Huertas-Bastidas et al. (2026) Enhancing AquaCrop-OSPy yield predictions with UAV-based remote sensing data: a case study on broccoli

Identification

• Journal: Computers and Electronics in Agriculture
• Year: 2026
• Date: 2026-01-05
• Authors: Jesús Huertas-Bastidas, M.A. Jiménez-Bello, Diego S. Intrigliolo, Ramírez-Cuesta Juan Miguel
• DOI: 10.1016/j.compag.2025.111402

Research Groups

• Research Institute of Water and Environmental Engineering (IIAMA), Universitat Politècnica de València, València, Spain
• Department of Ecology and Global Change, Desertification Research Centre (CIDE-CSIC-UV-GV), Moncada, Spain

Short Summary

This study evaluated how unmanned aerial vehicle (UAV) multispectral and thermal observations improved AquaCrop-OSPy simulations for irrigated broccoli. Assimilating UAV-derived canopy cover (CC) and actual evapotranspiration (ETa) modestly improved yield predictions, reducing the root-mean-square error by up to 11.9%.

Objective

• To evaluate how unmanned aerial vehicle (UAV) multispectral and thermal observations improved AquaCrop-OSPy simulations of canopy cover (CC), actual evapotranspiration (ETa), and yield for irrigated broccoli under Mediterranean conditions.

Study Configuration

• Spatial Scale: 0.2 hectare field
• Temporal Scale: Two growing seasons

Methodology and Data

• Models used: AquaCrop-OSPy (FAO AquaCrop formulation in Python), pyTSEB (two-source energy balance model)
• Data sources: UAV multispectral imagery, UAV thermal imagery, lysimeter measurements (for ETa validation), FAO Penman-Monteith (for reference evapotranspiration driver)

Main Results

• The pyTSEB model reproduced actual evapotranspiration (ETa) with a root-mean-square error (RMSE) of 0.39 mm d⁻¹ and a Nash-Sutcliffe efficiency (NSE) of 0.93 when compared with lysimeter measurements.
• Baseline AquaCrop-OSPy showed an ETa RMSE of 1.24 mm d⁻¹ and an NSE of 0.27.
• Without data assimilation, AquaCrop-OSPy reproduced mean yield but failed to capture subplot variability (RMSE 1.67 tonnes ha⁻¹).
• Assimilating UAV-derived canopy cover (CC) reduced yield RMSE by 8.9%.
• Joint assimilation of CC and ETa achieved the lowest yield RMSE (1.47 tonnes ha⁻¹), representing an 11.9% reduction.
• An Irrigation Advisor (IA) strategy applied 20.6% more water than farmer practice across seasons, without consistent benefits in yield or water productivity.

Contributions

• Demonstrated the utility of high-resolution UAV-based multispectral and thermal remote sensing data for enhancing crop model (AquaCrop-OSPy) predictions of canopy cover, actual evapotranspiration, and yield.
• Evaluated the effectiveness of different data assimilation strategies (CC, ETa, or both) for improving AquaCrop-OSPy yield predictions.
• Provided a case study for irrigated broccoli under Mediterranean conditions, highlighting the potential and limitations of integrating UAV data into crop simulation models.

Funding

• Not explicitly mentioned in the provided text.

Citation

@article{HuertasBastidas2026Enhancing,
  author = {Huertas-Bastidas, Jesús and Jiménez-Bello, M.A. and Intrigliolo, Diego S. and Miguel, Ramírez-Cuesta Juan},
  title = {Enhancing AquaCrop-OSPy yield predictions with UAV-based remote sensing data: a case study on broccoli},
  journal = {Computers and Electronics in Agriculture},
  year = {2026},
  doi = {10.1016/j.compag.2025.111402},
  url = {https://doi.org/10.1016/j.compag.2025.111402}
}