Artificial neural networks (ANNs) have been employed to solve a broad variety of tasks. The selection of an ANN model with appropriate weights is important in achieving accurate results. This paper presents an optimization strategy for ANN model selection based on the cuckoo search (CS) algorithm, which is rooted in the obligate brood parasitic actions of some cuckoo species. In order to enhance the convergence ability of basic CS, some modifications are proposed. The fraction Pa of the n nests replaced by new nests is a fixed parameter in basic CS. As the selection of Pa is a challenging issue and has a direct effect on exploration and therefore on convergence ability, in this work the Pa is set to a maximum value at initialization to achieve more exploration in early iterations and it is decreased during the search to achieve more exploitation in later iterations until it reaches the minimum value in the final iteration. In addition, a novel master-leader-slave multi-population strategy is used where the slaves employ the best fitness function among all slaves, which is selected by the leader under a certain condition. This fitness function is used for subsequent Lévy flights. In each iteration a copy of the best solution of each slave is migrated to the master and then the best solution is found by the master. The method is tested on benchmark classification and time series prediction problems and the statistical analysis proves the ability of the method. This method is also applied to a real-world water quality prediction problem with promising results.
Thermal structure and water quality in a large and shallow lake in Malaysia were studied between January 2012 and June 2013 in order to understand variations in relation to water level fluctuations and in-stream mining activities. Environmental variables, namely temperature, turbidity, dissolved oxygen, pH, electrical conductivity, chlorophyll-A and transparency, were measured using a multi-parameter probe and a Secchi disk. Measurements of environmental variables were performed at 0.1 m intervals from the surface to the bottom of the lake during the dry and wet seasons. High water level and strong solar radiation increased temperature stratification. River discharges during the wet season, and unsustainable sand mining activities led to an increased turbidity exceeding 100 NTU, and reduced transparency, which changed the temperature variation and subsequently altered the water quality pattern.
The pollution status of the downstream section of the Jakara River was investigated. Dissolved oxygen (DO), 5-day biochemical oxygen demand (BOD(5)), chemical oxygen demand (COD), suspended solids (SS), pH, conductivity, salinity, temperature, nitrogen in the form of ammonia (NH(3)), turbidity, dissolved solids (DS), total solids (TS), nitrates (NO(3)), chloride (Cl) and phosphates (PO(3-)(4)) were evaluated, using both dry and wet season samples, as a measure of variation in surface water quality in the area. The results obtained from the analyses were correlated using Pearson's correlation matrix, principal component analysis (PCA) and paired sample t-tests. Positive correlations were observed for BOD(5), NH(3), COD, and SS, turbidity, conductivity, salinity, DS, TS for dry and wet seasons, respectively. PCA was used to investigate the origin of each water quality parameter, and yielded 5 varimax factors for each of dry and wet seasons, with 70.7 % and 83.1 % total variance, respectively. A paired sample t-test confirmed that the surface water quality varies significantly between dry and wet season samples (P < 0.01). The source of pollution in the area was concluded to be of anthropogenic origin in the dry season and natural origins in the wet season.
Poor water quality is a serious problem in the world which threatens human health, ecosystems, and plant/animal life. Prediction of surface water quality is a main concern in water resource and environmental systems. In this research, the support vector machine and two methods of artificial neural networks (ANNs), namely feed forward back propagation (FFBP) and radial basis function (RBF), were used to predict the water quality index (WQI) in a free constructed wetland. Seventeen points of the wetland were monitored twice a month over a period of 14 months, and an extensive dataset was collected for 11 water quality variables. A detailed comparison of the overall performance showed that prediction of the support vector machine (SVM) model with coefficient of correlation (R(2)) = 0.9984 and mean absolute error (MAE) = 0.0052 was either better or comparable with neural networks. This research highlights that the SVM and FFBP can be successfully employed for the prediction of water quality in a free surface constructed wetland environment. These methods simplify the calculation of the WQI and reduce substantial efforts and time by optimizing the computations.
This case study uses several univariate and multivariate statistical techniques to evaluate and interpret a water quality data set obtained from the Klang River basin located within the state of Selangor and the Federal Territory of Kuala Lumpur, Malaysia. The river drains an area of 1,288 km(2), from the steep mountain rainforests of the main Central Range along Peninsular Malaysia to the river mouth in Port Klang, into the Straits of Malacca. Water quality was monitored at 20 stations, nine of which are situated along the main river and 11 along six tributaries. Data was collected from 1997 to 2007 for seven parameters used to evaluate the status of the water quality, namely dissolved oxygen, biochemical oxygen demand, chemical oxygen demand, suspended solids, ammoniacal nitrogen, pH, and temperature. The data were first investigated using descriptive statistical tools, followed by two practical multivariate analyses that reduced the data dimensions for better interpretation. The analyses employed were factor analysis and principal component analysis, which explain 60 and 81.6% of the total variation in the data, respectively. We found that the resulting latent variables from the factor analysis are interpretable and beneficial for describing the water quality in the Klang River. This study presents the usefulness of several statistical methods in evaluating and interpreting water quality data for the purpose of monitoring the effectiveness of water resource management. The results should provide more straightforward data interpretation as well as valuable insight for managers to conceive optimum action plans for controlling pollution in river water.
The study presents a baseline variability and climatology study of measured hydrodynamic, water properties and some water quality parameters of West Johor Strait, Singapore at hourly-to-seasonal scales to uncover their dependency and correlation to one or more drivers. The considered parameters include, but not limited by sea surface elevation, current magnitude and direction, solar radiation and air temperature, water temperature, salinity, chlorophyll-a and turbidity. FFT (Fast Fourier Transform) analysis is carried out for the parameters to delineate relative effect of tidal and weather drivers. The group and individual correlations between the parameters are obtained by principal component analysis (PCA) and cross-correlation (CC) technique, respectively. The CC technique also identifies the dependency and time lag between driving natural forces and dependent water property and water quality parameters. The temporal variability and climatology of the driving forces and the dependent parameters are established at the hourly, daily, fortnightly and seasonal scales.
Rivers play a significant role in providing water resources for human and ecosystem survival and health. Hence, river water quality is an important parameter that must be preserved and monitored. As the state of Selangor and the city of Kuala Lumpur, Malaysia, are undergoing tremendous development, the river is subjected to pollution from point and non-point sources. The water quality of the Klang River basin, one of the most densely populated areas within the region, is significantly degraded due to human activities as well as urbanization. Evaluation of the overall river water quality status is normally represented by a water quality index (WQI), which consists of six parameters, namely dissolved oxygen, biochemical oxygen demand, chemical oxygen demand, suspended solids, ammoniacal nitrogen and pH. The objectives of this study are to assess the water quality status for this tropical, urban river and to establish the WQI trend. Using monthly WQI data from 1997 to 2007, time series were plotted and trend analysis was performed by employing the first-order autocorrelated trend model on the moving average values for every station. The initial and final values of either the moving average or the trend model were used as the estimates of the initial and final WQI at the stations. It was found that Klang River water quality has shown some improvement between 1997 and 2007. Water quality remains good in the upper stream area, which provides vital water sources for water treatment plants in the Klang valley. Meanwhile, the water quality has also improved in other stations. Results of the current study suggest that the present policy on managing river quality in the Klang River has produced encouraging results; the policy should, however, be further improved alongside more vigorous monitoring of pollution discharge from various point sources such as industrial wastewater, municipal sewers, wet markets, sand mining and landfills, as well as non-point sources such as agricultural or urban runoff and commercial activity.
This article describes design and application of feed-forward, fully-connected, three-layer perceptron neural network model for computing the water quality index (WQI)(1) for Kinta River (Malaysia). The modeling efforts showed that the optimal network architecture was 23-34-1 and that the best WQI predictions were associated with the quick propagation (QP) training algorithm; a learning rate of 0.06; and a QP coefficient of 1.75. The WQI predictions of this model had significant, positive, very high correlation (r=0.977, p<0.01) with the measured WQI values, implying that the model predictions explain around 95.4% of the variation in the measured WQI values. The approach presented in this article offers useful and powerful alternative to WQI computation and prediction, especially in the case of WQI calculation methods which involve lengthy computations and use of various sub-index formulae for each value, or range of values, of the constituent water quality variables.
This paper describes the design of an artificial neural network (ANN) model to predict the water quality index (WQI) using land use areas as predictors. Ten-year records of land use statistics and water quality data for Kinta River (Malaysia) were employed in the modeling process. The most accurate WQI predictions were obtained with the network architecture 7-23-1; the back propagation training algorithm; and a learning rate of 0.02. The WQI forecasts of this model had significant (p < 0.01), positive, very high correlation (ρs = 0.882) with the measured WQI values. Sensitivity analysis revealed that the relative importance of the land use classes to WQI predictions followed the order: mining > rubber > forest > logging > urban areas > agriculture > oil palm. These findings show that the ANNs are highly reliable means of relating water quality to land use, thus integrating land use development with river water quality management.
A computer-aided multivariate water quality index is developed based on partial least squares (PLS) regression. The index is termed as the partial least squares water quality index (PLS-WQI). Briefly, a training set was computationally generated based on the guideline of National Water Quality Standards for Malaysia (NWQS) to predict the water quality. The index is benchmarked with the well-established index developed by the Department of Environment, Malaysia (DOE-WQI). The PLS-WQI is a continuous variable with the value closer to I indicating good water quality and closer to V indicating poor water quality. Unlike other conventional indexing methods, the algorithm calculates the index in a multivariate manner. The algorithm allows rapid processing of a large dataset without tedious calculation; it can be an efficient tool for spatial and temporal routine monitoring of water quality. Although the algorithm is designed based on the guideline of NWQS, it can be easily adapted to accommodate other guidelines. The algorithm was evaluated and demonstrated on the simulated and real datasets. Results indicate that the algorithm is robust and reliable. Based on six parameters, the overall ratings derived are inversely correlated to DOE-WQI. When the number of parameter is increased, the overall ratings appear to provide better insights into the water quality.
Microbes in groundwater play a key role in determining the drinking water quality of the water. The study aims to interpret the sources of microbes in groundwater and its relationship to geochemistry. The study was carried out by collecting groundwater samples and analyzed to obtain various cations and anions, where HCO3-, Cl- and NO3- found to be higher than permissible limits in few samples. Microbial analysis, like total coliform (TC), total viable counts (TVC), fecal coliforms (FC), Vibrio cholera (V. cholerae) and total Streptococci (T. streptococci) were analyzed, and the observations reveal that most of the samples were found to be above the permissible limits adopted by EU, BIS, WHO and USEPA standards. Correlation analysis shows good correlation between Mg2+-HCO3-, K+-NO3-, TVC- V. cholerae and T. streptococci-FC. Major ions like Mg+, K+, NO3, Ca2+ and PO4 along with TS and FC were identified to control the geochemical and microbial activities in the region. The magnesium hardness in the groundwater is inferred to influence the TVC and V. cholerae. The mixing of effluents from different sources reflected the association of Cl with TC. Population of microbes T. streptococci and FC was mainly associated with Ca and Cl content in groundwater, depicting the role of electron acceptors and donors. The sources of the microbial population were observed with respect to the land use pattern and the spatial distribution of hydrogeochemical factors in the region. The study inferred that highest microbial activity in the observed in the residential areas, cultivated regions and around the landfill sites due to the leaching of sewage water and fertilizers runoff into groundwater. The concentrations of ions and microbes were found to be above the permissible limits of drinking water quality standards. This may lead to the deterioration in the health of particular coastal region.
In order to evaluate the water quality of one of the most polluted urban river in Malaysia, the Penchala River, performance of eight biotic indices, Biomonitoring Working Party (BMWP), BMWPThai, BMWPViet, Average Score Per Taxon (ASPT), ASPTThai, BMWPViet, Family Biotic Index (FBI), and Singapore Biotic Index (SingScore), was compared. The water quality categorization based on these biotic indices was then compared with the categorization of Malaysian Water Quality Index (WQI) derived from measurements of six water physicochemical parameters (pH, BOD, COD, NH3-N, DO, and TSS). The river was divided into four sections: upstream section (recreational area), middle stream 1 (residential area), middle stream 2 (commercial area), and downstream. Abundance and diversity of the macroinvertebrates were the highest in the upstream section (407 individual and H' = 1.56, respectively), followed by the middle stream 1 (356 individual and H' = 0.82). The least abundance was recorded in the downstream section (214 individual). Among all biotic indices, BMWP was the most reliable in evaluating the water quality of this urban river as their classifications were comparable to the WQI. BMWPs in this study have strong relationships with dissolved oxygen (DO) content. Our results demonstrated that the biotic indices were more sensitive towards organic pollution than the WQI. BMWP indices especially BMWPViet were the most reliable and could be adopted along with the WQI for assessment of water quality in urban rivers.
Multivariate statistical techniques such as hierarchical Agglomerated cluster analysis (HACA), discriminant analysis (DA), principal component analysis (PCA), and factor analysis (FA) were applied to identify the spatial variation and pollution sources of Jakara River, Kano, Nigeria. Thirty surface water samples were collected: 23 along Getsi River and 7 along the main channel of River Jakara. Twenty-three water quality parameters, namely pH, temperature, turbidity, electrical conductivity (EC), dissolved oxygen (DO), 5-day biochemical oxygen demand (BOD(5)), Faecal coliform, total solids (TS), nitrates (NO(3)(-)), phosphates (PO(4)(3-)), cobalt (Co), iron (Fe), nickel (Ni), manganese (Mn), copper (Cu), sodium (Na), potassium (K), mercury (Hg), chromium (Cr), cadmium (Cd), lead (Pb), magnesium (Mg), and calcium(Ca) were analysed. HACA grouped the sampling points into three clusters based on the similarities of river water quality characteristics: industrial, domestic, and agricultural water pollution sources. Forward and backward DA effectively discriminated 5 and 15 water quality variables, respectively, each assigned with 100% correctness from the original 23 variables. PCA and FA were used to investigate the origin of each water quality parameter due to various land use activities, 7 principal components were obtained with 77.5% total variance, and in addition PCA identified 3 latent pollution sources to support HACA. From this study, one can conclude that the application of multivariate techniques derives meaningful information from water quality data.