article · Discover Applied Sciences
Streamflow forecasting is essential for water resource planning, particularly across semi-arid zones facing rising water demand and severe drought. An assessment in the mountainous Rheraya sub-basin of Morocco tested daily streamflow prediction between 2003 and 2016 using support vector regression, random forest, and multiple linear regression models. Support vector regression achieved the strongest overall performance, recording the highest Nash-Sutcliffe efficiency score of 0.59, followed by random forest and multiple linear regression. However, the limited duration of the available observational time series constrained the models from capturing the complete range of streamflow variability, which degraded predictive accuracy outside the calibration conditions. When calibrated on sufficiently representative historical data, machine learning methods, especially support vector regression, demonstrate clear potential to enhance daily streamflow estimates and aid water allocation strategies in data-scarce, arid catchments.
Reliable streamflow forecasting is critical for managing drinking water, irrigation, and reservoirs in drought-prone regions facing growing demand. In data-scarce mountainous basins, traditional hydrologic models can struggle. Demonstrating that machine learning techniques such as support vector regression can deliver usable streamflow estimates provides water authorities with practical tools to improve resource planning despite constrained hydrological monitoring networks.
This research demonstrates an applied modelling approach that could inform software tools for water resource managers, basin agencies, and agricultural authorities operating in water-stressed, semi-arid catchments. The work remains at an applied research stage, having been tested on historical data in a single sub-basin. Operational adoption requires integration into hydrological monitoring workflows and retraining on longer, more representative time series to handle hydrological variability effectively.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Streamflow prediction is a key variable for water resources management. It becomes more important in semi-arid regions such as the Tensift river basin in Morocco, where water resources are facing a severe drought and the demand is continuously increasing. The present analysis focuses on evaluating Machine Learning techniques, namely support vector regression (SVR) and Random Forest (RF) against the multiple linear regression (MLR) for daily streamflow forecasting in the mountainous sub-basin of Rheraya between 2003 and 2016. The results show that SVR performed best, followed by RF and MLR. In measurable terms and regarding mean performance, SVR exhibited the higher Nash–Sutcliffe efficiency score (NSE = 0.59) and a lower root mean squared error (RMSE = 1.18 $$\text {m}^3\,\text {s}^{-1}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msup><mml:mtext>m</mml:mtext><mml:mn>3</mml:mn></mml:msup><mml:mspace/><mml:msup><mml:mtext>s</mml:mtext><mml:mrow><mml:mo>-</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:msup></mml:mrow></mml:math> ) compared to RF (NSE = 0.53, RMSE = 1.18 $$\text {m}^3\,\text {s}^{-1}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msup><mml:mtext>m</mml:mtext><mml:mn>3</mml:mn></mml:msup><mml:mspace/><mml:msup><mml:mtext>s</mml:mtext><mml:mrow><mml:mo>-</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:msup></mml:mrow></mml:math> ) and MLR (NSE = 0.54, RMSE = 1.01 $$\text {m}^3\,\text {s}^{-1}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:msup><mml:mtext>m</mml:mtext><mml:mn>3</mml:mn></mml:msup><mml:mspace/><mml:msup><mml:mtext>s</mml:mtext><mml:mrow><mml:mo>-</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:msup></mml:mrow></mml:math> ). Furthermore,the available time series was too short to properly capture the full range of streamflow variability, which reduced the prediction performance outside of the calibration conditions. These findings suggest that ML algorithms, particularly SVR, can provide accurate streamflow estimation useful for water resources management when trained on a representative period. The results highlight the capacity of Machine Learning algorithms, specifically SVR, to augment streamflow prediction for enhanced water resource management in arid regions.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1007/s42452-024-05994-z
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
