HIGHLIGHTS. Discussing the spatiotemporal variations of drought using in situ observations and satellite-based datasets.Applying the hybrid MODWT–EEMD–K-means m

Sediment transportation and accurate estimation of its rate is a significant issue for river engineers and researchers. In this study, the capability of kernel based approaches including Kernel Extreme Learning Machine (KELM) and Gaussian Process Regression (GPR) was assessed for predicting the river daily Suspended Sediment Discharge (SSD). For this aim, the Mississippi River, with three consecutive hydrometric stations, was selected as the case study. Based on the sediment and flow characteristics during the period of 2005–2008, several models were developed and tested under two scenarios (i.e. modeling based on each station's own data or the previous stations' data). Two post-processing techniques, namely Wavelet Transform (WT) and Ensemble Empirical Mode Decomposition (EEMD), were used for enhancing the SSD modeling capability. Also, data post-proceeding was done using Simple Linear Averaging (SLAM) and Nonlinear Kernel Extreme Learning Machine Ensemble (NKELME) methods. Obtained results indicated that the integrated models resulted in more accurate outcomes. Data processing enhanced the models' capability up to 35%. It was found that SSD modeling based on the station's own data led to better results; however, using the integrated approaches, the previous station's data could be applied successfully for the SSD modeling when a station's own data were not available.

Due to the drought negative impacts, accurate forecasting of drought indices is important. This study focused on the short- to long-term Standardized Precipitation Index (SPI) forecasting in sites with different climates using newly integrated hybrid pre-post-processing techniques. Four sites in Iran's northwest were selected and the SPIs series with time scales of 3, 9, and 24 months were forecasted during the period of 1978–2017. For improving the modeling efficiency, wavelet transform and ensemble empirical mode decomposition (EEMD) pre-processing methods were used. In this regard, temporal features of the SPIs series were decomposed via wavelet transform (WT), then, the obtained sub-series were further broken down into intrinsic mode functions using EEMD. Also, simple linear averaging and nonlinear neural ensemble post-processing methods were applied to ensemble the outputs of hybrid models. The results showed that data pre-processing enhanced the models' capability up to 40%. Also, integrated pre-post-processing models improved the models' efficiency by approximately 50%. The root mean square errors' criteria distribution range decreased from 0.337–1.03 (in raw data) to 0.195–0.714 (in decomposed data). The results proved the capability of applied methods in modeling the SPIs series. In increasing the models' accuracy, data pre-processing was more effective than data post-processing.