Bounajra et al. (2026) Reference evapotranspiration variability and trends in relation to irrigation demand in a semi-arid agricultural region of Morocco

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2026
• Date: 2026-03-26
• Authors: Afaf Bounajra, Abdelhamid Amahouch, Kamal El Guemmat, Khalifa Mansouri
• DOI: 10.1016/j.ejrh.2026.103377

Research Groups

• Laboratory of Modelling and simulation of intelligent industrial systems (M2S2I), Ecole Normale Sup´erieure de l’Enseignement Technique (ENSET), Mohammedia, Morocco
• LPHE-Modeling and Simulation, Faculty of Sciences, Mohammed V University in Rabat, Morocco
• Laboratory of Informatique, Artificial Intelligence and Cyber Security (2IACS), Ecole Normale Sup´erieure de l’Enseignement Technique (ENSET), Mohammedia, Morocco

Short Summary

This study investigates the spatio-temporal variability and long-term trends of reference evapotranspiration (ET₀) in the semi-arid Chichaoua agricultural province of Morocco over 45 years, explicitly linking ET₀ dynamics to operational drip irrigation design practices. It reveals a significant upward trend in ET₀ driven by thermo-radiative factors, which directly translates into increased irrigation water requirements based on current engineering assumptions.

Objective

• To characterize the spatio-temporal variability and long-term trends of reference evapotranspiration (ET₀) in the Chichaoua province over the period 1981–2025 using ERA5-Land data.
• To identify and rank the relative contributions of key meteorological drivers controlling ET₀ through a combination of linear correlation, mutual information, and permutation-based Random Forest analysis.
• To explicitly link ET₀ variability and trends to operational irrigation design by comparing reanalysis-based estimates with ET₀ values derived from technical calculation notes used for drip irrigation system sizing.

Study Configuration

• Spatial Scale: Chichaoua agricultural province, south-central Morocco, covering five rural municipalities (Saidate, Sidi M′Hamed Dalil, Majjat, Lamzoudia, and Ahdil) and 37 representative agricultural fields. The topography ranges from approximately 200 to 800 meters above sea level.
• Temporal Scale: A 45-year period from 1981 to 2025, analyzed at daily, monthly, annual, and seasonal intervals.

Methodology and Data

• Models used:
  • FAO-56 Penman-Monteith formulation (for ET₀ estimation).
  • Linear correlation (Pearson's coefficient).
  • Mutual Information (k-nearest-neighbors-based nonparametric estimator).
  • Random Forest regression (for permutation importance, ΔRMSE).
  • Mann–Kendall test (nonparametric trend detection).
  • Sen's slope estimator (trend magnitude quantification).
  • Pettitt breakpoint test (abrupt change detection).
  • Benjamini–Hochberg procedure (False Discovery Rate for multiple comparisons).
• Data sources:
  • ERA5-Land reanalysis (ECMWF): hourly air temperature at 2 m, dew point temperature, surface pressure, zonal and meridional wind components at 10 m, and downward solar radiation (approx. 0.1° / 9 km spatial resolution).
  • CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data): daily precipitation (0.05° / 5 km resolution).
  • MERRA-2 (NASA) reanalysis (for intercomparison).
  • GLDAS Noah v2.1 (NASA) reanalysis (for intercomparison).
  • SRTM 30 m Digital Elevation Model (USGS SRTMGL1_003).
  • Technical calculation notes from the Provincial Directorate of Agriculture (PDA) for drip irrigation system design (operational ET₀ values, gross irrigation requirements, cultivated surface areas).
  • Google Earth Engine (GEE) for data extraction and processing.

Main Results

• Spatial Variability: Mean annual ET₀ ranges from approximately 4.0–4.8 mm⋅day⁻¹, with seasonal maximums approaching 6.8 mm⋅day⁻¹ in low-altitude agricultural zones. Spatial patterns are primarily controlled by altitudinal and aerodynamic contrasts (wind speed), with temperature and vapor pressure deficit playing secondary roles, while solar radiation shows low spatial variability.
• Meteorological Drivers: ET₀ variability is predominantly controlled by radiative and thermal factors (solar radiation and air temperature), with secondary contributions from vapor pressure deficit and wind speed. Solar radiation is the most stable and influential factor throughout the year, while wind speed becomes a dominant control during summer.
• Temporal Trends (1981-2025):
  • Annual ET₀ shows a statistically significant upward trend of +28.6 mm per decade (p < 0.001).
  • Significant positive trends were observed for mean air temperature (+0.35 °C per decade), maximum temperature (+0.43 °C per decade), minimum temperature (+0.28 °C per decade), and vapor pressure deficit (+0.059 kPa per decade). Solar radiation showed a marginal increase (+0.12 MJ⋅m⁻²⋅day⁻¹ per decade).
  • Wind speed showed no significant long-term trend. Surface atmospheric pressure increased slightly but significantly (+0.030 kPa per decade).
  • Annual precipitation showed a marginally significant downward trend (−14.3 mm per decade).
  • Seasonal ET₀ increased significantly in winter (+3.1 mm per decade), spring (+13.6 mm per decade), and summer (+8.7 mm per decade), with no significant trend in autumn.
  • Temporal breakpoints around the late 1990s indicate a shift towards higher evaporative demand.
• Irrigation Sensitivity: Gross irrigation requirements are directly and proportionally sensitive to ET₀, as embedded in current irrigation design practices. Peak summer ET₀ values (approximately 7.0 mm⋅day⁻¹) are associated with gross irrigation requirements of ~6.2 mm⋅day⁻¹ for intensive vegetable crops and ~3.8–4.0 mm⋅day⁻¹ for perennial tree crops. This sensitivity is strongly modulated by crop type and cultivated area.

Contributions

• Provides a comprehensive assessment of spatio-temporal ET₀ variability and trends in a data-scarce semi-arid region (Chichaoua, Morocco) using robust reanalysis data and nonparametric statistical methods.
• Explicitly links observed ET₀ variability and long-term trends to operational irrigation design practices by analyzing technical calculation notes for drip irrigation systems, thereby bridging climate science and irrigation engineering.
• Demonstrates how increases in atmospheric evaporative demand are directly incorporated into current irrigation system sizing assumptions, highlighting a potential mismatch between evolving climatic conditions and static design values.
• Quantifies the crop-dependent sensitivity of irrigation demand to ET₀, revealing heterogeneous pressures on irrigation systems at the field and commune scales.
• Offers a transferable framework for interpreting ET₀ trends in terms of irrigation demand and supporting adaptive water management strategies in semi-arid agricultural regions.

Funding

• CMAPS (Moroccan Company of Agriculture & Solar Pumping) as a corporate partner within the 'PhD-Associate Scholarship – PASS' program.

Citation

@article{Bounajra2026Reference,
  author = {Bounajra, Afaf and Amahouch, Abdelhamid and Guemmat, Kamal El and Mansouri, Khalifa},
  title = {Reference evapotranspiration variability and trends in relation to irrigation demand in a semi-arid agricultural region of Morocco},
  journal = {Journal of Hydrology Regional Studies},
  year = {2026},
  doi = {10.1016/j.ejrh.2026.103377},
  url = {https://doi.org/10.1016/j.ejrh.2026.103377}
}