30 millimeters, increasing in winter and decreasing in summer, with the northeastern part having significantly lower soil moisture.
• Evapotranspiration exhibits seasonal variations (5 to 96 millimeters per month), peaking in spring and lowest in summer. GEE-derived ET correlates with field data (0.71), with a slight underestimation.
• Surface runoff shows a decreasing trend from 2013 to 2023, correlating with reduced precipitation, with highest values in the plateau and lowest in the plain.
• Groundwater levels in a representative piezometer dropped by approximately 7 meters between 2008 and 2022, primarily due to excessive water pumping.

Contributions

• Demonstrates the significant potential and feasibility of using open Earth observation data and cloud-based platforms (Google Earth Engine) for effective water resource management in data-scarce regions.
• Provides a spatially explicit and comprehensive understanding of water balance components and water scarcity dynamics at seasonal and interannual scales for the Saïss Plain.
• Overcomes limitations of conventional field-based monitoring (e.g., data gaps, resource constraints) by validating the reliability of satellite-based approaches for precipitation and evapotranspiration.
• Identifies specific spatial hotspots of water scarcity, offering actionable insights for adaptive strategies such as irrigation optimization, aquifer recharge interventions, and land-use planning.
• Proposes feasible mitigation strategies for future water scarcity, including replenishment from the M’dez dam and the reuse of treated wastewater.
• Highlights GEE's role in supporting early warning systems for drought and informing policy reforms towards sustainable water use and climate resilience.

Funding

This research received no external funding.

Citation

@article{Garouani2025Earth,
  author = {Garouani, Manal El and El-Omairi, Mohamed Ali and El-Yazidi, Maryame and Lahrach, Abderrahim and Oulidi, Hassane Jarar},
  title = {Earth observation open data in Google Earth Engine for water resource management in the Saïss Plain, Morocco},
  journal = {Mediterranean Geoscience Reviews},
  year = {2025},
  doi = {10.1007/s42990-025-00204-3},
  url = {https://doi.org/10.1007/s42990-025-00204-3}
}