This is to certify that the thesis entitled "SURFACE WATER QUALITY ASSESSMENT THROUGH ENVIRONMETRICS APPROACH" submitted by me to the Indian Institute of Technology Guwahati for the award of the degree of Doctor of Philosophy is a bona fide work carried out by me . In the present study, various statistical methods were used as instruments for water quality assessment of the Brahmaputra River and its seven tributaries (Baralia, Puthimari, Pagladia, Beki, Manas, Kolong and Kameng River) as well as Deepor Beel, Assam (India). In the second phase, cluster analysis (CA), discriminant analysis (DA) and principal component analysis (PCA) were applied to the observed water quality datasets.
7.5 (a) Disorder indices and Isoinformation lines for EWQI in Deepor Beel 128 7.5 (b) Disorder indices and Isoinformation lines for BOD in Deepor Beel 128 7.5 (c) Disorder indices and Isoinformation lines for COD in Deepor Beel 5 (2d) Disorder 7. indices and Isoinformation lines for EHCI in 128 Deepor Beel.
LIST OF TABLES
INTRODUCTION
The effectiveness of MCDMs has been regularly reported in the fields of water quality monitoring and assessment (Hernandez and Uddameri, 2013; Khan and Maity, 2016) and health risk assessment (Li et al., 2016). Monitoring programs are also needed to determine trends in water bodies on a time scale, as well as for efficient design of water quality monitoring networks (Ozkul et al., 2000). However, there is a significant gap between the information required and the information produced by current water quality monitoring systems (Ozkul et al., 2000).
OBJECTIVES OF PRESENT STUDY
It is a measure of the unpredictability of a random event, or equivalently the average information derived from its occurrence. Reliable information on water quality and identification of pollution sources are essential for the prevention and control of surface water pollution (Simeonov et al., 2003; Shrestha and Kazama, 2007; Bu et al., 2010). It is important to re-evaluate the various components of water quality monitoring and explore and develop methods to optimize between information needs and information collected in an effective and efficient manner.
NEED OF THE STUDY
SCOPE OF THE STUDY
THESIS ORGANISATION
BACKGROUND
Phosphorus) damages a water body from uses such as recreation, drinking, irrigation and fishing (Carpenter et al., 1998). The appearance of trace pollutants and microplastics in the environment poses a serious threat to aquatic environments and human life (Wang et al., 2017). Pesticide residues reach water bodies through runoff, flushing and careless disposal of empty pesticide containers (Konstantinou et al., 2006).
WATER QUALITY MONITORING
The goal of QA is to obtain an assessment of the accuracy and precision of the data.
MULTIVARIATE STATISTICAL TECHNIQUES (MSTs)
- Cluster Analysis
- Discriminant Analysis
- Principal Component Analysis
Hierarchical agglomerative clustering (Ward's method of association (Fig. 2.5) with Euclidean distance as a measure of similarity) is the most common approach and is usually demonstrated by a dendrogram (Fig. 2.6), which provides a visual overview of the processes of clustering (Einax et al., 1998 and McKenna, 2003). The fourth step is to calculate the discriminant functions (DF). 2.2) where i is the number of groups (G), ki is the natural constant for each group, n is the number of objects used to classify a data set into a certain group, 𝑤𝑖𝑗 is the weight coefficient, determined by DA for a given selected parameter (Pj). This function is used to classify the new object into one of the known populations.
STUDIES ON MSTs USED FOR WATER QUALITY ASSESSMENT
Hajigholizadeh and Melesse (2017) assessed the spatial and temporal variations of water quality of the Miami Channel, Kissimmee River, and Caloosahatchee River in South Florida using CA and DA. FA identified the five factors of water quality parameters, which explain the majority of the experimental data. 2018) assessed temporal and spatial variations in water quality of the Jialing River watershed in Guangyuan City, China, using MSTs.
WATER QUALITY INDICES (WQIs)
STUDIES ON DEVELOPMENT OF WQIs
Delphi technique was used for the selection of water quality parameters and was based on water quality data from Oregon's Willamette River watershed. This index was based on the assumption that water quality parameters are randomly variable, and is therefore better treated with probability theory. Furthermore, the D-WQI model was applied to the water quality assessment of the Changjiang River in Sanjiangying.
INDICES FOR HEAVY METALS AND THEIR APPLICATION
The river's HPI varied widely from 3.55 to 388.9 due to the significant seasonality of heavy metal concentrations. The IHO was developed to resolve the shortcomings in the conflicts between HPI and CI and classified water quality with respect to a multiple of the average as criteria. The indices contradicted each other and the results were ambiguous to assess heavy metal contamination levels in the groundwater regime of the active mining area.
INFORMATION ENTROPY
It can be intuitively concluded that the greater amount of information obtained with the knowledge of the outcome means a greater uncertainty associated before its occurrence. Information entropy is a measure of the unpredictability of a random event, or equivalently the average information gained from its occurrence. It can also be defined as a measure of the uncertainty, disorder and diversity of a particular trait.
APPLICATION OF INFORMATION ENTROPY IN WATER QUALITY AND HYDROLOGICAL STUDIES
- Entropy Weighted Water Quality Index (EWQI)
- Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) In recent years, MCDMs have shown efficient performance in the fields such as
- Entropy based Diversity and Disorder Indices
It is measured in bits or nats according to the base of the logarithm used to determine entropy. EWQI is calculated by multiplying the quality rating scale by the entropy weight of the parameter and is classified according to the quality grading scale. Unlike variance, entropy-based diversity indices measure the distribution of the probability density function, independent of mean concentration metrics, and may provide a better measure of variability or uncertainty in a data set (Mishra et al. , 2009).
RECENT LITERATURE ON THE USE OF SHANNON ENTROPY IN WATER QUALITY AND HYDROLOGICAL STUDIES
Shannon entropy was applied to measure the inherent uncertainty of the water quality information. 2014) studied the groundwater quality and its suitability for drinking in the Lenjanat Plain aquifer, Iran. Zahedi (2017) modified conflicts between Drinking Water Quality Index (DWQI) and Irrigation Water Quality Index (IWQI) in the water quality ranking of shared extraction wells using MCDMs.
RESEARCH GAP AND HYPOTHESIS
Most studies using information entropy in the assessment of space-time variability are limited to hydro-meteorological variables. Can information entropy be applied to arbitrary water quality datasets to represent the water quality of a particular water body as well as to understand the random hydrometeorological processes that occur in nature? The results of the experimental research coupled with those of mathematical models will contribute to the refinement of the research questions, methodologies and theory.
DESIGN OF RESEARCH WORK
RECONNAISANCE SURVEY AND DATA ACQUISITION 3.2 STUDY AREA (BRAHMAPUTRA RIVER AND TRIBUTARIES)
- SAMPLING STRATEGIES AND ANALYSIS
- Sampling strategies
- Collection of water samples
- Analytical Procedures
The rainfall distribution in the Kolong river basin follows a typical monsoon pattern with maximum rainfall during monsoon and little rainfall in winter. The total length of the river is approximately 264 km and its drainage basin is approximately 11,843 km2. Major agricultural activities in the catchment are tea plantation and rice cultivation (Khound and Bhattacharyya, 2017).
It was essential during the planning phase to conduct a survey of the study area, noting all sources of waste, all incoming tributaries that could pose a potential threat of pollutants, and all uses and withdrawals of the water. Water samples were collected from well-mixed areas of the rivers and lakes using a weighted bottle sampler. Cleaned bottles with a capacity of one liter, free from dust and dirt, were used to collect water samples.
In preservative samples, nitric acid at 2mL/L was added to make the pH less than or equal to 2 to avoid precipitation of metals and adsorption to the container wall (APHA, 2012). Conservation samples were used for heavy metal analysis and non-conservation samples were used for other water quality parameters. After collecting samples in the field, bottles are placed in an insulated cooler with ice.
All chemicals and reagents used in the analyzes were of analytical grade unless otherwise stated.
IDENTIFICATION OF LATENT POLLUTION SOURCES 3.4 MULTIVARIATE STATISTICAL TECHNIQUES (MSTs)
PHASE 3: SURFACE WATER QUALITY EVALUATION EMPLOYING SHANNON ENTROPY
- ENTROPY WEIGHTED WATER QUALITY INDEX (EWQI)
- RISK ASSESSMENT TO HUMAN HEALTH
- IRRIGATION WATER QUALITY PARAMETERS
- TOPSIS
Once the overall EWQI score is known, it can be compared to the following scale proposed by Wu et al., (2011) to determine how healthy the water is at a given time period. Water supplies with ratings that fall in the good or excellent range can support a high diversity of aquatic life. In addition, the water will also be suitable for all forms of relaxation, including those involving direct contact with the water.
The risk characterization and assessment of heavy metals for human health included the two main routes: ingestion and dermal absorption. The enumeration of potential non-carcinogenic risks was done by Hazard Quotient (HQ), estimated by. A HQ greater than 1 indicates potential concern associated with metals coming from that pathway and the sum of HQs from individual pathway results in the Hazard Index (HI) which represents the total potential (HQingestion + HQdermal) of non-carcinogenic risk of the water source evaluate .
Water quality for irrigation suitability was evaluated using permeability index (PI), Kelley ratio (KR), magnesium absorption ratio (MgR), sodium absorption ratio (SAR), percentage of soluble sodium (SSP) and residual sodium carbonate (RSC). "Alternatives" (sampling locations) and "criteria" (parameters) were specified for both rivers, which would be assigned a ranking according to their pollution status.
The weight of the criteria was calculated on the basis of information entropy techniques as per the following equations
A weighted normalized decision matrix was constructed (𝑉) as follows
The PIS and the NIS of the alternatives were computed as
The closeness coefficient (𝐶𝐶) of each alternative was computed as
The alternatives were finally ranked according to their closeness coefficients
IDENTIFICATION OF IDEAL MONITORING LOCATIONS
- STUDY AREA (DEEPOR BEEL)
- SHANNON’S DIVERSITY INDEX
- ENTROPY BASED DISORDER INDICES
- STATISTICAL SUMMARY OF OBSERVED WATER QUALITY DATA
- GENERAL CHARACTERISTICS OF PHYSICO-CHEMICAL PARAMETERS
- HEAVY METALS AND ASSOCIATED HEALTH RISK ASSESSMENT
- CHAPTER CONCLUSION
- CLUSTER ANALYSIS (CA)
- DISCRIMINANT ANALYSIS (DA)
- PRINCIPAL COMPONENT ANALYSIS (PCA)
- CHAPTER CONCLUSION
- SPATIAL VARIATION OF WATER QUALITY IN TERMS OF
- HEAVY METALS AND ITS INDEXING
- DEVELOPMENT OF ENTROPY WEIGHTED IRRIGATION WATER QUALITY INDEX (EIWQI)
- OVERALL RANKING OF SAMPLING SITES
- CHAPTER CONCLUSION
- IDENTIFICATION OF IDEAL MONITORING LOCATIONS
- SPATIAL VARIATIONS OF DIFFERENT PHYSICO-CHEMICAL PARAMETERS AND EWQI
- SPATIAL VARIABILITY OF HEAVY METALS AND EHCI
- GEOSPATIAL ANALYSIS OF WATER QUALITY VARIABILITY
- RISK ASSESSMENT OF HEAVY METALS ON HUMAN HEALTH
- OVERALL RANKING OF SAMPLING SITES
- CHAPTER CONCLUSION
- CONCLUSIONS
- FUTURE RECOMMENDATIONS
The data of the analyzed water quality parameters along with their statistical summary are described in table 4.1 (a–h) and table 4.2 (a–h). Dissolved oxygen (DO) is one of the most important water quality parameters for surface water quality assessment. Therefore, DA allowed a reduction in the dimensionality of the large data set, defining several indicative parameters responsible for large changes in water quality.
In this 2nd phase of the study, different MSTs were used to evaluate the spatial variations in the surface water quality of the Brahmaputra River and its tributaries. The cumulative sum of PI and NI was calculated to classify the water quality of a sampling site as m-HPI. Under-indexing: The description of water quality required under-indexing of the water quality parameters.
To the authors' knowledge, Misaghi et al., (2017) introduced the first water quality index for assessing river water for irrigation purposes based on the National Sanitation Foundation Water Quality Index (NSFWQI). The TOPSIS array of the sampling sites for the rivers is shown in Fig. Site 2 identifies sites located near small industries and settlements on the periphery of Beel.
31.57% of the sampling sites were of "poor" and "extremely poor" water quality in the summer season. However, the wetland was most polluted in the winter season, with all Beel sites having "poor" or "extremely poor" water quality. TOPSIS rankings of sampling sites in Deepor Beel are shown in Fig.
Application of multivariate statistical techniques in the assessment of surface water quality in Lake Uluabat, Turkey. Application of multivariate statistical analysis in the evaluation of the surface water quality of a tropical river. Multivariate Statistical Techniques for the Evaluation of Surface Water Quality of the Himalayan Foothills, Pakistan.
Multivariate statistical analysis of surface water quality based on correlations and variations in the data set. Assessment of surface water quality using multivariate statistical techniques: A case study of the Fuji River Basin, Japan. Water quality assessment and pollution source apportionment in the Gomti River (India) using multivariate statistical techniques - a case study.
Assessment of physicochemical and microbiological surface water quality using multivariate statistical techniques: a case study of the Wadi El-Bey River, Tunisia. Assessment of surface water quality using multivariate statistical techniques: a case study of Behrimaz Stream, Turkey. Assessment of surface water quality via multivariate statistical techniques: a case study of the Songhua River Harbin Region, China.
Surface water quality assessment using multivariate statistical techniques in the red soil hilly region: a case study of the Xiangjiang watershed, China.
