Number Of Water Quality Variables Research Articles

Prediction enhancement for surface water sodium adsorption ratio using limited inputs: Implementation of hybridized stacked ensemble model with feature selection algorithm

The Sodium Adsorption Ratio (SAR) is a widely used variable in water quality research, particularly in agriculture and environmental studies. In many cases, the key variables required for SAR calculation, namely Na+, Mg+2, and Ca+2, are not available. Consequently, the potential to calculate SAR using a limited number of water quality variables becomes critically important. The study implemented the Multilayer Perceptron Neural Network (MLPNN), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN) models at level-0 for prediction purposes, along with the Boruta model for variable selection. A stacked ensemble learning model at level-1 enhanced the prediction accuracy. The discharge and water quality dataset from the Zarrin-Gol River in northern Iran was utilized to implement the modeling procedure. Results obtained from the variable selection process using the Boruta model revealed that using a limited number of water quality variables can effectively predict SAR even without the principal variables. Further investigation of the input combinations for the level-0 models demonstrated that, for the MLPNN, KNN, and SVR models, 4, 3, and 1 input variables, respectively, yielded optimal predictions. Among the level-0 models, the MLPNN model exhibited the highest accuracy, with RMSE = 0.54, MBE = 0.26, MAE = 0.44, R = 0.84, IA = 0.67, and KGE = 0.79. Implementing the stacked ensemble learning model at level-1 significantly improved the SAR prediction compared to the level-0 models. The ensemble-NN model yielded the best performance in estimating SAR within the range of recorded data, with RMSE = 0.53, MBE = 0.29, MAE = 0.41, R = 0.87, IA = 0.70, and KGE = 0.82. Residual analysis further confirmed the superior predictive capability of the level-1 models compared to the level-0 models. The generalized-logistic probability distribution function is used to estimate the extreme values data. The Ensemble-KNN model best predicted extreme values data, with RMSE = 0.69, MBE = −0.61, MAE = 0.61, R = 0.61, IA = 0.26, and KGE = 0.37. The findings underscore the substantial advancements achieved through stacked ensemble methods in enhancing the modeling of SAR across various aspects, including total data, extreme values, and models' residuals.

Shellfish farm closure prediction and cause identification using machine learning methods

Shellfish farms are needed to be closed if they are contaminated during their production as otherwise it may lead to serious health hazard. The authorities monitor a number of water quality variables to check the health of shellfish farms and to decide on the closure of the farms. The research presented in this paper aims to automate this process by developing novel algorithms to identify the cause of closure and also predicting the closure. As the frequency of closure is relatively very small, the labelled data sets are imbalanced in nature. We have developed a novel ensemble feature ranking algorithm that explicitly deals with class imbalance problem and identifies the cause of closure. We have also presented a class balancing ensemble classifier to predict shellfish farm closure. The class balancing ensemble classifier predicts closure/opening with as high as 71.69% accuracy and achieves best balancing act with decision tree base classifier in 75% locations. Rain and salinity are found to be the key causes of closure and the causality depends of the properties of the locations.

A study on the effect ofin situpurification on coal mine drainage by loess

AbstractTwo in situ tests, in an unsaturated loess zone and an unconfined aquifer (northern Shaanxi Loess Plateau, Shaanxi, China), were conducted to investigate the potential of loess to purify coal mine drainage water. The first test involved the analysis of loess samples that were collected at the start of the test and at 20 days after the termination of the test. Results indicated general declines in the loess of two heavy metals: Cd declined from ∼3 to <1.5 mg/kg and Pb from ∼4 to <1.0 mg/kg. This demonstrated the capacity for removal of Cd and Pb through the irrigation of unsaturated loess. A second test indicated that, although the concentrations of a number of water quality variables remained high in the groundwater of the unconfined aquifer, after irrigation of loess there was a steady decline in the concentrations of Cd and petroleum, as well as a decline in electrical conductivity (EC). From the drainage source to a lower elevation site (of 5 m), the concentration of Cd and petroleum declined f...

A probabilistic water quality index for river water quality assessment: a case study

Available water quality indices have some limitations such as incorporating a limited number of water quality variables and providing deterministic outputs. This paper presents a hybrid probabilistic water quality index by utilizing fuzzy inference systems (FIS), Bayesian networks (BNs), and probabilistic neural networks (PNNs). The outputs of two traditional water quality indices, namely the indices proposed by the National Sanitation Foundation and the Canadian Council of Ministers of the Environment, are selected as inputs of the FIS. The FIS is trained based on the opinions of several water quality experts. Then the trained FIS is used in a Monte Carlo analysis to provide the required input-output data for training both the BN and PNN. The trained BN and PNN can be used for probabilistic water quality assessment using water quality monitoring data. The efficiency and applicability of the proposed methodology is evaluated using water quality data obtained from water quality monitoring system of the Jajrood River in Iran.

A statistical approach for the rationalization of water quality indicators in surface water quality monitoring networks

Summary Despite several decades of operation and the increasing importance of water quality monitoring (WQM) networks, authorities still rely on subjective or semi-subjective decision processes to identify the optimal combination of water quality variables to measure. For this purpose, a statistical approach is developed for assessment and selection of the optimal combination of water quality variables. The proposed approach overcomes deficiencies in the conventional correlation–regression approach used to assess and eventually reduce the number of water quality variables in WQM networks. For the reduction of water quality variables, criteria developed from record-augmentation procedures are integrated with correlation analysis and cluster analysis to identify highly associated water quality variables. This step is followed by the application of an information performance index to systematically identify the optimal combination of variables to be continuously measured and those to be discontinued. The linear regression and maintenance of variance (MOVE) record-extension techniques are employed to reconstitute information about discontinued variables. The proposed approach is applied for rationalization of the water quality variables in the Nile Delta surface WQM network in Egypt. Results indicate that the proposed approach represents a useful decision support tool for the optimized selection of water quality variables. The MOVE record-extension technique is shown to result in better performance than regression for the estimation of discontinued variables.

An empirical study on sea water quality prediction

This paper studies the problem of predicting future values for a number of water quality variables, based on measurements from under-water sensors. It performs both exploratory and automatic analysis of the collected data with a variety of linear and nonlinear modeling methods. The paper investigates issues, such as the ability to predict future values for a varying number of days ahead and the effect of including values from a varying number of past days. Experimental results provide interesting insights on the predictability of the target variables and the performance of the different learning algorithms.

Improved Method for Aggregation of Water Quality Subindices

Alternative methods to describe water quality using an aggregate index consisting of subindices for individual water quality variables are examined. Most aggregation methods suffer from three shortcomings: Ambiguity, eclipsing, and rigidity. Ambiguity problems exist when all the subindices indicate acceptable water quality for a given use, but the aggregated index does not. Eclipsing problems exist when the aggregated index fails to reflect poor water quality of one or more water quality variables. Rigidity problems exist when additional variables are included in the index to address specific water quality concerns, but the faulty aggregation function might artificially reduce the value of the water quality index such that it does not accurately reflect the true water quality. As the number of water quality variables increases, the magnitude of the aggregated index decreases raising the issue of ambiguity again. The writers developed a mathematical formulation for aggregate indices that avoids the problems of ambiguity, eclipsing, and rigidity with respect to the number of water quality variables required to be aggregated in a given index.

Odour emissions from anaerobic piggery ponds. 2: Improving estimates of emission rate through recognition of spatial variability

Odour emission rates were measured for seven different anaerobic ponds treating piggery wastes at six to nine discrete locations across the surface of each pond on each sampling occasion over a thirteen month period. Significant variability in emission rates were observed for each pond. Measurement of a number of water quality variables in pond liquor samples collected at the same time and from the same locations as the odour samples indicated that the composition of the pond liquor was also variable. The results indicated that spatial variability was a real phenomenon and could have a significant impact on odour assessment practices. Considerably more odour samples would be required to characterise pond emissions than currently recommended by most practitioners, or regulatory agencies.

Persistence and stability of lotic invertebrate communities in New Zealand

1. Persistence and stability of lotic invertebrate communities were determined at an annual time scale over a 9‐year period (1990–98) at 26 river sites over the northern half of New Zealand. In addition, a number of water quality variables were measured monthly and flow information collected continuously over the same period at each site.2. The aim of the study was to describe the levels of interannual variability in invertebrate communities, and relate community changes to variability in environmental conditions. The consequences of this temporal variability for the sensitivity of predictive models were also investigated.3. Levels of change in environmental conditions varied significantly between years, but were relatively similar across sites. In contrast, community persistence (similarity between years in species assemblage composition), and stability (similarity between years with respect to relative abundance of species in the community) both varied significantly between sites, but changes between years were similar. Community stability was highest at sites with relatively harsh flow conditions (high coefficient of variation, high relative size of floods), and was also greater in communities dominated by Ephemeroptera.4. Relationships between change in environmental conditions and changes in community composition and structure were relatively weak for most individual sites. However, when average levels of change for each of the 26 sites were used, communities showed greater persistence under conditions where flow conditions remained relatively constant. Water quality changes had no significant effect on community persistence when assessed for all 26 sites combined.5. Results from this study suggest that lotic invertebrate communities fluctuate around a relatively stable state, at least over a 9‐year period. However, the extent of interannual variation in community composition and structure observed, along with the relatively low degree of cluster fidelity observed within a single region, suggests that predictive models based on reference site conditions extrapolated over several years should be applied with caution in New Zealand streams.

Optimization of Water Quality Monitoring Networks

At the moment the water quality monitoring network in the main surface waters in The Netherlands includes almost 400 sampling locations with a sampling interval of 1 to 4 weeks. The number of water quality variables analysed varies per location from 15 up to 100. Recent developments, such as limiting financial and laboratory capacities and changing objectives of the routine water quality investigations, necessitate an optimization of this monitoring network. Being an essential element in the optimization procedure, a relationship has to be found between the cost of obtaining information from the network and the effectiveness of the information, the latter being strongly dependent on the objective(s) of the network. In this paper a general optimization approach is presented. Also a method is proposed, worked out and applied, that relates the effectiveness of the information to the sampling frequency of the water quality monitoring network. This method can be used for the optimization of the sampling frequency for the main objectives of the routine water quality research i.e. the detection of trends in water quality constituents.

Water Quality in Channel Catfish Production Ponds

AbstractA number of water quality variables were measured at 1‐ to 2‐week intervals in five channel catfish (Ictalurus punctatus L.) production ponds at Auburn, Alabama, during the 1978 growing season. Concentrations of most variables increased with time since feeding rates were progressively increased as fish grew. The DO concentration at dawn was often lower than 2.0 mg/liter during August and September. Emergency aeration was used one or more times in each pond to minimize the danger of fish kills because of low DO. In addition, unionized ammonia (NH3) and carbon dioxide concentrations were occasionally high enough to possibly have had adverse effects on fish. A computer simulation model for predicting the early morning DO concentration in ponds proved efficient. The average difference between measured and predicted DO concentration, disregarding sign, was 0.47 mg/liter for 30 trials. Spillway discharge from channel catfish ponds was more concentrated in pollutants than streams near Auburn, Alabama.