In the present paper, we propose a new approach for monthly streamflow prediction based on the extreme learning machine (ELM) and the metaheuristic Bat algorithm (BAT-ELM). The performance of the BAT-ELM was compared to that of ELM, support vector regression (SVR), Gaussian process regression (GPR), multilayer perceptron neural network (MLPNN), and the generalized regression neural network (GRNN). The proposed models were applied using data from three hydrometric stations located in the Cheliff Basin, Algeria. It was shown that the BAT-ELM was more satisfactory than the standalone models. The numerical results showed that the BAT-ELM achieved the highest numerical performance with correlation coefficient and Nash-Sutcliffe efficiency ranging from 0.927 to 0.973 and from 0.846 to 0.944, respectively, much higher than the values obtained using the MLPNN, GRNN, SVR, GPR and ELM approaches, respectively. The obtained results demonstrated that the BAT-ELM presented an interesting alternative algorithm for predicting high extreme streamflow data.
This study aimed to improve daily streamflow forecasting by combining machine learning (ML) models with signal decomposition techniques. Four ML models were hybridized with five families of maximum overlap discrete wavelet transforms (MODWTs). The hybrid models were applied to predict daily streamflow at the Bir Ouled Taher station in northern Algeria. Model performance was evaluated using multiple statistical metrics and compared to standalone ML models. The hybrid MODWT-Gaussian process regression (GPR) model using Symlet wavelets (MODWT-GPR3 sym4) achieved the best performance, with R = 0.99 and NSE = 0.98 during validation. This significantly outperformed the standalone models tested and other hybrid combinations. The MODWT-GPR3 sym4 model demonstrated a superior ability to capture nonlinearities and predict peak flows. Hybridization of ML models with wavelet transforms, particularly the MODWT-GPR approach, can substantially improve daily streamflow prediction accuracy compared to standalone models. However, model performance may vary between watersheds due to differences in hydrological characteristics. Consideration of catchment concentration time when selecting model inputs could further enhance forecasting capabilities.
28 March 2024 in Acque Sotterranee - Italian Journal of Groundwater
Water scarcity is a big issue in arid and semi-arid regions. This challenge is particularly evident in the Upper Cheliff plain in Algeria, where the alluvial aquifer plays a vital role in drinking water supply and supporting irrigation. This aquifer faces high demand and quality issues. A study was conducted in this context, employing a cartographic approach to assess potential groundwater recharge from precipitation into the alluvial aquifer. The current study aimed at mapping zones with potential natural groundwater recharge zones by applying the Analytic Hierarchy Process (AHP) integrated within a Geographic Information System (GIS) environment, combining various factors that can influence recharge, such as rainfall, surface soil type, slope degree, land use and land cover, unsaturated zone, groundwater depth, and curve number. The map resulting from the analysis indicates that only 22% of the assessed area covers zones with very low and low potential recharge, 35% with moderate potential recharge zones, and 43% with high and very high potential recharge zones. This map reveals that the eastern region of the plain, from the cities of Djendel to Ain Soltane, is moderately to highly favorable for recharge. This is due to the natural recharge from rainfall and watercourse infiltration during dam release periods, excess irrigation water, and recharge from the Miocene sandstone aquifer in areas with direct aquifer contact. A validation process was performed using data from 66 wells distributed in this plain and it indicated that 48 wells exhibited good agreement with the resulting map, while 18 wells showed slight deviations. The results indicate an agreement of 72.72% between expected and exist number value of wells which confirming the good prediction of the AHP technique.