EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Forecasting Evaporation Trends Amid Climate Change for Sustainable Water Management in Semi-Arid Regions

Ban Al-Hasani, Mawada Abdellatif, Iacopo Carnacina, Clare Harris, Ayad M. Fadhil Al-Quraishi, Muthanna M. A. Al-Shammari

Water · 01.04.2025 · 17 (7) · 1039

DOI: 10.3390/w17071039 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Water, online erschienen am 01.04.2025. Verfasst von Ban Al-Hasani und 5 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Evapotranspiration plays a vital role in the design of irrigation systems, water resource management, and hydrological modeling, especially in arid and semi-arid regions. This study focuses on projecting evaporation rates using three machine learning models: a Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Gaussian Process Regression (GPR), in combination with Principal Component Analysis (PCA) for dimensionality reduction. Meteorological data from 1980 to 2022, including the minimum and maximum temperatures, rainfall, and solar radiation, were used to train and test the models. Projections were made for Kirkuk Governorate by downscaling five global climate models under two climate scenarios: SSP2-4.5 and SSP5-8.5. These scenarios were used to predict future evaporation rates at a rainwater harvesting site for four future periods (P1, P2, P3, and P4) and compare them to the historical reference period (RP). The performance of the models was evaluated using three statistical metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the regression coefficient (R2). Among the models, the MLP demonstrated superior predictive accuracy, with values of MAE = 0.02 mm, RMSE = 0.10 mm, and R2 = 0.95. The SVM model showed a slightly lower performance, with MAE = 0.21 mm, RMSE = 0.13 mm, and R2 = 0.92. The GPR model’s performance was comparable, yielding MAE = 0.22 mm, RMSE = 0.37 mm, and R2 = 0.91. The historical reference period (RP) showed an average evaporation rate of 1370.9 mm per year. Under the SSP2-4.5 scenario, evaporation is projected to increase by 57.2%, while under SSP5-8.5, the increase is projected to be 85.9%. Under the SSP2-4.5 scenario, the evaporation rate for period P1 (2031–2050) showed a slight increase of 1.61%, while for periods P2 (2051–2070) and P3 (2071–2090), the increases were smaller, at 1.89% and 1.93%, respectively. The highest increase occurred in P4 (2091–2100), with a rate of 2.68%, compared to an observed value increase of 1.33%. These findings suggest that climate change will significantly elevate evaporation rates in the region, emphasizing the need for adaptive water resource management strategies.

Zitieren

APA

Al-Hasani, B., Abdellatif, M., Carnacina, I., Harris, C., Al-Quraishi, A. M. F., & Al-Shammari, M. M. A. (2025). Forecasting Evaporation Trends Amid Climate Change for Sustainable Water Management in Semi-Arid Regions. Water, 17(7), 1039. https://doi.org/10.3390/w17071039

Vancouver

Al-Hasani B, Abdellatif M, Carnacina I, Harris C, Al-Quraishi AMF, Al-Shammari MMA. Forecasting Evaporation Trends Amid Climate Change for Sustainable Water Management in Semi-Arid Regions. Water. 2025;17(7):1039. doi:10.3390/w17071039

BibTeX
@article{AlHasani2025,
  author = {Al-Hasani, Ban and Abdellatif, Mawada and Carnacina, Iacopo and Harris, Clare and Al-Quraishi, Ayad M. Fadhil and Al-Shammari, Muthanna M. A.},
  title = {Forecasting Evaporation Trends Amid Climate Change for Sustainable Water Management in Semi-Arid Regions},
  journal = {Water},
  year = {2025},
  volume = {17},
  number = {7},
  pages = {1039},
  doi = {10.3390/w17071039}
}
Kostenlos starten →