Accurate prediction of daily streamflow plays an essential role in various applications of water resources engineering, such as flood mitigation and urban and agricultural planning. This study investigated a hybrid ensemble decomposition technique based on ensemble empirical mode decomposition (EEMD) and variational mode decomposition (VMD) with gene expression programming (GEP) and random forest regression (RFR) algorithms for daily streamflow simulation across three mountainous stations, Siira, Bilghan, and Gachsar, in Karaj, Iran. To determine the appropriate corresponding input variables with optimal lag time the partial auto-correlation function (PACF) and auto-correlation function (ACF) were used for streamflow prediction purpose. Calibration and validation datasets were separately decomposed by EEMD that eventually improved standalone predictive models. Further, the component of highest pass (IMF1) was decomposed by the VMD approach to breakdown the distinctive characteristic of the variables. Results suggested that the EEMD-VMD algorithm significantly enhanced model calibration. Moreover, the EEMD-VMD-RFR algorithm as a hybrid ensemble model outperformed better than other techniques (EEMD-VMD-GEP, RFR and GEP) for daily streamflow prediction of the selected gauging stations. Overall, the proposed methodology indicated the superiority of hybrid ensemble models compare to standalone in predicting streamflow time series particularly in case of high fluctuations and different patterns in datasets.

This study presents the first attempt to link the multi-algorithm genetically adaptive search method (AMALGAM) proposed by Vrugt and Robinson (2007) with a groundwater model to define pumping rates within a well distributed set of Pareto solutions. The pumping rates along with three minimization objectives, i.e. minimizing shortage affected by the failure to supply, modified shortage index and minimization of extent of drawdown within prespecified regions, were chosen to define an optimal solution for groundwater drawdown and subsidence. Hydraulic conductivity, specific yield parameters of a modular three-dimensional finite-difference (MODFLOW) groundwater model were first optimized using cuckoo optimization algorithm by minimizing the sum of absolute deviation between the observed and simulated water table depths. These parameters were then applied in AMALGAM to optimize the pumping rate variables for an arid groundwater system in Iran. The Pareto parameter sets yielded satisfactory results when maximum and minimum drawdowns of the aquifer were defined in a range of −40 to +40 cm/year. Overall, ‘Modelling – Optimization – Simulation’ procedure was capable to compute a set of optimal solutions displayed on a Pareto front. The proposed optimal solution provides sustainable groundwater management alternatives to decision makers in arid region.

Floods are among the most destructive natural hazards in the world, posing numerous risks to societies and economies globally. Accurately understanding and modeling floods driven by extreme rainfall events has long been a challenging task in the domains of hydrologic science and engineering. Unusual catchment responses to flooding cause great difficulty in predicting the variability and magnitude of floods, as well as proposing solutions to manage large volumes of overland flow. The usage of nature-based solutions (NBSs) has proved to be effective in the mitigation of flood peak rate and volume in urban or coastal areas, yet it is still not widely implemented due to limited knowledge and testing compared to traditional engineering solutions. This research examined an integrated hydrological and hydraulic modeling system to understand the response of an at-risk watershed system to flooding and evaluate the efficacy of NBS measures. Using the Hydrologic Engineering Center Hydrologic Modeling System and River Analysis System (HEC-HMS and HEC-RAS) software, an integrated hydrologic–hydraulic model was developed for Hurricane Matthew- (2016) and Florence-driven (2018) floods across the Little Pee Dee–Lumber River watershed, North and South Carolina (the Carolinas), US. The focus was on Nichols, a small town that has disproportionately been impacted by flooding during these two hurricane events. The present article proposes a methodology for selecting, modeling, and evaluating the performance of NBS measures within a catchment, which can be extended to other case studies. Different NBS measures, including flood storage ponds, riparian reforestation, and afforestation in croplands, were designed, modeled, and evaluated. Hurricane Matthew's flooding event was used for evaluating the NBS scenarios given its high simulation accuracy in flood inundation compared to the less accurate results obtained for Hurricane Florence. The scenario comparison evidenced that large-scale natural interventions, such as afforestation in croplands, can reduce the inundated area in Nichols by 8 % to 18 %. On the contrary, the smaller-scale interventions such as riparian reforestation and flood storage ponds showed a negligible effect of only 1 % on flood mitigation.

The use of automated methods for detecting and classifying different types of labels in flood images have important applications in hydrologic prediction. In this research, we propose a fully automated end‐to‐end image detection system to predict flood stage data using deep neural networks across two US Geological Survey (USGS) gauging stations, that is, the Columbus and the Sweetwater Creek, Georgia, USA. The images were driven from the USGS live river web cameras, which were strategically located nearby the monitoring stations and refreshed roughly every 30 s. To estimate the flood stage, a U‐Net Convolutional Neural Network (U‐Net CNN) was first stacked on top of a segmentation model for noise and feature reduction that diminished the number of images needed for training. A Long Short‐Term Memory (LSTM), a dense model, and a CNN were then trained to predict the flood stage time series data in near real‐time (6, 12, 24, and 48 hr). The results revealed that the U‐Net CNN has a higher accuracy for image segmentation if the algorithm is stacked in front of the network. The absolute error with the U‐Net was 0.0654 feet at the Columbus while it was 0.0035 feet at the Sweetwater Creek, which were practically low for flood stage estimation. For time series prediction, among three models, the LSTM predicted the flood stage values more accurately during both historical (2015–2022) as well as real‐time forecasts, particularly for 24 and 48 hr timescales. We extensively evaluated the proposed flood stage prediction system against current state‐of‐the‐art methodologies partly crowd‐sourced and mined in real‐time.