Support Vector Regression for Rainfall-Runoff Modeling in Urban Drainage: A Comparison with the EPA’s Storm Water Management Model

Coupling machine learning with signal process techniques and particle swarm optimization for forecasting flood routing calculations in the Eastern Black Sea Basin, Türkiye

With the effect of global warming, the frequency of floods, one of the most important natural disasters, increases, and this increases the damage it causes to people and the environment. Flood routing models play an important role in predicting floods so that all necessary precautions are taken before floods reach the region, loss of life and property in the region is prevented, and agricultural lands are protected. This research aims to compare the performance of hybrid machine learning models such as least-squares support vector machine technique hybridized with particle swarm optimization, empirical mode decomposition, variational mode decomposition, and discrete wavelet transform processes for flood routing estimation models in Ordu, Eastern Black Sea Basin, Türkiye. In addition, it is aimed to examine the effect of data division in flood forecasting. Accordingly, 70%, 80%, and 90% of the data were used for training, respectively. For this purpose, the flood data of 2009 and 2013 in Ordu were used. The performance of the established models was evaluated with the help of statistical indicators such as mean bias error, mean absolute percentage error, determination coefficient, Nash-Sutcliffe efficiency, Taylor Diagrams, and boxplot. As a result of the study, the particle swarm optimization least-squares support vector machine technique was chosen as the most successful model in predicting flood routing results. In addition, the optimum data partition ratio was found to be Train:70:Test:30 in the flood routing calculation. The findings are essential regarding flood management and taking necessary precautions before the flood occurs.

Comparison of Rainfall-Runoff Simulation between Support Vector Regression and HEC-HMS for a Rural Watershed in Taiwan

To better understand the effect and constraint of different data lengths on the data-driven model training for the rainfall-runoff simulation, the support vector regression (SVR) approach was applied to the data-driven model as the core algorithm in the present study. Various features selection strategies and different data lengths were employed in the training phase of the model. The validated results of the SVR were compared with the rainfall-runoff simulation derived from a physically based hydrologic model, the Hydrologic Modeling System (HEC-HMS). The HEC-HMS was considered a conventional approach and was also calibrated with a dataset period identical to the SVR. Our results showed that the SVR and HEC-HMS models could be adopted for short and long periods of rainfall-runoff simulation. However, the SVR model estimated the rainfall-runoff relationship reasonably well even if the observational data of one year or one typhoon event was used. In contrast, the HEC-HMS model needed more parameter optimization and inference processes to achieve the same performance level as the SVR model. Overall, the SVR model was superior to the HEC-HMS model in the performance of the rainfall-runoff simulation.

Anomaly Detection in Dam Behaviour with Machine Learning Classification Models

Dam safety assessment is typically made by comparison between the outcome of some predictive model and measured monitoring data. This is done separately for each response variable, and the results are later interpreted before decision making. In this work, three approaches based on machine learning classifiers are evaluated for the joint analysis of a set of monitoring variables: multi-class, two-class and one-class classification. Support vector machines are applied to all prediction tasks, and random forest is also used for multi-class and two-class. The results show high accuracy for multi-class classification, although the approach has limitations for practical use. The performance in two-class classification is strongly dependent on the features of the anomalies to detect and their similarity to those used for model fitting. The one-class classification model based on support vector machines showed high prediction accuracy, while avoiding the need for correctly selecting and modelling the potential anomalies. A criterion for anomaly detection based on model predictions is defined, which results in a decrease in the misclassification rate. The possibilities and limitations of all three approaches for practical use are discussed.

Flood Stage Forecasting Using Machine-Learning Methods: A Case Study on the Parma River (Italy)

Real-time river flood forecasting models can be useful for issuing flood alerts and reducing or preventing inundations. To this end, machine-learning (ML) methods are becoming increasingly popular thanks to their low computational requirements and to their reliance on observed data only. This work aimed to evaluate the ML models’ capability of predicting flood stages at a critical gauge station, using mainly upstream stage observations, though downstream levels should also be included to consider backwater, if present. The case study selected for this analysis was the lower stretch of the Parma River (Italy), and the forecast horizon was extended up to 9 h. The performances of three ML algorithms, namely Support Vector Regression (SVR), MultiLayer Perceptron (MLP), and Long Short-term Memory (LSTM), were compared herein in terms of accuracy and computational time. Up to 6 h ahead, all models provided sufficiently accurate predictions for practical purposes (e.g., Root Mean Square Error < 15 cm, and Nash-Sutcliffe Efficiency coefficient > 0.99), while peak levels were poorly predicted for longer lead times. Moreover, the results suggest that the LSTM model, despite requiring the longest training time, is the most robust and accurate in predicting peak values, and it should be preferred for setting up an operational forecasting system.

Development and evaluation of an innovative Enhanced River Pollution Index model for holistic monitoring and management of river water quality

The present study was conceptualized to develop the Enhanced River Pollution Index (ERPI) model. The ERPI model was used to evaluate the river water quality (RWQ) for its beneficial usage, i.e., drinking with (DCD) and without (DD) conventional treatment, outdoor-bathing (OB), wildlife and fisheries (WF), and industrial and irrigation (IIW). The adequacy of multiple linear regression (MLR) and support vector regression (SVR) models was also investigated to predict the ERPI for estimating the RWQ. The accuracy of the MLR and SVR models was tested by using the statistical parameters, i.e., root mean squared error (RMSE), coefficient of determination (R²), and mean absolute error (MAE). The results revealed that the MLR models performed well (RMSE = 0.004 ± 0.0043, R² = 0.998 ± 0.001, and MAE = 0.002 ± 0.003) for the DD, DCD, and OB. However, the SVR models estimated the RWQ more accurately (RMSE = 0.041 ± 0.001, R² = 0.962 ± 0.010, and MAE = 0.026 ± 0.002) than the MLR models for WF and IIW. Moreover, this study disclosed that the RWQ was not excellent for DD, OB, and DCD. However, the RWQ was categorized from excellent to poor classes for WF, while it was suitable for IIW.

Forecasting of Extreme Storm Tide Events Using NARX Neural Network-Based Models

The extreme values of high tides are generally caused by a combination of astronomical and meteorological causes, as well as by the conformation of the sea basin. One place where the extreme values of the tide have a considerable practical interest is the city of Venice. The MOSE (MOdulo Sperimentale Elettromeccanico) system was created to protect Venice from flooding caused by the highest tides. Proper operation of the protection system requires an adequate forecast model of the highest tides, which is able to provide reliable forecasts even some days in advance. Nonlinear Autoregressive Exogenous (NARX) neural networks are particularly effective in predicting time series of hydrological quantities. In this work, the effectiveness of two distinct NARX-based models was demonstrated in predicting the extreme values of high tides in Venice. The first model requires as input values the astronomical tide, barometric pressure, wind speed, and direction, as well as previously observed sea level values. The second model instead takes, as input values, the astronomical tide and the previously observed sea level values, which implicitly take into account the weather conditions. Both models proved capable of predicting the extreme values of high tides with great accuracy, even greater than that of the models currently used.

Runoff Volume Reduction Using Green Infrastructure

Uncontrolled urbanization is a frequent cause behind the local flooding of catchment areas. This also results in a degradation of water quality in receivers, as well as causing a disruption of the natural water cycle in the catchment. Classical solutions, such as retention, do not prove to be sufficient under all conditions. An alternative solution is the application of low impact development (LID), which, in the analysed case, takes the form of rain gardens, infiltration trenches and controlled unsealing of catchment components. The work presents the influence of a few variants of solutions on a selected urbanized catchment located in Gorzów Wielkopolski. The assessment was developed using a simulation model, making use of EPA’s Storm Water Management Model (SWMM) software. The nalysed design variants are compared with the described existing state before the implementation of modernization works. Previous results showing that LID may be ineffective as the only solution in systems overloaded with runoff generated by rainfall of relatively low intensities were confirmed. In the case of existing systems, LID should be applied in combination with classical retention systems or in a treatment train and every opportunity to implement LID whether on a property or urban site must be taken. Such solutions in the analysed cases will allow for a reduction of the maximum outflow intensity from the analysed subcatchment by 9 to 17% depending on the analysed rainfall. The results are similar to those obtained in other implementations. However, the interpretation of the results is not as simple and obvious for overloaded systems. In addition to flow rate reduction, reduction of surcharge in the sewer network and reduction of the volume of local flooding must be considered. LID solutions should also, whenever possible, be looked into as early as the stage of planning the land development of the infrastructure.

Prediction of Active Microwave Backscatter Over Snow-Covered Terrain Across Western Colorado Using a Land Surface Model and Support Vector Machine Regression

The main objective of this article is to develop a physically constrained support vector machine (SVM) to predict C-band backscatter over snow-covered terrain as a function of geophysical inputs that reasonably represent the relevant characteristics of the snowpack. Sentinel-1 observations, in conjunction with geophysical variables from the Noah-MP land surface model, were used as training targets and input datasets, respectively. Robustness of the SVM prediction was analyzed in terms of training targets, training windows, and physical constraints related to snow liquid water content. The results showed that a combination of ascending and descending overpasses yielded the highest coverage of prediction (15.2%) while root mean square error (RMSE) ranged from 2.06 to 2.54 dB and unbiased RMSE ranged from 1.54 to 2.08 dB, but that the combined overpasses were degraded compared with ascending-only and descending-only training target sets due to the mixture of distinctive microwave signals during different times of the day (i.e., 6 A.M. versus 6 P.M. local time). Elongation of the training window length also increased the spatial coverage of prediction (given the sparsity of the training sets), but resulted in introducing more random errors. Finally, delineation of dry versus wet snow pixels for SVM training resulted in improving the accuracy of predicted backscatter relative to training on a mixture of dry and wet snow conditions. The overall results suggest that the prediction accuracy of the SVM was strongly linked with the first-order physics of the electromagnetic response of different snow conditions.

Implications of Experiment Set-Ups for Residential Water End-Use Classification

With an increasing need for secured water supply, a better understanding of the water consumption behavior is beneficial. This can be achieved through end-use classification, i.e., identifying end-uses such as toilets, showers or dishwashers from water consumption data. Previously, both supervised and unsupervised machine learning (ML) techniques are employed, demonstrating accurate classification results on particular datasets. However, a comprehensive comparison of ML techniques on a common dataset is still missing. Hence, in this study, we are aiming at a quantitative evaluation of various ML techniques on a common dataset. For this purpose, a stochastic water consumption simulation tool with high capability to model the real-world water consumption pattern is applied to generate residential data. Subsequently, unsupervised clustering methods, such as dynamic time warping, k-means, DBSCAN, OPTICS and Hough transform, are compared to supervised methods based on SVM. The quantitative results demonstrate that supervised approaches are capable to classify common residential end-uses (toilet, shower, faucet, dishwasher, washing machine, bathtub and mixed water-uses) with accuracies up to 0.99, whereas unsupervised methods fail to detect those consumption categories. In conclusion, clustering techniques alone are not suitable to separate end-use categories fully automatically. Hence, accurate labels are essential for the end-use classification of water events, where crowdsourcing and citizen science approaches pose feasible solutions for this purpose.

sEMG-Based Neural Network Prediction Model Selection of Gesture Fatigue and Dataset Optimization

The fatigue energy consumption of independent gestures can be obtained by calculating the power spectrum of surface electromyography (sEMG) signals. The existing research studies focus on the fatigue of independent gestures, while the research studies on integrated gestures are few. However, the actual gesture operation mode is usually integrated by multiple independent gestures, so the fatigue degree of integrated gestures can be predicted by training neural network of independent gestures. Three natural gestures including browsing information, playing games, and typing are divided into nine independent gestures in this paper, and the predicted model is established and trained by calculating the energy consumption of independent gestures. The artificial neural networks (ANNs) including backpropagation (BP) neural network, recurrent neural network (RNN), and long short-term memory (LSTM) are used to predict the fatigue of gesture. The support vector machine (SVM) is used to assist verification. Mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) are utilized to evaluate the optimal prediction model. Furthermore, the different datasets of the processed sEMG signal and its decomposed wavelet coefficients are trained, respectively, and the changes of error functions of them are compared. The experimental results show that LSTM model is more suitable for gesture fatigue prediction. The processed sEMG signals are appropriate for using as the training set the fatigue degree of one-handed gesture. It is better to use wavelet decomposition coefficients as datasets to predict the high-dimensional sEMG signals of two-handed gestures. The experimental results can be applied to predict the fatigue degree of complex human-machine interactive gestures, help to avoid unreasonable gestures, and improve the user's interactive experience.

Spectrophotometric Online Detection of Drinking Water Disinfectant: A Machine Learning Approach

The spectra fingerprint of drinking water from a water treatment plant (WTP) is characterised by a number of light-absorbing substances, including organic, nitrate, disinfectant, and particle or turbidity. Detection of disinfectant (monochloramine) can be better achieved by separating its spectra from the combined spectra. In this paper, two major focuses are (i) the separation of monochloramine spectra from the combined spectra and (ii) assessment of the application of the machine learning algorithm in real-time detection of monochloramine. The support vector regression (SVR) model was developed using multi-wavelength ultraviolet-visible (UV-Vis) absorbance spectra and online amperometric monochloramine residual measurement data. The performance of the SVR model was evaluated by using four different kernel functions. Results show that (i) particles or turbidity in water have a significant effect on UV-Vis spectral measurement and improved modelling accuracy is achieved by using particle compensated spectra; (ii) modelling performance is further improved by compensating the spectra for natural organic matter (NOM) and nitrate (NO₃) and (iii) the choice of kernel functions greatly affected the SVR performance, especially the radial basis function (RBF) appears to be the highest performing kernel function. The outcomes of this research suggest that disinfectant residual (monochloramine) can be measured in real time using the SVR algorithm with a precision level of ± 0.1 mg L⁻¹.

Evaluation of the Impact of Rainfall Inputs on Urban Rainfall Models: A Systematic Review

Over the past several decades, urban flooding and other water-related disasters have become increasingly prominent and serious. Although the urban rain flood model’s benefits for urban flood simulation have been extensively documented, the impact of rainfall input to model simulation accuracy remains unclear. This systematic review aims to provide structured research on how rain inputs impact urban rain flood model’s simulation accuracy. The selected 48 peer-reviewed journal articles published between 2015 and 2019 on the Web of Science™ database were analyzed by key factors, including rainfall input type, calibration times and verification times. The results from meta-analysis reveal that when a traditional rain measurement was used as the rainfall input, model simulation accuracy was higher, i.e., the Nash–Sutcliffe efficiency coefficient (NSE) of traditional technology for rain measurement was higher than the 0.18 for the new technology rain measurement with respect to flow simulation. In addition, the single-field sub-flood calibration model was better than the multi-field sub-flood calibration model. NSE was higher than 0.14. The precision was better for the verification period; NSE of the calibration value showed a 0.07 higher verification value on average in flow simulation. These findings have certain significance for the development of future urban rain flood models and propose the development direction of the future urban rain flood model. Finally, in view of the rainfall input problem of the urban storm flood model, we propose the future development direction of the urban storm flood model.

Flood Evacuation Routes Based on Spatiotemporal Inundation Risk Assessment

For flood risk assessment, it is necessary to quantify the uncertainty of spatiotemporal changes in floods by analyzing space and time simultaneously. This study designed and tested a methodology for the designation of evacuation routes that takes into account spatial and temporal inundation and tested the methodology by applying it to a flood-prone area of Seoul, Korea. For flood prediction, the non-linear auto-regressive with exogenous inputs neural network was utilized, and the geographic information system was utilized to classify evacuations by walking hazard level as well as to designate evacuation routes. The results of this study show that the artificial neural network can be used to shorten the flood prediction process. The results demonstrate that adaptability and safety have to be ensured in a flood by planning the evacuation route in a flexible manner based on the occurrence of, and change in, evacuation possibilities according to walking hazard regions.

A Hybrid Approach Combining Conceptual Hydrological Models, Support Vector Machines and Remote Sensing Data for Rainfall-Runoff Modeling

Understanding catchment response to rainfall events is important for accurate runoff estimation in many water-related applications, including water resources management. This study introduced a hybrid model, the Tank-least squared support vector machine (LSSVM), that incorporated intermediate state variables from a conceptual tank model within the least squared support vector machine (LSSVM) framework in order to describe aspects of the rainfall-runoff (RR) process. The efficacy of the Tank-LSSVM model was demonstrated with hydro-meteorological data measured in the Yongdam Catchment between 2007 and 2016, South Korea. We first explored the role of satellite soil moisture (SM) data (i.e., European Space Agency (ESA) CCI) in the rainfall-runoff modeling. The results indicated that the SM states inferred from the ESA CCISWI provided an effective means of describing the temporal dynamics of SM. Further, the Tank-LSSVM model’s ability to simulate daily runoff was assessed by using goodness of fit measures (i.e., root mean square error, Nash Sutcliffe coefficient (NSE), and coefficient of determination). The Tank-LSSVM models’ NSE were all classified as “very good” based on their performance during the training and testing periods. Compared to individual LSSVM and Tank models, improved daily runoff simulations were seen in the proposed Tank-LSSVM model. In particular, low flow simulations demonstrated the improvement of the Tank-LSSVM model compared to the conventional tank model.

Inundation Map Prediction with Rainfall Return Period and Machine Learning

To date, various methods of flood prediction using numerical analysis or machine learning have been studied. However, a methodology for simultaneously predicting the rainfall return period and an inundation map for observed rainfall has not been presented. Simultaneous prediction of the return period and inundation map would be a useful technique for responding to floods in real-time and could provide an expected inundation area by return period. In this study, return period estimation for observed rainfall was performed via PNN (probabilistic neural network). SVR (support vector regression) and a SOM (self-organizing map) were used to predict flood volume and inundation maps. The study area was the Gangnam area, which has experienced extensive urbanization. The database for training SVR and SOM was constructed by one- and two-dimensional flood analysis with consideration of 120 probable rainfall events. The probable rainfall events were composed with 2–100 year return periods and 1–3 hour durations. The SVR technique was used to predict flood volume according to the rainfall return period, and the SOM was used to cluster various expected flood patterns to be used for predicting inundation maps. The prediction results were compared with the simulation results of a two-dimensional flood analysis model. The highest fitness of the predicted flood maps in the study area was calculated at 85.94%. The proposed method was found to constitute a practical methodology that could be helpful in improving urban flood response capabilities.

Accounting for Uncertainty and Reconstruction of Flooding Patterns Based on Multi-Satellite Imagery and Support Vector Machine Technique: A Case Study of Can Tho City, Vietnam

One of the most frequent natural perils affecting the world today is flooding, and over the years, flooding has caused a large loss of life and damage to property. Remote sensing technology and satellite imagery derived data are useful in mapping the inundated area, which is useful for flood risk management. In the current paper, commonly used satellite imagery from the public domain for flood inundated extent capturing are studied considering Can Tho City as a study area. The differences in the flood inundated areas from different satellite sensors and the possible reasons are explored. An effective and relatively advanced method to address the uncertainties—inundated area capture from different remote sensing sensors—was implemented while establishing the inundated area pattern between the years 2000 and 2018. This solution involves the usage of a machine learning technique, Support Vector Machine Regression (SVR) which further helps in filling the gaps whenever there is lack of data from a single satellite data source. This useful method could be extended to establish the inundated area patterns over the years in data-sparse regions and in areas where access is difficult. Furthermore, the method is economical, as freely available data are used for the purpose.

Urban Flood Prediction Using Deep Neural Network with Data Augmentation

Data-driven models using an artificial neural network (ANN), deep learning (DL) and numerical models are applied in flood analysis of the urban watershed, which has a complex drainage system. In particular, data-driven models using neural networks can quickly present the results and be used for flood forecasting. However, not a lot of data with actual flood history and heavy rainfalls are available, it is difficult to conduct a preliminary analysis of flood in urban areas. In this study, a deep neural network (DNN) was used to predict the total accumulative overflow, and because of the insufficiency of observed rainfall data, 6 h of rainfall were surveyed nationwide in Korea. Statistical characteristics of each rainfall event were used as input data for the DNN. The target value of the DNN was the total accumulative overflow calculated from Storm Water Management Model (SWMM) simulations, and the methodology of data augmentation was applied to increase the input data. The SWMM is one-dimensional model for rainfall-runoff analysis. The data augmentation allowed enrichment of the training data for DNN. The data augmentation was applied ten times for each input combination, and the practicality of the data augmentation was determined by predicting the total accumulative overflow over the testing data and the observed rainfall. The prediction result of DNN was compared with the simulated result obtained using the SWMM model, and it was confirmed that the predictive performance was improved on applying data augmentation.

Application of the deep learning for the prediction of rainfall in Southern Taiwan

Precipitation is useful information for assessing vital water resources, agriculture, ecosystems and hydrology. Data-driven model predictions using deep learning algorithms are promising for these purposes. Echo state network (ESN) and Deep Echo state network (DeepESN), referred to as Reservoir Computing (RC), are effective and speedy algorithms to process a large amount of data. In this study, we used the ESN and the DeepESN algorithms to analyze the meteorological hourly data from 2002 to 2014 at the Tainan Observatory in the southern Taiwan. The results show that the correlation coefficient by using the DeepESN was better than that by using the ESN and commercial neuronal network algorithms (Back-propagation network (BPN) and support vector regression (SVR), MATLAB, The MathWorks co.), and the accuracy of predicted rainfall by using the DeepESN can be significantly improved compared with those by using ESN, the BPN and the SVR. In sum, the DeepESN is a trustworthy and good method to predict rainfall; it could be applied to global climate forecasts which need high-volume data processing.

Urban Drainage Networks Rehabilitation Using Multi-Objective Model and Search Space Reduction Methodology

The drainage network always needs to adapt to environmental and climatic conditions to provide best quality services. Rehabilitation combining pipes substitution and storm tanks installation appears to be a good solution to overcome this problem. Unfortunately, the calculation time of such a rehabilitation scenario is too elevated for single-objective and multi-objective optimization. In this study, a methodology composed by search space reduction methodology whose purpose is to decrease the number of decision variables of the problem to solve and a multi-objective optimization whose purpose is to optimize the rehabilitation process and represent Pareto fronts as the result of urban drainage networks optimization is proposed. A comparison between different model results for multi-objective optimization is made. To obtain these results, Storm Water Management Model (SWMM) is first connected to a Pseudo Genetic Algorithm (PGA) for the search space reduction and then to a Non-Dominated Sorting Genetic Algorithm II (NSGA-II) for multi-objective optimization. Pareto fronts are designed for investment costs instead of flood damage costs. The methodology is applied to a real network in the city of Medellin in Colombia. The results show that search space reduction methodology provides models with a considerably reduced number of decision variables. The multi-objective optimization shows that the models’ results used after the search space reduction obtain better outcomes than in the complete model in terms of calculation time and optimality of the solutions.

Research on the Construction Method of the Service-Oriented Web-SWMM System

On a global scale, with the acceleration of urbanization and the continuous expansion of cities, the problem of urban flooding has become increasingly prominent. An increasing number of experts and scholars have begun to focus on this phenomenon and build corresponding models to solve the problem. The storm water management model 5 (SWMM5) is a dynamic rainfall-runoff simulation model developed by the US Environmental Protection Agency (EPA); this model simulates urban flooding and drainage well and is widely favored by researchers. However, the use of SWMM5 is relatively cumbersome and limited by the operational platform, and these factors hinder the further promotion and sharing of SWMM5. Based on the OpenGMS platform, this study first encapsulates, deploys, and publishes SWMM5 and further builds the Web-SWMM system for the model. With Web-SWMM, the user can conveniently use network data resources online and call SWMM5 to carry out calculations, avoiding the difficulties caused by the localized use of SWMM5 and enabling the sharing and reuse of SWMM5.

Quantification of Stream Drying Phenomena Using Grid-Based Hydrological Modeling via Long-Term Data Mining throughout South Korea including Ungauged Areas

The Drying Stream Assessment Tool and Water Flow Tracking (DrySAT-WFT) were modified to simulate the hydrological components of water loss databases (DBs) affecting stream drying phenomena. In this study, the phenomenon is defined based on a method using the 10-day minimum flow (reference Q355). Prior to identifying the method using reference Q355, the DrySAT-WFT model was calibrated and verified for its performance with the total runoff (TQ), evapotranspiration (ET), and soil moisture (SM) at 12 streamflow locations, 3 ET locations, and 58 SM locations. The average R2 for TQ in 2005 to 2015 were 0.66 to 0.84, which demonstrates good performance. Moreover, Nash Sutcliffe model efficiency (NSE) values were 0.52 to 0.72, which are also good. After verifying the DrySAT-WFT model for hydrologic components, in order to apply the method, this study defined the drying progress which was analyzed by the stream drying index (SDI) as decision criteria. In this study, the criteria for the estimation of SDI were calculated as reference Q355 coming from the 10-day minimum flow considering only weather changes from 1976 to 2015. Then, SDI grades were determined by counting the number of days below a reference Q355 from TQ considering all water loss databases (DBs) such as weather changes, groundwater uses, forest heights, soil depths, land use, and road network. On the other hand, SDI represents how many days below the reference Q355 increased when all water loss DBs were applied, in comparison to when only weather changes were applied. The DrySAT-WFT model simulated the hydrological components of the water balance based on each water loss DB, including the application of all DBs. As a result, the change ratios for TQ were measured: −4.8% for groundwater use (GWU), −1.3% for forest height (FH), −0.3% for road network (RN), −0.1% for land use (LU) and −0.1% for soil depth (SD). Overall, TQ values decreased by -8.4%. The change ratios for ET were measured: −2.0% for GWU, +10.5% for FH, +5.6% for RN, −1.8% for LU and +0.3% for SD. Overall, the ET values increased by +14.7%. In addition, based on all water loss DBs, the SDI was evaluated for all watersheds, which intensified recently (2006–2015). Under weather DB conditions, the average SDI was measured as 2.0 for all watersheds. Stream drying processes remained limited, requiring only monitoring. Given baseline conditions, stream drying intensified to grades of 3.1 (1976–1985), 3.2 (1986–1995), 3.3 (1996–2005) and 3.5 (2006–2015) by all water loss DBs.

Optimizing the Water Treatment Design and Management of the Artificial Lake with Water Quality Modeling and Surrogate-Based Approach

The tradeoff between engineering costs and water treatment of the artificial lake system has a significant effect on engineering decision-making. However, decision-makers have little access to scientific tools to balance engineering costs against corresponding water treatment. In this study, a framework integrating numerical modeling, surrogate models and multi-objective optimization is proposed. This framework was applied to a practical case in Chengdu, China. A water quality model (MIKE21) was developed, providing training datasets for surrogate modeling. The Artificial Neural Network (ANN) and Support Vector Machine (SVM) were utilized for training surrogate models. Both surrogate models were validated with the coefficient of determinations (R2) greater than 0.98. SVM performed more stably with limited training data sizes while ANN demonstrated higher accuracies with more training samples. The multi-objective optimization model was developed using the genetic algorithm, with targets of reducing both engineering costs and target aquatic pollutant concentrations. An optimal target concentration after treatment was identified, characterized by the ammonia concentration (1.3 mg/L) in the artificial lake. Furthermore, scenarios with varying water quality in the upstream river were evaluated. Given the assumption of deteriorated upstream water quality in the future, the optimal proportion of pre-treatment in the total costs is increasing.

An Improved LSSVM Model for Intelligent Prediction of the Daily Water Level

Daily water level forecasting is of significant importance for the comprehensive utilization of water resources. An improved least squares support vector machine (LSSVM) model was introduced by including an extra bias error control term in the objective function. The tuning parameters were determined by the cross-validation scheme. Both conventional and improved LSSVM models were applied in the short term forecasting of the water level in the middle reaches of the Yangtze River, China. Evaluations were made with both models through metrics such as RMSE (Root Mean Squared Error), MAPE (Mean Absolute Percent Error) and index of agreement (d). More accurate forecasts were obtained although the improvement is regarded as moderate. Results indicate the capability and flexibility of LSSVM-type models in resolving time sequence problems. The improved LSSVM model is expected to provide useful water level information for the managements of hydroelectric resources in Rivers.

Flood Prediction Using Machine Learning Models: Literature Review

Floods are among the most destructive natural disasters, which are highly complex to model. The research on the advancement of flood prediction models contributed to risk reduction, policy suggestion, minimization of the loss of human life, and reduction of the property damage associated with floods. To mimic the complex mathematical expressions of physical processes of floods, during the past two decades, machine learning (ML) methods contributed highly in the advancement of prediction systems providing better performance and cost-effective solutions. Due to the vast benefits and potential of ML, its popularity dramatically increased among hydrologists. Researchers through introducing novel ML methods and hybridizing of the existing ones aim at discovering more accurate and efficient prediction models. The main contribution of this paper is to demonstrate the state of the art of ML models in flood prediction and to give insight into the most suitable models. In this paper, the literature where ML models were benchmarked through a qualitative analysis of robustness, accuracy, effectiveness, and speed are particularly investigated to provide an extensive overview on the various ML algorithms used in the field. The performance comparison of ML models presents an in-depth understanding of the different techniques within the framework of a comprehensive evaluation and discussion. As a result, this paper introduces the most promising prediction methods for both long-term and short-term floods. Furthermore, the major trends in improving the quality of the flood prediction models are investigated. Among them, hybridization, data decomposition, algorithm ensemble, and model optimization are reported as the most effective strategies for the improvement of ML methods. This survey can be used as a guideline for hydrologists as well as climate scientists in choosing the proper ML method according to the prediction task.

Efficiency of Flood Control Measures in a Sewer System Located in the Mediterranean Basin

Pollution induced by surface runoff in urban areas constitutes a significant problem. The adoption of control measures aimed at improving the quality of recipient water bodies is a fundamental issue in the management of Mediterranean Basin sewer systems. Previous research in Mediterranean areas using small virtual basins has shown that rainfall regimes have a limited impact on the pollutant load and discharge flowing into a receiving body. The aim of our research was to identify a sizing methodology for stormwater tanks located in the Mediterranean Basin. To achieve this, a numerical model of a sewer system, located in the Southern Iberian Peninsula, was developed. Different patterns related to peak periods of rainfall were considered. Furthermore, efficiency indices were used to evaluate and compare the effects of having a stormwater tank in the system. In our study (which considered a real area), significantly varied values were obtained for the pollution load removal rate (η) and the receiver overflow rate (θ). We nevertheless observed that, in our catchment, at a specific volume of V = 60 m3/ha, η and θ reached constant values without experiencing any significant improvement (η = 0.673 and θ = 0.133). Based on our model, this volume was proposed for the stormwater tank. The ATV (German Association for Water Pollution Control) A 128 standard was applied in order to validate the results, and the specific volume obtained (V = 60 m3/ha) matched with the one proposed. Thus, our proposed methodology is simple and different, and it is very easy to apply by obtaining the values of the efficiency indices η and θ through the development of a Storm Water Management Model (SWMM).

Future Predictions of Rainfall and Temperature Using GCM and ANN for Arid Regions: A Case Study for the Qassim Region, Saudi Arabia

Future predictions of rainfall patterns in water-scarce regions are highly important for effective water resource management. Global circulation models (GCMs) are commonly used to make such predictions, but these models are highly complex and expensive. Furthermore, their results are associated with uncertainties and variations for different GCMs for various greenhouse gas emission scenarios. Data-driven models including artificial neural networks (ANNs) and adaptive neuro fuzzy inference systems (ANFISs) can be used to predict long-term future changes in rainfall and temperature, which is a challenging task and has limitations including the impact of greenhouse gas emission scenarios. Therefore, in this research, results from various GCMs and data-driven models were investigated to study the changes in temperature and rainfall of the Qassim region in Saudi Arabia. Thirty years of monthly climatic data were used for trend analysis using Mann–Kendall test and simulating the changes in temperature and rainfall using three GCMs (namely, HADCM3, INCM3, and MPEH5) for the A1B, A2, and B1 emissions scenarios as well as two data-driven models (ANN: feed-forward-multilayer, perceptron and ANFIS) without the impact of any emissions scenario. The results of the GCM were downscaled for the Qassim region using the Long Ashton Research Station’s Weather Generator 5.5. The coefficient of determination (R2) and Akaike’s information criterion (AIC) were used to compare the performance of the models. Results showed that the ANNs could outperform the ANFIS for predicting long-term future temperature and rainfall with acceptable accuracy. All nine GCM predictions (three models with three emissions scenarios) differed significantly from one another. Overall, the future predictions showed that the temperatures of the Qassim region will increase with a specified pattern from 2011 to 2099, whereas the changes in rainfall will differ over various spans of the future.

The Integration of Nature-Inspired Algorithms with Least Square Support Vector Regression Models: Application to Modeling River Dissolved Oxygen Concentration

The current study investigates an improved version of Least Square Support Vector Machines integrated with a Bat Algorithm (LSSVM-BA) for modeling the dissolved oxygen (DO) concentration in rivers. The LSSVM-BA model results are compared with those obtained using M5 Tree and Multivariate Adaptive Regression Spline (MARS) models to show the efficacy of this novel integrated model. The river water quality data at three monitoring stations located in the USA are considered for the simulation of DO concentration. Eight input combinations of four water quality parameters, namely, water temperature, discharge, pH, and specific conductance, are used to simulate the DO concentration. The results revealed the superiority of the LSSVM-BA model over the M5 Tree and MARS models in the prediction of river DO. The accuracy of the LSSVM-BA model compared with those of the M5 Tree and MARS models is found to increase by 20% and 42%, respectively, in terms of the root-mean-square error. All the predictive models are found to perform best when all the four water quality variables are used as input, which indicates that it is possible to supply more information to the predictive model by way of incorporation of all the water quality variables.

Application of Least-Squares Support Vector Machines for Quantitative Evaluation of Known Contaminant in Water Distribution System Using Online Water Quality Parameters

In water-quality, early warning systems and qualitative detection of contaminants are always challenging. There are a number of parameters that need to be measured which are not entirely linearly related to pollutant concentrations. Besides the complex correlations between variable water parameters that need to be analyzed also impairs the accuracy of quantitative detection. In aspects of these problems, the application of least-squares support vector machines (LS-SVM) is used to evaluate the water contamination and various conventional water quality sensors quantitatively. The various contaminations may cause different correlative responses of sensors, and also the degree of response is related to the concentration of the injected contaminant. Therefore to enhance the reliability and accuracy of water contamination detection a new method is proposed. In this method, a new relative response parameter is introduced to calculate the differences between water quality parameters and their baselines. A variety of regression models has been examined, as result of its high performance, the regression model based on genetic algorithm (GA) is combined with LS-SVM. In this paper, the practical application of the proposed method is considered, controlled experiments are designed, and data is collected from the experimental setup. The measured data is applied to analyze the water contamination concentration. The evaluation of results validated that the LS-SVM model can adapt to the local nonlinear variations between water quality parameters and contamination concentration with the excellent generalization ability and accuracy. The validity of the proposed approach in concentration evaluation for potassium ferricyanide is proven to be more than 0.5 mg/L in water distribution systems.

Modeling the Effects of Introducing Low Impact Development in a Tropical City: A Case Study from Joinville, Brazil

In tropical countries like Brazil, fast and uncontrolled urbanization, together with high rainfall intensities, makes flooding a frequent event. The implementation of decentralized stormwater controls is a promising strategy aiming to reduce surface runoff and pollution through retention, infiltration, filtration, and evapotranspiration of stormwater. Although the application of such controls has increased in the past years in developed countries, they are still not a common approach in developing countries, such as Brazil. In this paper we evaluate to what extend different low impact development (LID) techniques are able to reduce the flood risk in an area of high rainfall intensities in a coastal region of South Brazil. Feasible scenarios of placing LID units throughout the catchment were developed, analyzed with a hydrodynamic solver, and compared against the baseline scenario to evaluate the potential of flood mitigation. Results show that the performance improvements of different LID scenarios are highly dependent on the rainfall events. On average, a total flood volume reduction between 30% and 75% could be achieved for seven LID scenarios. For this case study the best results were obtained when using a combination of central and decentral LID units, namely detention ponds, infiltration trenches, and rain gardens.

Evaluating the Hydrologic Performance of Low Impact Development Scenarios in a Micro Urban Catchment

As urbanization progresses, increasingly impervious surfaces have changed the hydrological processes in cities and resulted in a major challenge for urban stormwater control. This study uses the urban stormwater model to evaluate the performance and costs of low impact development (LID) scenarios in a micro urban catchment. Rainfall-runoff data of three rainfall events were used for model calibration and validation. The pre-developed (PreDev) scenario, post-developed (PostDev) scenario, and three LID scenarios were used to evaluate the hydrologic performance of LID measures. Using reduction in annual runoff as the goal, the best solutions for each LID scenario were selected using cost-effectiveness curves. The simulation results indicated that the three designed LID scenarios could effectively reduce annual runoff volumes and pollutant loads compared with the PostDev scenario. The most effective scenario (MaxPerf) reduced annual runoff by 53.4%, followed by the sponge city (SpoPerf, 51.5%) and economy scenarios (EcoPerf, 43.1%). The runoff control efficiency of the MaxPerf and SpoPerf scenarios increased by 23.9% and 19.5%, respectively, when compared with the EcoPerf scenario; however, the costs increased by 104% and 83.6%. The reduction rates of four pollutants (TSS, TN, TP, and COD) under the MaxPerf scenario were 59.8-61.1%, followed by SpoPerf (53.9-58.3%) and EcoPerf (42.3-45.4%), and the costs of the three scenarios were 3.74, 3.47, and 1.83 million yuan, respectively. These results can provide guidance to urban stormwater managers in future urban planning to improve urban water security.

Machine Learning Algorithms for the Forecasting of Wastewater Quality Indicators

Stormwater runoff is often contaminated by human activities. Stormwater discharge into water bodies significantly contributes to environmental pollution. The choice of suitable treatment technologies is dependent on the pollutant concentrations. Wastewater quality indicators such as biochemical oxygen demand (BOD5), chemical oxygen demand (COD), total suspended solids (TSS), and total dissolved solids (TDS) give a measure of the main pollutants. The aim of this study is to provide an indirect methodology for the estimation of the main wastewater quality indicators, based on some characteristics of the drainage basin. The catchment is seen as a black box: the physical processes of accumulation, washing, and transport of pollutants are not mathematically described. Two models deriving from studies on artificial intelligence have been used in this research: Support Vector Regression (SVR) and Regression Trees (RT). Both the models showed robustness, reliability, and high generalization capability. However, with reference to coefficient of determination R2 and root‐mean square error, Support Vector Regression showed a better performance than Regression Tree in predicting TSS, TDS, and COD. As regards BOD5, the two models showed a comparable performance. Therefore, the considered machine learning algorithms may be useful for providing an estimation of the values to be considered for the sizing of the treatment units in absence of direct measures.

Lost in calibration: why people still do not calibrate their models, and why they still should - a case study from urban drainage modelling

From a scientific point of view, it is unquestioned that numerical models for technical systems need to be calibrated. However, in sufficiently calibrated models are still used in engineering practice. Case studies in the scientific literature that deal with urban water management are mostly large cities, while little attention is paid to the differing boundary conditions of smaller municipalities. Consequently, the aim of this paper is to discuss the calibration of a hydrodynamic model of a small municipality (15,000 inhabitants). To represent the spatial distribution of precipitation, three distributed rain gauges were used for model calibration. To show the uncertainties imminent to the calibration process, 17 scenarios, differing in assumptions for calibration, were distinguished. To compare the impact of the different calibration scenarios on actual design values, design rainfall events were applied. The comparison of the model results using the different typical design storm events from all the surrounding data points showed substantial differences for the assessment of the sewers regarding urban flooding, emphasizing the necessity of uncertainty analysis for hydrodynamic models. Furthermore, model calibration is of the utmost importance, because uncalibrated models tend to overestimate flooding volume and therefore result in larger diameters and retention volumes.