Ghazouani et al. (2025) Assessment of AquaCrop Inputs from ERA5-Land and Sentinel-2 for Soil Water Content Estimation and Durum Wheat Yield Prediction: A Case Study in a Tunisian Field

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

Identification

• Journal: Water
• Year: 2025
• Date: 2025-12-12
• Authors: Hiba Ghazouani, Dario De Caro, Matteo Ippolito, Fulvio Capodici, Giuseppe Ciraolo
• DOI: 10.3390/w17243522

Research Groups

Not explicitly mentioned in the provided text, but the research implies an agricultural research institution or university department in Tunisia.

Short Summary

This study compares AquaCrop model performance using various input combinations, including ERA5-Land reanalysis and Sentinel-2 derived crop cover, demonstrating its feasibility for durum wheat yield estimation in data-scarce Mediterranean regions.

Objective

• To compare the performance of the AquaCrop model using four different input combinations, specifically evaluating the effectiveness of using ERA5-Land meteorological data and satellite-derived crop cover in place of continuous field-based observations.

Study Configuration

• Spatial Scale: Field experiment in northwest Tunisia.
• Temporal Scale: One growing season (2024–2025).

Methodology and Data

• Models used: AquaCrop
• Data sources:
  • In situ meteorological variables (temperature, precipitation, reference evapotranspiration (ETo)).
  • ERA5-Land (E5L) reanalysis for meteorological variables.
  • Measured green Canopy Cover (gCC).
  • Sentinel-2 NDVI-derived gCC.
  • In situ Soil Water Content (SWC).
  • In situ final yields (biological and grain).

Main Results

• ERA5-Land reproduced temperature with a Root Mean Square Error (RMSE) less than 2.4 °C and a Nash-Sutcliffe Efficiency (NSE) greater than 0.72.
• ERA5-Land reproduced ETo with an RMSE of 0.57 mm d⁻¹ and an NSE of 0.79.
• ERA5-Land precipitation showed larger discrepancies with an RMSE of 4.14 mm d⁻¹ and an NSE of 0.58.
• Sentinel-2 effectively captured gCC dynamics with an RMSE of 15.65% and an NSE of 0.73, improving AquaCrop performance (RMSE = 5.29%, NSE = 0.93).
• AquaCrop, across all input combinations, reproduced yields within acceptable deviations.
• Simulated biological yield ranged from 9.7 to 11.0 t ha⁻¹ compared to the observed 10.3 t ha⁻¹.
• Simulated grain yield ranged from 3.0 to 3.5 t ha⁻¹ against the observed 3.3 t ha⁻¹.
• The use of ERA5-Land meteorological data coupled with Sentinel-2 NDVI-derived gCC produced realistic yield estimates, highlighting its feasibility for crop monitoring in data-scarce environments.

Contributions

• Demonstrates the practical application and reliability of the AquaCrop model for durum wheat yield assessment in Mediterranean regions using exclusively remote sensing (Sentinel-2) and reanalysis (ERA5-Land) data, addressing the challenge of limited in situ observations.
• Provides a viable alternative for crop management decision-making in data-scarce agricultural areas under climate change.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Ghazouani2025Assessment,
  author = {Ghazouani, Hiba and Caro, Dario De and Ippolito, Matteo and Capodici, Fulvio and Ciraolo, Giuseppe},
  title = {Assessment of AquaCrop Inputs from ERA5-Land and Sentinel-2 for Soil Water Content Estimation and Durum Wheat Yield Prediction: A Case Study in a Tunisian Field},
  journal = {Water},
  year = {2025},
  doi = {10.3390/w17243522},
  url = {https://doi.org/10.3390/w17243522}
}