Search

Drought is one of the most severe climatic calamities, affecting many aspects of the environment and human existence. Effective planning and decision making in disaster-prone areas require accurate and reliable drought predictions globally. The selection of an effective forecasting model is still challenging due to the lack of information on model performance, even though data-driven models have been widely employed to anticipate droughts. Therefore, this study investigated the application of simple extreme learning machine (ELM) and wavelet-based ELM (W-ELM) algorithms in drought forecasting. Standardized runoff index was used to model hydrological drought at different timescales (1-, 3-, 6-, 9-, and 12-month) at five Wadi Mina Basin (Algeria) hydrological stations. A partial autocorrelation function was adopted to select lagged input combinations for drought prediction. The results suggested that both algorithms predict hydrological drought well. Still, the performance of W-ELM remained superior at most of the hydrological stations with an average coefficient of determination = 0.74, root mean square error = 0.36, and mean absolute error = 0.43. It was also observed that the performance of the models in predicting drought at the 12-month timescale was higher than at the 1-month timescale. The proposed hybrid approach combined ELM's fast-learning ability and discrete wavelet transform's ability to decompose into different frequency bands, producing promising outputs in hydrological droughts. The findings indicated that the W-ELM model can be used for reliable drought predictions in Algeria. [ABSTRACT FROM AUTHOR]

Climate change may cause severe hydrological droughts, leading to water shortages which will require to be assessed using high-resolution data. Gravity Recovery and Climate Experiment (GRACE) satellite Terrestrial Water Storage (TWSA) estimates offer a promising solution to monitor hydrological drought, but its coarse resolution (1°) limits its applications to small regions of the Indus Basin Irrigation System (IBIS). Here we employed machine learning models such as Extreme Gradient Boosting (XGBoost) and Artificial Neural Network (ANN) to downscale GRACE TWSA from 1° to 0.25°. The findings revealed that the XGBoost model outperformed the ANN model with Nash Sutcliff Efficiency (NSE) (0.99), Pearson correlation (R) (0.99), Root Mean Square Error (RMSE) (5.22 mm), and Mean Absolute Error (MAE) (2.75 mm) between the predicted and GRACE-derived TWSA. Further, Water Storage Deficit Index (WSDI) and WSD (Water Storage Deficit) were used to determine the severity and episodes of droughts, respectively. The results of WSDI exhibited a strong agreement when compared with the Standardized Precipitation Evapotranspiration Index (SPEI) at different time scales (1-, 3-, and 6-months) and self-calibrated Palmer Drought Severity Index (sc-PDSI). Moreover, the IBIS had experienced increasing drought episodes, e.g., eight drought episodes were detected within the years 2010 and 2016 with WSDI of −1.20 and −1.28 and total WSD of −496.99 mm and −734.01 mm, respectively. The Partial Least Square Regression (PLSR) model between WSDI and climatic variables indicated that potential evaporation had the largest influence on drought after precipitation. The findings of this study will be helpful for drought-related decision-making in IBIS. [ABSTRACT FROM AUTHOR]

Evapotranspiration (ET) is a significant aspect of the hydrologic cycle, notably in irrigated agriculture. Direct approaches for estimating reference evapotranspiration (ET0) are either difficult or need a large number of inputs that are not always available from meteorological stations. Over a 6-year period (2006–2011), this study compares Feed Forward Neural Network (FFNN), Radial Basis Function Neural Network (RBFNN), and Gene Expression Programming (GEP) machine learning approaches for estimating daily ET0 in a meteorological station in the Lower Cheliff Plain, northwest Algeria. ET0 was estimated using the FAO-56 Penman–Monteith (FAO56PM) equation and observed meteorological data. The estimated ET0 using FAO56PM was then used as the target output for the machine learning models, while the observed meteorological data were used as the model inputs. Based on the coefficient of determination (R2), root mean square error (RMSE), and Nash–Sutcliffe efficiency (EF), the RBFNN and GEP models showed promising performance. However, the FFNN model performed the best during training (R2 = 0.9903, RMSE = 0.2332, and EF = 0.9902) and testing (R2 = 0.9921, RMSE = 0.2342, and EF = 0.9902) phases in forecasting the Penman–Monteith evapotranspiration. [ABSTRACT FROM AUTHOR]

Drought is among the most hazardous climatic disasters that significantly influence various aspects of the environment and human life. Qualitative and reliable drought forecasting is important worldwide for effective planning and decision-making in disaster-prone regions. Data-driven models have been extensively used for drought forecasting, but due to the inadequacy of information on model performance, the selection of an appropriate forecasting model remains a challenge. This study concerns a comparative analysis of six machine learning (ML) techniques widely used for hydrological drought forecasting. The standardized runoff index (SRI) was calculated at a seasonal (3-month) time scale for the period 1973 to 2016 in four selected watersheds of the Han River basin in South Korea. The ML models employed were built-ins, using precipitation, temperature, and humidity as input variables and the SRI as the target variable. The results indicated that all the ML models were able to map the relationship for seasonal SRI using the applied input vectors. The decision tree (DT) technique outperformed in all the watersheds with an average mean absolute error (MAE)=0.26 , root mean square error (RMSE)=0.34 , Nash-Sutcliffe efficiency (NSE)=0.87 , and coefficient of determination (R2)=0.89. The performances of the support vector machine (SVM) and deep learning neural network (DLNN) were similar, whereas the fuzzy rule-based system (FRBS) performed very well compared to the artificial neural network (ANN) and the adaptive neuro-fuzzy inference system (ANFIS). The overall findings of this study indicate that, considering performance criteria and computation time, the DT was the most accurate ML technique for hydrological drought forecasting in the Han River basin. [ABSTRACT FROM AUTHOR]

• Developed computationally efficient three new variants of the CN model. • Introduced flexible initial abstraction to avoid sudden jumps in runoff estimation. • New weighted score-based statistical indicators for the models' evaluation. • The proposed models outperformed in almost all watersheds. The globally adopted and computationally efficient curve number (CN) model fills an active hydrological professional niche and has a well-documented history. However, it is structurally inconsistent and fails to reliably estimate runoff from rainfall. This is mainly due to the much-debated fixed initial abstraction (λ) and associated sudden jumps in runoff based on CN obtained from the documented tables. In this study, three new variants (M4, M5, M6) of the CN model are proposed that consider the hydrological imbalance between pre-storm soil moisture and initial abstraction after a rainfall event. A total of 1837 rainfall-runoff events were analyzed from 41 steep-slope watersheds in South Korea to test the robustness of the proposed CN models. The results were compared to the recently updated original CN model (M1) and two other recent variants (M2, M3) of this model. These models were evaluated using root mean squared error (RMSE), Nash-Sutcliffe efficiency (NSE), percent bias (PB), Kling-Gupta efficiency (KGE), a proposed overall weighted score, 1:1 line graph, and the FITEVAL tool. Using data from 41 watersheds, the lowest mean (median) RMSE of M5, M6, M4 [15.54(16.56), 15.84(11.45), 18.00(16.56)], PB for M4, M6, M5 [−1.28(−1.03), 1.29(0.03), −7.06(−7.27)]; the highest mean (median) NSE for M5, M6, M4 [0.86(0.87), 0.85(0.87), 0.81(0.83)], KGE for M6, M5, M4 [0.81(0.84), 0.79(0.81), 0.79(0.81)] and other graphical assessments show a better agreement between the observed and the runoff estimated by the proposed models. The corresponding mean (median) of RMSE, PB, NSE, and KGE statistics for M2 [19.26(17.69), 9.50(10.34), 0.78(0.80), 0.77(0.79)], and M3 [19.99(18.47), 05.56(7.17), 0.76(0.79), 0.77(0.79)] models show comparatively inferior results. Based on the same statistics, the M1 [24.60(22.48), 33.62(32.77), 0.63(0.68), 0.60(0.64)] model yields unrealistic results. It is inferred that both λ and CN should be kept flexible for a systematic and region-specific revision of the CN model to improve runoff estimation. In addition, a structurally consistent model with a stable soil moisture accounting (SMA) procedure is vital to get more reliable runoff estimates without compromising the simplicity and applicability to ungauged watersheds. [ABSTRACT FROM AUTHOR]

Drought has negative impacts on water resources, food security, soil degradation, desertification and agricultural productivity. The meteorological and hydrological droughts prediction using standardized precipitation/runoff indices (SPI/SRI) is crucial for effective water resource management. In this study, we suggest ANFISWCA, an adaptive neuro-fuzzy inference system (ANFIS) optimized by the water cycle algorithm (WCA), for hydrological drought forecasting in semi-arid regions of Algeria. The new model was used to predict SRI at 3-, 6-, 9-, and 12-month accumulation periods in the Wadi Mina basin, Algeria. The results of the model were assessed using four criteria; determination coefficient, mean absolute error, variance accounted for, and root mean square error, and compared with those of the standalone ANFIS model. The findings suggested that throughout the testing phase at all the sub-basins, the proposed hybrid model outperformed the conventional model for estimating drought. This study indicated that the WCA algorithm enhanced the ANFIS model's drought forecasting accuracy. The proposed model could be employed for forecasting drought at multi-timescales, deciding on remedial strategies for dealing with drought at study stations, and aiding in sustainable water resources management. • Adaptive neuro-fuzzy inference system optimized by the water cycle algorithm (ANFISWCA) for drought prediction. • Meteorological and hydrological droughts (SPI/SRI) were predicted by ANFISWCA in five sub-basins. • Proposed ANFISWCA was compared with ordinary ANFIS. • SPI and SRI were monitored and predicted on various time scales. [ABSTRACT FROM AUTHOR]