Water Quality Monitoring Data Research Articles

Prediction of water quality time series data based on least squares support vector machine

Actual water quality monitoring sites due to inadequate and abnormal by the impact to water quality warning has brought new challenges. How do based on limited monitoring data to predict the water quality to address the shortage of river water quality monitoring sites and data coverage of false alarms caused by abnormal, early warning of water pollution are of great significance. According to river water monitoring data, the small sample properties, is proposed based on least squares support vector machine prediction of water quality parameters, this method has strong ability to predict the true value, and the global optimization and good generalization. This method is applied in the river water quality measurement data, after training the LS-SVM model for water quality parameters of water quality monitoring system to predict, in the same sample under the BP network and RBF network prediction. Experimental results show that the small sample case with noise, least squares support vector machine method is better than multi-layer BP and RBF neural network, to better meets the requirements of water quality prediction. Comparing data of the two experiments shows that new module makes the interest mining more effective.

Assessing temporal representativeness of water quality monitoring data

The effectiveness of different monitoring methods in detecting temporal changes in water quality depends on the achievable sampling intervals, and how these relate to the extent of temporal variation. However, water quality sampling frequencies are rarely adjusted to the actual variation of the monitoring area. Manual sampling, for example, is often limited by the level of funding and not by the optimal timing to take samples. Restrictions in monitoring methods therefore often determine their ability to estimate the true mean and variance values for a certain time period or season. Consequently, we estimated how different sampling intervals determine the mean and standard deviation in a specific monitoring area by using high frequency data from in situ automated monitoring stations. Raw fluorescence measurements of chlorophyll a for three automated monitoring stations were calibrated by using phycocyanin fluorescence measurements and chlorophyll a analyzed from manual water samples in a laboratory. A moving block bootstrap simulation was then used to estimate the standard errors of the mean and standard deviations for different sample sizes. Our results showed that in a temperate, meso-eutrophic lake, relatively high errors in seasonal statistics can be expected from monthly sampling. Moreover, weekly sampling yielded relatively small accuracy benefits compared to a fortnightly sampling. The presented method for temporal representation analysis can be used as a tool in sampling design by adjusting the sampling interval to suit the actual temporal variation in the monitoring area, in addition to being used for estimating the usefulness of previously collected data.

Water Environmental Quality Evaluation based on Visualize Classification for Jian-gang Reservoir of South-North Water Transfer Project

Abstract According to the water quality monitoring data for Bashang section of Jian-gang reservoir conveying water for South-North Water Transfer Project from 2006 to 2009, individual evaluation, overall merit and evaluations of flood season and non-flood season were employed for water environmental quality evaluation with visualize classification on the ground of water quality standard classification, the results indicated that the evaluation indexes of flood season were higher than that of non-flood season and overall merit were with class I, thus overall water quality was good enough for satisfying water quality requirements for supplying of South-North Water Transfer Project, the water quality of non-flood season was better than the flood season.

Design of sampling locations for mountainous river monitoring

River water quality monitoring data are essential to water basin development and decision making. The location of sampling stations is one of the most critical factors in monitoring network design. Most previous studies have focused on mathematical aspects of monitoring network design, whereas human activities within a river basin are often not considered comprehensively. In this study, a new water monitoring network design procedure is introduced to identify representative river water quality sampling locations. The new procedure combines river mixing length, human activities, and geographic information systems (GIS) to locate position of sampling points. An implementation of a new potential pollution score (PPS) model of land use helps to classify the importance of each sampling point prior to selecting the most appropriate locations for an entire river system. With the study area, urban and agriculture are mainly concentrated in the downstream area and were the primary non-point pollution sources in this region. Seven sampling points are proposed to monitor river water quality in the study area.

Trend analysis and assessment of water quality and quantity monitoring data in lignite mines of Western Macedonia, Greece

PERSEAS research team in cooperation with Public Power Corporation (PPC-DEH) of Greece, installed, operates and maintains four R.E.MO.S. monitoring telemetric stations (MTS) in the plant of the mines of Kardia and Notio Pedio in Western Macedonia, Greece. Two of the MTS monitor the stagnant surface waters in the bottom-pit of the mines, while the other two monitor ground water in specific well positions. The monitoring parameters are water temperature – Tw , electrical conductivity of water – ECw , pH, water level – H and discharge – Q. This study is focused on: a) The goodness-of-fit test, of the time series data to the normal distribution, using the Kolmogorov–Smirnov (KS) [1], non-parametric test of hypotheses; b) The assessment of the water quality and quantity monitoring daily data, based on approximately 2.5 years of monitoring period (2007–2009); c) The trend existence for the monitoring parameters, according to the method of the least squares estimator, apart from temperatures which present period...

MTBE in California's Public Drinking Water Wells: Have Past Predictions Come True?

Extensive water quality monitoring data in California are now available for methyl tertiary butyl ether (MTBE) that span the past 16 years. The statewide database includes most of the active (and some inactive) public drinking water wells in the state, most of which have been repeatedly sampled for MTBE during 1995 to 2010. The cumulative detection frequency of MTBE in public drinking water wells in California at any concentration level is approximately 1.6% and 0.9% based on a one-detection criterion and two-detection criterion, respectively. The detection frequency of MTBE is approximately two to three times lower at concentrations above the state drinking water standards. Since 1996, the annual detection frequency of MTBE at any concentration level has ranged from 0.4 to 0.9% (one-detection criterion) and 0.2% to 0.7% (two-detection criterion). Despite repeated sampling, MTBE has been detected in only 1 or 2 years for most public drinking water wells that have ever been found to contain MTBE at any concentration level. The annual average detected concentration of MTBE has ranged from 5 to 16 μg/L (arithmetic mean) and 2 to 5.5 μg/L (median), excluding two outlier samples. MTBE detections are more prevalent in select geographic locations, rather than widespread across the entire state. The cumulative detection frequency for MTBE is approximately 1% for wells that belong to very small and small public drinking water systems (i.e., that serve between 25 and 3,300 people) and is approximately 2% for wells that belong to large and very large public drinking water systems (i.e., that serve >10,000 people). No detections of tertiary butyl alcohol (TBA) were found from 1996 to 2000, although few samples of TBA were collected during this time period. From 2001 to 2010, the cumulative detection frequency of TBA in public drinking water wells in California at any concentration level was approximately 1.4% and 0.2% based on a one-detection criterion and two-detection criterion, respectively. The cumulative detection frequency of tetrachloroethylene (PCE) in public drinking water wells in California is at least six times greater than that of MTBE and TBA, and unlike the trends observed for MTBE and TBA, PCE is often consistently detected in the same wells over time. The results of this analysis can be used to assess historical and current detections of MTBE in public drinking water wells in California. Also, if coupled with other chemical and site-specific information, the results can be used to assess potential future impacts of MTBE on drinking water supplies.

Measuring treatment effectiveness of urban wetland using hybrid water quality — Artificial neural network (ANN) model

Constructed wetlands are now commonly used as tertiary treatment for urban stormwater. The wetlands have primary advantage over other forms of treatment as they remove dissolved organics and heavy metals in conjunction with other pollutants. The effectiveness of a wetland is a primary concern for validating its compliance with design objectives and regulatory requirements. The treatment in a wetland is however complex and is dependent on input pollutants, hydraulics, physico-chemical balance and biota within the wetland. Several models are available for wetlands but have limitations in simulating the physico-chemical and biological processes within the wetland. The aim of this paper is to introduce a hybrid modelling approach that involves both a deterministic model and artificial neural network (ANN) for testing the effectiveness of a constructed wetland at Olympic Park, Homebush, Sydney, Australia. This novel approach allows a combination of calibrated water quality and neural based models to predict the water quality from the wetland. The models were calibrated and validated using water quality monitoring data measured for eight months in both influent and effluent streams of the wetland. The calibrated hybrid models were then tested for treatment effectiveness for range of wet, dry and median flows conditions within the catchments. A water quality index was developed and used to quantify the effectiveness of the wetland.

Influence of drought and total phosphorus on diel pH in wadeable streams: Implications for ecological risk assessment of ionizable contaminants

Climatological influences on site-specific ecohydrology are particularly germane in semiarid regions where instream flows are strongly influenced by effluent discharges. Because many traditional and emerging aquatic contaminants, such as pharmaceuticals, are ionizable, we examined diel surface water pH patterns (i.e., change in pH over a 24-h period) at 23 wadeable streams in central Texas, USA, representing a gradient of nutrient enrichment during consecutive summers of 2006 and 2007. The years of our study were characterized by decidedly different instream flows, which likely affected production:respiration dynamics and led to distinctions in diel pH patterns between 2006 and 2007. Site-specific ambient water quality criteria for NH(3) and the aquatic toxicity of the model weak base pharmaceutical sertraline were predicted using continuous water quality monitoring data from the sites. Drought conditions of 2006 significantly increased (p<0.05) diel pH changes compared to high instream flows of 2007,and the magnitude of diel pH variability was most pronounced at nutrient-enriched sites in 2006. Differences in diel pH change patterns between 2006 and 2007 affected predictions of the environmental fate and effects for model weak base pharmaceuticals and NH(3). Overall, site-specific diel pH was more variable at some sites than the difference in mean surface water pH between the 2 summers. Diel pH variability affected regulatory criteria, because 20% of the study sites in 2006 experienced greater than 5-fold differences in National Ambient Water Quality Criteria for NH(3) over 24-h periods. Our study emphasizes the potential uncertainty that diel pH variability may introduce in site-specific assessments and provides recommendations for environmental assessment of ionizable contaminants.

Analysis on Water Quality Change and Influence Factors in the Yitong River Basin

Using survey information about pollution sources and water quality monitoring data, water quality condition was analyzed systematically. The result displayed that water quality in the upper reach was obviously better than that in the middle-lower reaches. Water quality was Inferior Class V and showed worse trend, with serious excess of contents of DO, COD, CODMn, BOD5, NH3-N in the middle-lower reaches. Based on the water quality change trend, influence which affected water pollution possibly on land use, urban life and industry activities were discussed. As application rate and loss of fertilizers and content of NH3-N heightened with the increase of cultivated land area, rural activity was the major factor affecting water quality in village and town. Discharge of living pollution and industrial pollution led to water quality deterioration.

Application of EFDC and WASP7 in Series for Water Quality Modeling of the Yongdam Lake, Korea

This study aims to test the feasibility of combined use of EFDC (Environmental Fluid Dynamics Code) hydrodynamic model and WASP7.3 (Water Quality Analysis Program) model to improve accuracy of water quality predictions of the Yongdam Lake, Korea. The orthogonal curvilinear grid system was used for EFDC model to represent riverine shape of the study area. Relationship between volume, surface and elevation results were checked to verify if the grid system represents morphology of the lake properly. Monthly average boundary water quality conditions were estimated using the monthly monitored water quality data from Korean Ministry of Environment DB system. Monthly tributary flow rates were back-routed using dam discharge data and allocated in proportion to each basin area as direct measurements were not available. The optimum number of grid system was determined to be 372 horizontal cells and 10 vertical layers of the site for 1 year simulation of hydrodynamics and water quality out of iterative trials. Monthly observed BOD, TN, TP and Chl-a concentrations inside the lake were used for calibration of WASP7.3 model. This study shows that EFDC and WASP can be used in series successfully to improve accuracy in water quality modeling. However, it was observed that the amount of data to develop inflow water quality and flow rate boundary conditions and water quality data inside lake for calibration were not enough for accurate modeling. It is suggested that object-oriented data collection systems would be necessary to ensure accuracy of EFDC-WASP model application and thus for efficient lake water quality management strategy development.

Social Disparities in Nitrate-Contaminated Drinking Water in California’s San Joaquin Valley

Background: Research on drinking water in the United States has rarely examined disproportionate exposures to contaminants faced by low-income and minority communities. This study analyzes the relationship between nitrate concentrations in community water systems (CWSs) and the racial/ethnic and socioeconomic characteristics of customers.Objectives: We hypothesized that CWSs in California’s San Joaquin Valley that serve a higher proportion of minority or residents of lower socioeconomic status have higher nitrate levels and that these disparities are greater among smaller drinking water systems.Methods: We used water quality monitoring data sets (1999–2001) to estimate nitrate levels in CWSs, and source location and census block group data to estimate customer demographics. Our linear regression model included 327 CWSs and reported robust standard errors clustered at the CWS level. Our adjusted model controlled for demographics and water system characteristics and stratified by CWS size.Results: Percent Latino was associated with a 0.04-mg nitrate-ion (NO3)/L increase in a CWS’s estimated NO3 concentration [95% confidence interval (CI), –0.08 to 0.16], and rate of home ownership was associated with a 0.16-mg NO3/L decrease (95% CI, –0.32 to 0.002). Among smaller systems, the percentage of Latinos and of homeownership was associated with an estimated increase of 0.44 mg NO3/L (95% CI, 0.03–0.84) and a decrease of 0.15 mg NO3/L (95% CI, –0.64 to 0.33), respectively.Conclusions: Our findings suggest that in smaller water systems, CWSs serving larger percentages of Latinos and renters receive drinking water with higher nitrate levels. This suggests an environmental inequity in drinking water quality.

A global Water Quality Index and hot-deck imputation of missing data

Water is an essential resource for life on Earth and available freshwater resources are emerging as a limiting factor not only in quantity but also in quality for human development and ecological stability in a growing number of locations. Water quality is a significant criterion in matching water demand and supply. Securing adequate freshwater quality for both human and ecological needs is thus an important aspect of integrated environmental management and sustainable development. The 2008 Environmental Performance Index (EPI) published by the Yale Center for Environmental Law and Policy (YCELP) and the Center for International Earth Science Information Network (CIESIN) at Columbia University includes a Water Quality Index (WATQI). The WATQI provides a first global effort at reporting and estimating water quality on the basis of five commonly reported quality parameters: dissolved oxygen, electrical conductivity, pH value, and total nitrogen and phosphorus concentrations. This paper explains the motivation and methodology of the EPI WATQI and demonstrates how hot-deck imputation of missing values can expand its geographical coverage and better inform decision-makers on the types and extents of water quality problems in the context of limited globally comparable water quality monitoring data.

An Expert System Applied in Construction Water Quality Monitoring

Problem statement: An untoward environmental impact of urban growth in Malaysia has been deterioration in a number of watercourses due to severe siltation and other pollutants from the construction site. Water quality monitoring is a plan for decision makers to take into account the adverse impacts of construction activities on the receiving water bodies. It is also a process for collecting the construction water quality monitoring, baseline data and standard level. Approach: In recent years, expert systems have been used extensively in different applications areas including environmental studies. In this study, expert system software -CWQM- developed by using Microsoft Visual Basic was introduced. CWQM to be used for water quality monitoring during construction activities was designed based on the legal process in Malaysia. Results: According to the water quality monitoring regulation enacted in Malaysia, construction activities require mandatory water quality monitoring plans duly approved by Department of Environment before staring activities. CWQM primarily aims to provide educational and support system for water quality monitoring engineers and decision-makers during construction activities. It displays water quality monitoring plan in report form, water sampling location in GIS format and water quality monitoring data in graph. Conclusion: When the use of CWQM in construction water quality monitoring becomes widespread, it is highly possible that it will be benefited in terms of having more accurate and objective decisions on construction projects which are mainly focused on reducing the stormwater pollution.

Non-indigenous macroinvertebrate species in Lithuanian fresh waters, Part 2: Macroinvertebrate assemblage deviation from naturalness in lotic systems and the consequent potential impacts on ecological quality assessment

The biological pressure represented by non-indigenous macroinvertebrate species (NIMS) should be addressed in the implementation of EU Water Framework Directive as this can have a direct impact on the ’naturalness’ of the invaded macroinvertebrate assemblage. The biocontamination concept allows assessment of this deviation from naturalness, by evaluation of abundance and disparity contamination of an assemblage. This study aimed to assess the biocontamination of macroinvertebrate assemblages in Lithuanian rivers, thereby revealing the most high-impact non-indigenous species, and to explore the relationship between biocontamination and conventional metrics of ecological quality. Most of the studied rivers appeared to be impacted by NIMS. The amphipods Pontogammarus robustoides, Chelicorophium curvispinum and snail Litoglyphus naticoides were revealed as high-impact NIMS for Lithuanian lotic systems. Metrics of ecological quality which largely depend upon the richness of indicator taxa, such as the biological monitoring working party (BMWP) score and Ephemeroptera/Plecoptera/Trichoptera (EPT) taxa number, were negatively correlated with biocontamination, implying they could provide unreliable ecological quality estimates when NIMS are present. Routine macroinvertebrate water quality monitoring data are sufficient for generation of the biocontamination assessment and thus can provide supplementary information, with minimal extra expense or effort. We therefore recommend that biocontamination assessment is included alongside established methods for gauging biological and chemical water quality.

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.

Assessment of the water quality of a river catchment by chemometric expertise

AbstractThe present paper deals with the successive application of self‐organizing map (SOM) classification and Hasse diagram technique (HDT) as chemometric tool for assessment of river water quality. The study is carried out by using long‐term water quality monitoring data from the Struma River catchment, Bulgaria. The advantages of the SOM algorithm for advanced visualization and classification of large data sets are used for proper selection of chemical parameters being most effective in quality assessment. The proper variable selection combined with some state directives for surface water quality parameters was then used for performing a new data classification separating the objects of interest (sampling sites) into specific patterns. The simultaneous application of the SOM methodology and partial order theory allows to visualize the spatial and temporal evolution of water quality parameters. Thus, it can be seen that the nitrate loads are decreasing with time and the high specificity of a certain sampling station with respect to its water quality data pattern is changing. Copyright © 2010 John Wiley &amp; Sons, Ltd.

Modeling of Phosphorus Loads in Sugarcane in a Low‐Relief Landscape Using Ontology‐based Simulation

Water flow and P dynamics in a low-relief landscape manipulated by extensive canal and ditch drainage systems were modeled utilizing an ontology-based simulation model. In the model, soil water flux and processes between three soil inorganic P pools (labile, active, and stable) and organic P are represented as database objects. And user-defined relationships among objects are used to automatically generate computer code (Java) for running the simulation of discharge and P loads. Our objectives were to develop ontology-based descriptions of soil P dynamics within sugarcane- (Saccharum officinarum L.) grown farm basins of the Everglades Agricultural Area (EAA) and to calibrate and validate such processes with water quality monitoring data collected at one farm basin (1244 ha). In the calibration phase (water year [WY] 99-00), observed discharge totaled 11,114 m3 ha(-1) and dissolved P 0.23 kg P ha(-1); and in the validation phase (WY 02-03), discharge was 10,397 m3 ha(-1) and dissolved P 0.11 kg P ha(-). During WY 99-00 the root mean square error (RMSE) for monthly discharge was 188 m3 ha(-1) and for monthly dissolved P 0.0077 kg P ha(-1); whereas during WY 02-03 the RMSE for monthly discharge was 195 m3 ha(-1) and monthly dissolved P 0.0022 kg P ha(-1). These results were confirmed by Nash-Sutcliffe Coefficient of 0.69 (calibration) and 0.81 (validation) comparing measured and simulated P loads. The good model performance suggests that our model has promise to simulate P dynamics, which may be useful as a management tool to reduce P loads in other similar low-relief areas.

Estimation of pollutant loads in an estuarine reservoir considering pollution source characteristics and seasonal variation

The Total Maximum Daily Load (TMDL) program is an integrated process of watershed assessment and management to address surface water quality impairment. The management of organic contaminants and nutrients is a primary concern in conserving surface water bodies. Watershed-scale pollutant loads simulation can assist stakeholders and watershed planners in making decisions on immediate and long-term land use schemes to improve water quality. However, the behavior of contaminants in a watershed needs to be characterized prior to such model applications. The objectives of this study were to characterize point and nonpoint pollutants runoff at a watershed scale and to develop a Pollutant Load Calculation Model (PLCM), which facilitates the estimation of pollutant delivery to a watershed outlet. The developed model was applied for the six sub-watersheds of the Saemangeum estuarine watershed in Korea, where a large tidal reclamation project has been underway. Two years stream flow and water quality data were used for the model calibration, while 1 year data were utilized for the model validation. The model calibration resulted in the R 2 values of 0.58, 0.53, and 0.35 for BOD, TN, and TP, respectively. Overall performance for the validation period was similar with that for the calibration period although the R 2 values were slightly decreased. The PLCM tends to substantially under or overestimate delivery pollutants loads during the summer rainy seasons when most rainfall events occur. This is probably because once-a-month-measured water quality data, which might not represent appropriately monthly water quality, particularly, for rainy seasons, were used for the loads calculation. Thus, more frequently monitored water quality data should be used for the delivery loads estimation at least for a rainy season in order to improve the PLCM performance. Nevertheless, the developed model took the pollutant reduction process into account, which is not allowed with the conventional unit loading method, and furthermore temporal variations of pollutant loads based on stream flows were also incorporated into the pollutant loads estimation. The developed PLCM can be a useful tool to assess pollutants delivery loads at a watershed scale and thus assist decision makers in developing watershed pollution management schemes.

Feasibility of reverse osmosis desalination of brackish agricultural drainage water in the San Joaquin Valley

The technical feasibility of reverse osmosis (RO) desalination of agricultural drainage (AD) water in California's San Joaquin Valley (SJV) was evaluated based on systematic analysis of water quality monitoring data and field water desalting tests in a laboratory plate-and-frame RO (PFRO) system. Thermodynamic solubility analysis and diagnostic PFRO desalting tests served to determine the feasible range of water recovery limits and to assess the mineral scaling potential. Analysis of the recovery limits imposed by scaling due to sparingly soluble salts (e.g. calcite, gypsum, silica) suggested feasible recoveries in the range of 46%–69%. Diagnostic PFRO desalting tests with five representative field water sources from the SJV (having gypsum and calcite saturation indices in the range of 0.12–1.03 and 2.9–9.5, respectively) confirmed the above recovery range. Mineral scale coverage was consistent with the observed flux decline. Deployment of RO technology for treatment of brackish SJV AD water would require site-specific process optimization given the geographic and temporal water quality variabilities. Therefore, RO operation with variable feed water quality (with respect to salinity and scaling propensity) and at sufficiently high recovery would require effective plant control, enabled by real-time mineral scale detection and adaptable process operation to mitigate mineral scaling.

Biocalcification in the Eastern Oyster (Crassostrea virginica) in Relation to Long-term Trends in Chesapeake Bay pH

Anthropogenic carbon dioxide (CO2) emissions reduce pH of marine waters due to the absorption of atmospheric CO2 and formation of carbonic acid. Estuarine waters are more susceptible to acidification because they are subject to multiple acid sources and are less buffered than marine waters. Consequently, estuarine shell forming species may experience acidification sooner than marine species although the tolerance of estuarine calcifiers to pH changes is poorly understood. We analyzed 23 years of Chesapeake Bay water quality monitoring data and found that daytime average pH significantly decreased across polyhaline waters although pH has not significantly changed across mesohaline waters. In some tributaries that once supported large oyster populations, pH is increasing. Current average conditions within some tributaries however correspond to values that we found in laboratory studies to reduce oyster biocalcification rates or resulted in net shell dissolution. Calcification rates of juvenile eastern oysters, Crassostrea virginica, were measured in laboratory studies in a three-way factorial design with 3 pH levels, two salinities, and two temperatures. Biocalcification declined significantly with a reduction of ∼0.5 pH units and higher temperature and salinity mitigated the decrease in biocalcification.