Achite, Mohammed; Nasiri, Hamid; Katipoğlu, Okan Mert; Abdallah, Mohammed; Moazenzadeh, Roozbeh; Mohammadi, Babak

Reference evapotranspiration (ET0) modeling is pivotal for irrigation scheduling and water resources planning. This study presents a hybrid approach integrating Extreme Gradient Boosting (XGB) with Marine Predators Algorithm (MPA) for daily ET0 estimation in northern Algeria. The proposed XGB-MPA model was evaluated against traditional empirical models and assessed using statistical methods. Shapley Additive Explanations (SHAP) was employed to enhance model interpretability. Various combinations of meteorological variables were tested as inputs, including air temperature, relative humidity, sunshine hours, wind speed, and extraterrestrial solar radiation. The XGB-MPA hybrid model achieved superior prediction accuracy during testing (R2 = 0.9958, RMSE = 0.1713 mm/day) compared to traditional empirical models and the standard XGB model. The study demonstrated that ET0 prediction accuracy increased with the number of meteorological inputs used. Our findings highlight the XGB-MPA hybrid model's potential for accurate ET0 estimation in northern Algeria, which can be used for water resource management and irrigation planning.

Abbaszadeh, Maryam; Bazrafshan, Ommolbanin; Katipoğlu, Okan Mert; Jamshid, Sajad

Understanding the streamflow fluctuation of the Minab River basin in the south of Iran is of great importance due to the presence of the Esteghlal Reservoir Dam at its outlet and agricultural activities that produce food for the residents and support neighboring cities. Water resources of this basin have been affected by climate change and it is expected to observe more effects in the future. Therefore, this study was conducted to evaluate the impact of climate change on river flow levels in the Minab River basin, comparing the future period (2020–2050) with the baseline period (1985–2014). Using the CanESM5 model, future climate scenarios SSP1-2.6, SSP3-7.0, and SSP5-8.5 were simulated to project changes in precipitation and maximum, and minimum temperatures. For streamflow simulation, five machine learning models including Generalized Linear Regression (GLR), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees (DT), and Random Forest (RF) were employed, with KNN showing superior performance. The findings indicate max and min temperatures could rise by 4.9℃ and 5.2℃ respectively under the SSP3-7.0 scenario until 2050. Mean annual streamflow will experience an increase of 33% and the highest values will probably be observed in fall, which can be the trigger for more floods and decreasing streamflow during warmer seasons can cause many challenges in the availability of water resources. These insights are crucial for water resource management by local authorities and organizations.

Öllükçü, Meva Nur; Katipoğlu, Okan Mert

Drought was defined as water scarcity resulting from below-normal precipitation in a specific region over an extended period. Analysis of droughts was of vital importance in terms of water resources planning and management, hydropower generation, irrigation planning, agricultural productivity, and reservoir planning. In this study, drought status was evaluated according to the Standard Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI), and De Martonne Index (DMI) values calculated between 1979 and 2022 for seasonal and annual periods for Amasya, Çorum, Samsun, and Tokat meteorological stations located in the Yeşilırmak Basin, and station-based drought percentages were expressed. In addition, trend and change point analyses were conducted for each drought index according to Şen’s Innovative Trend Analysis (ITA), Sequential Mann-Kendall (SQMK), and Innovative Polygon Trend Analysis (IPTA) methods. The analysis identified the trends and onset years of droughts in the region, revealing that severe droughts were dominant at a 15% level. ITA results indicated a significant decreasing trend of 5–10% in spring, summer, autumn, and annual periods, alongside an increasing drought trend. SQMK analyses confirmed a general decreasing trend at the stations, while IPTA highlighted the dominance of increasing drought intensities in SPI, SPEI, and DMI trends. These findings underscored the need for policymakers to take proactive measures to mitigate drought risks in the Yeşilırmak Basin.

Achite, Mohammed; Katipoglu, Okan Mert; Elbeltagi, Ahmed; Elshaboury, Nehal; Pandey, Kusum; Emami, Somayeh; Yongguo, Chen

Accurate estimation of evaporation from surface water bodies is essential for sustainable water resource planning and operational hydrology. However, direct measurement over reservoirs is complex and often impractical. In this study, an advanced machine learning framework based on Regression by Discretization (RD) was applied to predict daily reservoir evaporation at the Sidi-M’Hamed Ben Aouda Dam Basin in Algeria from 2010 to 2023. Among the models evaluated, the RD-Bagging model demonstrated superior performance with a correlation coefficient (CC) of 0.7849, mean absolute error (MAE) of 0.0139, root mean square error (RMSE) of 0.0172, and percent bias (PBIAS) of -1.42% during the testing period. For the validation period, RD-Bagging yielded CC, MAE, RMSE, and PBIAS values of 0.822, 0.0119, 0.0154, and − 1.18%, respectively, indicating high predictive accuracy and low bias. The results confirm the robustness and generalization ability of the RD-Bagging model, making it a reliable tool for reservoir evaporation prediction. This study provides a foundation for the development of data-driven decision support systems in hydrological modeling and climate resilience planning.

Akiner, Muhammed Ernur; Katipoğlu, Okan Mert; Çintaş, Emre

Urban air quality influences public health, ecosystem sustainability and economic productivity. This study focuses on predicting the Air Quality Index (AQI) in Başakşehir, Istanbul. The study proposes a hybrid artificial intelligence (AI) model that amalgamates Categorical Boosting (CatBoost), Shapley Additive Explanations (SHAP) and the feature importance analysis. The dataset encompasses various meteorological parameters, including Tempmax, Tempmin, Temp, Dew, Humidity, Precip, Windspeed, Sea level pressure, Cloud cover, Solar radiation, Solar energy and UV index, in addition to air quality parameters such as PM10, SO2, CO, NO2, NOX, NO and O3. These variables serve as inputs for models like ANN, BAT-ANN,- BBO-ANN, GWO-ANN, HCA-ANN, CatBoost and CNN; the intent is to enhance the accuracy of the AQI prediction. When the combined set of variables were employed as inputs, the most precise results emerged from the CNN model, which yielded an RMSE of 1.43, an AIC of 949.21 and an NSE and R2 of 0.99. The CatBoost model exhibited exceptional performance among the various input combinations, providing the most accurate results for these configurations. Non-parametric statistical Friedman and Nemenyi post-hoc tests were used for multi-model comparison, and it was concluded that there were significant performance differences between the models used according to the p-value values, both in general and based on pairs. While prior studies have explored hybrid AI models for AQI prediction, this study uniquely integrates CatBoost and SHAP for enhanced explainability and model performance evaluation. SHAP analysis provided transparent insights into variable contributions; however, PM10 emerged as the dominant predictor, achieving the highest mutual information score of 6.88. These findings underscore the importance of integrating pollutant and meteorological data. The proposed methodology aligns with global sustainability goals, including SDG 11 (Sustainable Cities and Communities) and SDG 13 (Climate Action).

Achite, Mohammed; Katipoğlu, Okan Mert; Javari, Majid; Caloiero, Tommaso

Spatial precipitation analysis is essential for effectively managing hydrological modeling, construction of water structures, and irrigation planning. In this study, the ordinary kriging (OK), simple kriging (SK), global polynomial interpolation (GPI), local polynomial interpolation (LPI), inverse distance weighted (IDW), radial basis functions (RBF), and artificial neural network (ANN)-based hybrid techniques were compared to determine the spatial variation of annual precipitation. Statistical indicators derived from Willmott’s index of agreement, root mean square error, mean absolute percentage error, and the violin plot and boxplot graphical approaches were used to determine the most effective technique for precipitation interpolation. According to the analysis results, it has been observed that the ANN model significantly improves the prediction performance of single interpolation methods. The OK-ANN hybrid model was determined to be the most accurate representation of precipitation distribution, with the GPI-ANN model coming in second. The most precise results were obtained using the deterministic method, RBF with inverse multiquadric kernel function, LPI with Epanechnikov kernel function, and GPI with 3rd-order polynomial interpolations. In addition, it was determined that deterministic approaches produce more successful results than geostatistical approaches in the basin due to the presence of homogeneous and densely distributed meteorological observation networks.

Berhail, Sabri; Katipoğlu, Okan Mert

This study explores hydrological and meteorological drought in the Coastal Constantine basin, the Cheffia and Mexa dams in northeast Algeria in particular, using a time series of monthly rainfall and dam inflow data over 54 years (1965–2018). Recently, these dams have experienced recurring droughts, significantly reducing water storage and severely impacting the population’s drinking water supply. To better assess the severity and trends of these droughts, we applied a range of methods (Mann-Kendall test, Sen’s slope estimator, Standardized Precipitation Index (SPI), Standardized Streamflow Index (SSI), and Innovative Trend Analysis (ITA) method). SPI and SSI were chosen because they can be applied across various timescales, easy to calculate, and have demonstrated accuracy. Additionally, the trend analysis techniques were preferred because they have few assumptions, are resistant to incomplete data, offer the advantage of visual interpretation, and have the ability to capture changes within data subgroups. The findings analysis shows an autumn rainfall increase (1.18 mm/year) and a significant spring decrease (-2.23 mm/year). In addition, an annual rainfall trend magnitude ranging from − 5.26 mm/year to 1.71 mm/year. The Cheffia Dam experiences a summer inflow decrease (-0.05 hm³/year) and an overall annual decline (-0.35 hm³/year), while the Mexa Dam shows a significant spring inflow decrease (-0.17 hm³/year) with a slight annual decline (-0.07 hm³/year). The ITA method revealed severe and prolonged droughts with increasing severity over time, as shown by the SPI and SSI drought trend analysis. These trends affirm the growing impact of reduced precipitation on water resources and emphasize the urgent need for proactive water management strategies in the region. The results highlight the extremely difficult conditions and the severe drought, particularly in central parts of the basin, over the past 54 years, and underscore ongoing challenges in managing water resources amid changing climatic conditions. This situation negatively impacts the Boulatane dam project and the under-construction Boukhroufa dam. Thus, it is necessary to implement necessary measures to mitigate the effects of prolonged droughts on water availability and infrastructure development. The study’s findings provide a foundation for spatial and temporal analysis and management of droughts in water-stressed regions globally, in the context of climate change with innovative techniques. The results emphasize the need for policymakers to prepare for droughts and adopt early warning systems. It is recommended that water conservation efforts be strengthened, new water structures be built, and climate change policies be adapted.

Erkol, Z. Ibrahim; Yesilyurt, S. Nur; Dalkilic, H. Yildirim

In this article, the affiliation details for author Z. Ibrahim Erkol was incorrectly given as ‘Environmental Sciences Institute, Boğaziçi University, Istanbul, Turkey’ but should have been ‘Brandenburg Technical University Cottbus-Senftenberg, Environmental Engineering Department, Cottbus, Germany’. As for H. Yildirim Dalkilic, it was incorrectly given as ‘Department of Civil Engineering, Dokuz Eylül University, Izmir, Turkey’ but should have been ‘Erzincan Binali Yıldırım University, Faculty of Engineering and Architecture, Department of Civil Engineering, Erzincan, Turkey’. The original article has been corrected.

Achite, Mohammed; Katipoğlu, Okan Mert; Jehanzaib, Muhammad; Kartal, Veysi; Mansour, Hamidi

Drought is a natural disaster characterised as precipitation much lower than the precipitation reported in actual periods. Many studies characterized drought as meteorological, hydrological, agricultural, or socioeconomic. When there is a long-term shortage of precipitation, deficits in surface and ground waters occur. In this study, a hydrological drought analysis has been performed for Wadi Mina Basin (4900 km2), which is the biggest sub-basin in Cheliff Basin, using the Streamflow Drought Index (SDI) for the time scales of 3, 6, 9, and 12-month. Monthly mean streamflow records for 05 stations are obtained from the National Water Resources Agency. The obtained SDI values were subjected to Run analysis and drought duration and severity values were calculated. According to the analysis, it has been observed that the maximum (duration: 70 months, severity: 92.78) and average (duration: 31 months, severity: 31.28) droughts occurred at the Sidi AEK Djillali station on a 12-month time scale. The average drought severity was 6.34, with a maximum value of 56.71 on a monthly time scale. However, on a 12-month time scale, the average drought severity increased to 31.28, with a maximum value of 92.78. Therefore, it can be said that the drought severity has increased with the increase in the time scale. When the temporal changes of drought indices are evaluated, it is noteworthy that extraordinary droughts prevailed in the basin in 2000 and 2007. When the scatter diagrams of drought characteristics were examined, it was seen that there was a significant linear relationship between drought duration and severity. In addition, the highest correlation was observed at the 9-month time scale at Ain Hamara (R:0.964) and Oued Abtal (R:0.904) stations. In contrast, the highest correlation was observed at the 12-month time scale at Sidi AEK Djillali (R:0.980) and Takhmaret (R:0.969) stations.

Achite, Mohammed; Katipoğlu, Okan Mert; Elshaboury, Nehal; Kartal, Veysi; Aktürk, Gaye; Ertugay, Neşe

Sodium adsorption rate (SAR), which significantly affects soil and plant health, is determined according to the concentration of sodium ions, calcium, and magnesium in irrigation water. Accurate estimation of SAR values is vital for agricultural production and irrigation. In this study, hybrid swarm intelligence-based neural networks are used to model sodium adsorption ratio in irrigation water quality parameters in a semi-arid environment at Sidi M’Hamed Ben Aouda (SMBA) dam, Algeria. For this, the nature-inspired optimization techniques of particle swarm optimization (PSO), genetic algorithm (GA), Jaya algorithm (JA), artificial bee colony (ABC), and firefly algorithm (FFA) and the signal processing technique of variational mode decomposition (VMD) have been combined with artificial neural networks (ANN). Correlation matrices were used to select the data entry structure in the established models. Water quality parameters with a statistically significant and medium to high relationship with SAR values were presented as input to the model. The overall performance was measured using various statistical metrics, scatter diagrams, Taylor diagrams, correlograms, boxplots, and line plots. In addition, the effect of input parameters on model estimation was evaluated according to Sobol sensitivity analysis. As a result, the GA-ANN algorithm demonstrated superior performance (MSE = 0.073, MAE = 0.193, MAPE = 0.048, MBE=-0.16, R2 = 0.934, WI = 0.968, and KGE = 0.866) based on the statistical indicators, indicating better results compared to other models. The second-best model, ABC-ANN (MSE = 0.084, MAE = 0.233, MAPE = 0.066, MBE=-0.135, R2 = 0.897, WI = 0.965, and KGE = 0.920) was also selected. The weakest prediction outputs were obtained from the VMD-ANN model. The accurate and reliable estimation of SAR in irrigation water has the potential to facilitate improvements in agricultural irrigation management and agricultural production efficiency for farmers, agricultural practitioners, and policymakers.