SYEED Et Al, 2023

Environmental and Sustainability Indicators 18 (2023) 100247
Contents lists available at ScienceDirect
Environmental and Sustainability Indicators

journal homepage: www.sciencedirect.com/journal/environmental-and-sustainability-indicators
Surface water quality profiling using the water quality index, pollution
index and statistical methods: A critical review
M M Mahbubul Syeed a, b, Md Shakhawat Hossain a, b, *, Md Rajaul Karim a,
Mohammad Faisal Uddin a, b, Mahady Hasan a, b, Razib Hayat Khan a, b
a
RIoT Research Center, Independent University, Bangladesh, Block A, Aftabuddin Ahmed Road, Bashundhara RA, Dhaka, Bangladesh
b
Department of CSE, Independent University, Bangladesh, Block A, Aftabuddin Ahmed Road, Bashundhara RA, Dhaka, Bangladesh
A R T I C L E I N F O A B S T R A C T
Keywords: Surface water is heavily exposed to contamination as this is the ubiquitous source for most of the water needs.
Pollution index (PI) This situation is exaggerated by the excessive population, heavy industrialization, rapid urbanization, and
Surface water quality improper sanitation. Comprehensive measurement and knowledge extraction of surface water quality is therefore
Water pollution assessment
pivotal for ensuring safe and hygienic water use. Consequently, surface water quality profiling has received
Water quality monitoring
remarkable academic attention in recent decades that produces an ample amount of research results. This study,
Water quality parameters
Water quality index (WQI) therefore, conducts a comprehensive systematic literature review to summarize and structure the existing
Water quality management system literature and to identify current research trends and hotspots. Reported results suggest that the terrain of fresh
surface water includes 13 distinct water sources that are predominantly used in 5 sectors. These sectors often
cause the water pollution in the form of industrial effluents, agricultural runoffs, and domestic sewage. For
profiling the water quality, around 23 Water Quality Index (WQI) models, and 10 Pollution Index (PI) models are
used in research. These models often use several water quality parameters. This study reports an exhaustive
taxonomy of 69 prominent quality parameters in three categories which will support their adoption for these
models. Finally, the limitations of the current manual water quality measurement approaches are summarized to
propose a set of seven requirements for the tech-intensive water quality profiling research and system
development.
1. Introduction becoming even more pivotal than ever before for the sustainability of
human civilization (Ahmed 2016). Surface water is therefore the ubiq
Surface water refers to anybody of liquid water found on the Earth’s uitous source for the majority of water needs, including drinking and
surface. This includes, the ocean water and the water deposited in the domestic purposes, industrial and research activities, irrigation and
inland repositories, e.g., rivers, streams, lakes, wetlands, reservoirs and agricultural production, horticulture, livestock farming and aquatic life
creeks (Dooge 2009). Liquid surface water accounts for more than 97% management including fish and fisheries (Ahmed 2016; Behmel et al.
of the Earth’s hydrosphere within which 96% is saltwater in the ocean 2016).
and only 1.1% is the fresh liquid water (Dooge 2009; Ball 2015). Of this However, with the rapid proliferation of population and socio-
1.1% fresh water, 99% is groundwater and only 1% is the fresh surface economic development, the use and scarcity of this limited fresh sur
water (Dooge 2009; Berner and Berner 1996). Nonetheless, fresh surface face water are increasing hastily (Dooge 2009; Najafi Saleh and
water is one of the most indispensable natural elements in shaping the AuthorAnonymous, 2020). A natural consequence of this is the deteri
environment and maintaining various forms of life on this planet orating water quality due to heavy exposure to contamination and
(Ahmed 2016). Fresh water is fundamental for all living organisms, to pollution (Ahmed 2016). This pollution can be due to both the natural
human health, to food production and to most industrial processes and human-related activities (Uddin et al., 2021). Natural factors that
(Ahmed 2016; Nguyen and Huynh 2022). With rapid urbanization, influence water quality are hydrological, atmospheric, climatic, topo
industrialization and agricultural production, the fresh surface water is graphical and lithological (Magesh et al. 2013; Mahmood 2018). Human
* Corresponding author. RIoT Research Center, Independent University Bangladesh, Block A, Aftabuddin Ahmed Road, Bashundhara RA, Dhaka, Bangladesh.
E-mail addresses: mahbubul.syeed@iub.edu.bd (M.M.M. Syeed), shakhawat@iub.edu.bd (M.S. Hossain), rajaul@iub.edu.bd (M.R. Karim), faisal@iub.edu.bd
(M.F. Uddin), mahady@iub.edu.bd (M. Hasan), rkhan@iub.edu.bd (R.H. Khan).
https://doi.org/10.1016/j.indic.2023.100247
Received 18 January 2023; Received in revised form 6 March 2023; Accepted 17 March 2023
Available online 22 March 2023
2665-9727/© 2023 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M.M.M. Syeed et al. Environmental and Sustainability Indicators 18 (2023) 100247
activities that adversely affect water quality are mining, livestock sectors. These sectors often cause the water pollution in the form of
farming, production and disposal of effluent water (e.g., industrial, industrial effluents, agricultural runoffs, and domestic sewage, among
municipal and agricultural), increased sediment run-off or soil erosion others. There are 23 Water Quality Index (WQI) models, and 10 Pollu
due to land-use change and heavy metal pollution (Yousefi et al., 2018; tion Index (PI) models that are used in research for water quality
Lobato et al. 2015; Sánchez et al. 2007). Colossal contamination and assessment and pollution measurement. Alongside, several statistical
pollution vary based on the establishment and its’ water usage pattern methods are applied for predictive analysis of the water quality. All
(Yousefi et al., 2018). For instance, water bodies close to the heavy in these models follow a four-step evaluation process for the water quality
dustrial zone are highly susceptible to heavy metals and hazardous measurement, that includes, selection of appropriate set of quality pa
substances that are discharged as a by-product of the production pro rameters, determining parameter sub-indices, and assigning relative
cess. The wetlands and water sources surrounding the agricultural lands weights to the parameters, and finally, applying an aggregation function
are exposed to fertilizers and residue of pesticides (e.g., organophos to compute the water quality or pollution index. However, eclipsing
phate, carbamate and organochlorine groups) (Chowdhury, Banik, problem and model uncertainty often lead to inconsistent and inaccurate
Uddin, Moniruzzaman, Karim and Gan 2012b; Syeed et al., 2020). Ac measurement for the models. Alongside, the selection of parameters for
cording to the statistics, one-quarter of the earth’s population has no a specific WQI and PI model is influenced by three critical factors, e.g.,
access to the safe water supply and one-half of the world’s population the natural properties of the parameters, the purpose for which the
has no access to adequate sanitary facilities (Dooge 2009; Ball 2015). water is to be used, and the environmental significance of a water
This adversary leads to the widespread of water-related diseases, quality parameter and the extent to which quality is to be ensured. This
claiming over 5 million deaths per year (Dooge 2009). Therefore, study therefore developed an exhaustive taxonomy of 69 water quality
persistent monitoring and control of the quality for fresh surface water is parameters in these three categories that can be adopted for the models.
considered as a top priority for all the countries in the world (Ly and Finally, the limitations and lack of practical usability of the current
Larsen, 2015; Behmel et al., 2016). Monitoring water quality helps manual water quality measurement approaches are summarized to
countries to assess, predict and control the water pollution and provides proposes a set of seven requirements for the tech-intensive water quality
an evidential means for planning the sustainable use of the water re profiling research and system development. To the best of our knowl
sources (Turner et al., 2009; Firoz 2007; Mama et al. 2021). However, in edge this is the first comprehensive review on the topic that binds
the recent times, developing countries have faced significant problems together all necessary perspectives of the water quality assessment for
in preserving the water quality when trying to improve water supply and an overarching understanding.
sanitation (Mama et al., 2021; Carvalho et al., 2011; Debels et al. 2005). This paper is organized as follows, in Section 2 the research rationale
Even developed countries are facing challenges to sustain their water in relation to the objectives and related research is presented, Section 3
quality due to nutrient enrichment and eutrophication issues (Ortega details the manifestation of the SLR research method for this study.
et al. 2016; Pham et al. 2022). Exclusive findings and observations are categorically reported in Sec
Since water is the key factor for the sustainability of human civili tions 4, 5 and 6. Section 7 details the recommendation and future
zation and the earth’s environment, researchers have carried out research work. Finally, the validation arguments of the study and
extensive research in favor of the maintenance and management of the concluding remarks are drawn in Section 8 and 9, respectively.
fresh surface water quality (Dooge 2009; Uddin et al., 2021). Conse
quently, plenty of research was conducted in the last two decades for 2. Research rationale
profiling the water quality to understand the overall health of an
ecosystem and the condition of the surface water (Drasovean and Mur Since the water quality measurement research has become prevalent,
ariu 2021). Other research works evaluated the water quality by there is a need to analyze the existing literature for revealing the do
leveraging distinct Water Quality Index (WQI) models (Uddin et al., main’s intellectual structure and to identify the critical research gaps
2021; Lumb et al., 2011), Pollution Index (PI) models (Kurnaz et al., (Akter et al. 2016; Bartram et al., 2001; Bo et al. 2022). There have been
2016; Kumar et al., 2019) and Statistical methods for periodic assess a few systematic literature reviews on the related topic to date that
ments and time-series analysis (Nguyen and Huynh 2022; Balla et al. investigate the WQI and PI models, their applications, and performance.
2022; Schreiber et al. 2022). These models and methods were populated For example, in (Uddin et al., 2021) a comparative discussion on the
with several water quality parameters that signify distinct characteris most prominent WQI models, their structure, the process of parame
tics of the water body (Yousefi et al. 2017; Schreiber et al., 2022; Parmar terization and model conceptualization is presented. Furthermore, the
and Bhardwaj 2014). Along this direction, research works were also issues concerning model performance and future research directions are
targeting tech-intensive system design and development for remote highlighted. Similar investigation is conducted in (Lumb et al., 2011;
sensing, measurement and monitoring of the surface water quality Poonam et al., 2013) and (Patil et al., 2012). Other studies have
(Drasovean and Murariu 2021; Islam et al., 2020). reviewed different methods (e.g., Hyperion, WQI and PI) in measuring
Largely, this plethora of research have produced significant results the lake water quality, and recommend that the combination of pollu
that require a systematic synthesis for an in-depth comprehension of the tion prevention, water re-use and recycling approach would be effective
surface water quality monitoring and management for practical use. In for the quality assurance (Schreiber et al., 2022; Bhateria and Jain
this research, an exhaustive systematic literature survey aka SLR has 2016). Alongside, the impact of physio-chemical parameters in deter
been conducted to summarize and structure the existing body of mining the surface water quality is discussed in (Patil et al., 2012). Few
knowledge on the concerned domain (Kitchenham et al. 2010). The of the studies reviewed the applicability of PI models and their effec
objective and thus the contribution of this study is to (a) draw the ho tiveness in measuring heavy metal (HM) and metalloid Pollution (Kar
listic landscape of the fresh surface water, their usage pattern and source ami et al., 2012; Sarkar et al., 2019; Prathumratana et al., 2008; Islam
of pollution, (b) derive a three-dimensional classification of the water et al. 2018; Hasan et al., 2019; Parvin et al., 2022). Finally, review on
quality parameters pertaining to their natural and indicative properties the suitability of AI and ML based models in evaluating surface water
and the usage of water, (c) develop encyclopedic documentation of the quality were conducted in (Lowe et al., 2022; Chen et al. 2020) and
water quality assessment models and statistical methods in relation to (Altalak, Ammad uddin, Alajmi and Rizg 2022). The findings suggest
the parameters, water sources, usage pattern and model performance that the ANFIS and ANN models perform the best in predicting the water
and finally, (d) shed light on the future research concerning model quality.
performance and tech-intensive water quality profiling. Although these reviews yield new and important insights, they are
Reported results suggest that the terrain of fresh surface water in mostly confined within a narrow domain of surface water quality mea
cludes 13 distinct water sources that are predominantly used in 5 surement and do not attempt to identify the most influential
2
contributions over a longer time frame. These studies fall short of por Table 1
traying an overarching comprehension of the surface water quality Study research questions and objectives.
monitoring process that connects all the required perspectives, e.g., an Research Question Addressed Objective
in-depth understanding of the water sources in relation to their usage In
and pollution pattern, providing a detailed taxonomy of the quality [RQ1.] What is the landscape of Section 4 To draw a holistic
assessment parameters and associated models in determining their in the fresh surface water in categorization of the fresh
fluence and contribution with respect to the water source, water use, relation to their sources, surface water in order to
geo-location and pollution. As a result, it is necessary to review the intended usage and the origin of comprehend the relationship
contamination? among the water sources, their
literature to reveal the current research foci, trends and hotspots to have usage and pollution pattern.
a comprehensive understanding of the surface water quality measure [RQ2.] What taxonomy of water Section 5.1 To develop an exhaustive three-
ment, monitoring and management process. quality parameters can be dimensional classification of
To fill the knowledge gap, this study leveraged the bibliographic derived in relation to their the water quality parameters
natural factors (e.g., physical, pertaining their usage or
literature review method for a rigorous quantitative and qualitative
chemical and biological influence in measuring the
analysis of the reported research at the intersection of surface water properties)? water quality.
landscape, water quality parameters and quality assessment approaches [RQ3.] What taxonomy of water Section 5.2
(e.g., methods, models and technologies) (Wanyama et al., 2022). It is quality parameters can be
argued that this study made several contributions to the existing liter derived for measuring water
quality from the perspective of
ature by examining the fresh surface water sources to define their usage its’ intended use?
pattern and the instigation of pollution. Then, provides a [RQ4.] What taxonomy of water Section 5.3
three-dimensional classification of the quality parameters pertaining to quality parameters can be
their natural and indicative properties and the usage of water in deter derived in relation to their
specific quality indicative
mining the water quality. Then, develops an encyclopedic documenta
properties?
tion of the water quality assessment models (e.g., WQI and PI models) [RQ5.] What are the Water Section 6.1 To materialize the taxonomy of
and statistical methods in relation to the parameters, water sources, Quality Index (WQI) models Section 6.2 water quality measurement
their usage pattern, and model performance. Finally, the gaps in current and PI Models used to measure models and associated
research are documented through an exclusive culmination of the re the water quality? statistical methods for
[RQ6.] What are the statistical Section 6.3 monitoring the water quality
ported results to suggest areas for further investigation. Two potential
methods explored for both in spatial and temporal
research directions are detailed, namely, the requirements for the design optimizing WQI model regional variations.
and development of a tech-savvy surface water quality profiling systems uncertainty, parameter
and the amelioration of the WQI and PI models in relation to their variability and predictive time-
series analysis of the water
performance and applicability. For fine-grained assessment and inves
quality?
tigation, a set of research questions analogous to these objectives are [RQ7.] What are the potential Section 7 To shed light on the future
defined. Table 1 and Section 3.1 detail these research questions. research directions that can be research concerning the
explored in relation to the performance of the WQI models
3. Methodology methods, models and tech- and PI models and derive the
intensive management of requirements for a tech
surface water quality? intensive system with AI
Evidence-based research relies on the aggregation of the best in integrated geo-tagged
stances of prior research for evaluating and interpreting all available intelligence for remote sensing,
research results relevant to a particular research question, or topic area, profiling and management of
the surface water quality. The
or phenomenon of interest (Syeed et al., 2013; Robinson et al., 2021;
system should facilitate
Kitchenham et al., 2010; Kitchenham and Charters, 2007). A prevalent automatic assessment,
research methodology for such research is the Systematic Literature predictive forecasting and
Review, predominantly abbreviated as SLR (Syeed et al., 2013; Keele derive factual observations to
et al., 2007; Kitchenham 2004). Conducting a SLR involves several plan for sustainable
management of water
discrete activities that should be adequately defined and must be
resources.
accomplished in an orderly manner for a through, impartial and fair
synthesis of the existing research (Kitchenham et al., 2010; Kitchenham
2004). In SLR, it is also possible to deploy meta-analytic techniques to 3.2. Article selection and quality assessment
detect real cause and effects in the research for drawing valid conclu
sions, which is otherwise, left unnoticed (Kitchenham and Charters, The article selection process is intended to identify those primary
2007; Petticrew and Roberts 2008). Following the guidelines for con studies (i.e., research articles) that provide direct evidence about the
ducting a comprehensive SLR (Robinson et al., 2021; Syeed et al., 2013; research questions (Kitchenham et al., 2010; Syeed et al., 2013). In order
Kitchenham and Charters, 2007; Keele et al., 2007), this study adopted to ensure the comprehensiveness of the collected articles and to reduce
the review process sketched in Fig. 1. This process is detailed in the the likelihood of bias, a well-defined strategy should be adopted
subsequent sections. (Wanyama et al., 2022). Therefore, following the recommendations this
study executed a five step process as defined below (Kitchenham et al.,
3.1. The review objectives and the research questions 2010; Wanyama et al., 2022).
The starting point of a review is to define a set of explicit research 3.2.1. Listing the digital libraries
objectives, which is already reported in Section 2. The next step is to The selection of appropriate digital libraries and associated search
bind these objectives against a set of research questions (RQs) for fine engines are pivotal for ensuring the authenticity of the articles (Kitch
grained assessment of the review articles. Table 1 defines the research enham and Charters, 2007). This selection process is influenced by the
questions and their mapping with the objectives that they address. reputation of the libraries, the scientific content they publish and the
relevance to the objective of this research. At the end of this process, five
libraries and one search engine are nominated, a list of which is pre
sented in Table 2. While searching with keywords in each of these
3
Fig. 1. Overview of the SLR methodology used in this research.
3.2.3. Inclusion criteria

Table 2
As reported in the literature, automated keyword search frequently
Digital libraries.
delivers a deceptive list of articles due to several deficiencies (Kitchen
Digital Libraries Search Content Search Duration ham 2004; Cornelissen et al. 2009). This includes a lack of a consistent
SpringerLink, ScienceDirect, IEEE, Title, Keyword and 23 years range and standardized set of keywords for article classification and a poorly
MDPI, ACM and Google Scholar? Abstract of each (1999–2022) formulated abstract. Therefore, to ensure the quality and relevance of
Article
the selected articles, explicit inclusion criteria should be defined (Cor
nelissen et al., 2009), an embodiment of which is presented in Table 4.
libraries, only the title, keyword and abstract of the papers are searched. The suitability of the articles is determined against these criteria during
This constraint increases the probability of shortlisting the most relevant the manual selection process.
articles (Kitchenham and Charters, 2007; Petticrew and Roberts 2008).
The search period is kept between January 1999 and April 2022. 3.2.4. Manual selection
As stated above, the articles identified through the automated search
3.2.2. Keywords and search string process might contain irrelevant ones, while some relevant articles
Automatic keyword search is the universally practiced approach to might be missing. Therefore, a manual assessment is conducted to
explore the digital libraries for collecting relevant articles (Kitchenham ensure the quality, relevance and completeness of the articles listed
2004; Beecham et al. 2008). Therefore, a broad automated keyword through automated keyword search (Kitchenham et al., 2010; Petticrew
search is performed to get the initial set of articles. Knowing the fact that and Roberts 2008). To accomplish this, each of the articles is assessed
the construction of search strings varies among the digital libraries, this against the inclusion criteria listed in Table 4. Only the title, keywords
study first defines the search terms (i.e., the keywords) according to the and abstract are assessed and in case of doubt, the conclusion is checked.
research questions and the study objectives (Syeed et al., 2013). Then However, this process is subject to reviewer bias and therefore, requires
these keywords are combined to form the search string by consulting the an impartial assessment by a domain expert external to the review team
guidelines of a specific digital library. The keywords and the generic (Robinson et al., 2021). For this, a professor external to the university is
search string are detailed in Table 3. approached and given the set of selected articles and the selection
criteria for his expert review (Robinson et al., 2021; Cornelissen et al.,
2009). Any disagreement is resolved through discussion.
Table 3 3.2.5. Reference checking and final selection

Set of keywords and the generic search string. Finally, to ensure the inclusion of other relevant but missing articles
(if any), a non-recursive search through the references of the manually
Focus Search Terms/Keywords Generic Search
String selected articles are carried out. This concludes the exhaustive search of
articles that resulted in 127 articles, including 123 journals and 4
Water Quality & Synonyms Set-1 = “Water Quality” or Synonyms Set-
Parameters “Surface Water Quality” or “fresh surface 1
water” or “Water Quality Parameters” or AND
“Water Quality profiling” or “Surface water Synonyms Set- Table 4
Monitoring” or “Surface Water 2 Article inclusion criteria.
Management”. AND
Selection Definition
Methods & Synonyms Set-2 = “Water Quality Models” Synonyms Set-
Category
Models or “Water Quality Index” or WQI or “Water 3
Quality Indicator” or “Water Pollution” or AND Subject Domain The subject area of the articles must unveil a strong focus on the
“Pollution Index” or “Pollution Index Synonyms Set- taxonomy of surface water, parameters, their usage and
Models” or “Statistical Methods” or 4 contamination patterns, detection and profiling of the same by
“Statistical Technique”. leveraging tech-savvy cutting-edge tools, methods and models.
Surface Water Synonyms Set-3 = “Potable Water” or Forums Articles published in referred journals and conferences are
Usage “Drinking Water” or Fisheries or Irrigation included for the review. Technical reports published by the
or “Industrial Use” or “Wastewater designated authorities, e.g., WHO, Water Development Board
management” or “Conservative Use” or (WDB)- Bangladesh, US Environment Protection Agency,
Agriculture or Undertaking or Ecological or Department of Environment (DoE) Bangladesh, Canada Water
“Aquatic Inhibition”. Agency etc. are used for reference. Like most SLRs, books are not
Tools & Synonyms Set-4 = “Remote Sensing” or AI considered for review.
Techniques or “Artificial Intelligence” or ML or Ranking/ For the Journal Q1, Q2, Q3 or Q4.
“Machine Learning” or IoT or “Internet of Quality For Conference CORE A, CORE B and CORE C.
Things” or “Geo Referencing” or GIS or Indexing Articles should be indexed at least in the Web of Science (WoS)
“System Design” or Automation. and/or SCOPUS.
4
conference papers. The categorical collection of the articles along the 2012; Prathumratana et al., 2008; Sarker et al., 2021). Consequently, the
digital libraries are presented in Table 5, a complete list of which is synopsis of the reviewed articles is drawn along this axis and is pre
recorded in the appendix (Table 9). sented in Fig. 3.
According to this figure, the terrain of surface water can be classified
3.3. Data extraction and synthesis in 13 categories relative to the sources (Uddin et al., 2021; Bo et al.,
2022; Sarker et al., 2021; Low et al. 2016; Zhao et al. 2013; Fallah and
In order to answer the research questions, each article is thoroughly Zamani-Ahmadmahmoodi 2017). For example, rivers, wetlands, ponds,
studied and interpreted to extract the most appropriate results and lakes, cascades, canals, wells, streams and others. Water extracted from
discussions relevant to the questions. For ensuring the quality and to these sources are primarily served in 5 sectors, as listed in the middle
reduce the reviewer bias, two independent teams from the author list column of Fig. 3. Among these, the Agriculture sector generally covers
have performed this task in two phases (Kitchenham et al., 2010; the irrigation and animal husbandry (Acharya et al. 2020; Gholizadeh
Kitchenham and Charters, 2007). In the initial phase, the first three et al., 2016). Usually, untreated water from rivers, lakes, ponds and
authors reviewed and recorded the relevant information against each canals are used for irrigation (Low et al., 2016; Whitehead et al., 2009).
research question, and in the later phase, the other three authors verified For farm stock-drinking and wild-life watering purposes, water sources
the collected data to validate the originality and completeness. that are free from excessive dissolved salts, not too turbid, and are not
contaminated/infested with chemical or biological pollutants, is suit
able. The Industrial Complexes are heavily reliant on the rivers for their
3.4. Demography of the collected articles
day-to-day operation (Sarker et al., 2021). Water from the river is
well-harnessed for industrial production (e.g., for steel mills, paper
Demographic assessment on the collected articles are carried out in
mills, manufacturing factories and food processing), hydro-electricity
two axis, namely, (a) through the major primary publication channels
generation and thermal power generation (Prathumratana et al., 2008;
and (b) in accordance with the intensity of the research progress with
Karami et al., 2012; Sarkar et al., 2019). Domestic consumption of
time. Fig. 2a exhibits the count of primary studies in the X-axis against
surface water includes drinking, and household usage, and most of the
the publication’s channels and publication type (e.g., journal and con
water sources are exploited for this purpose depending on their avail
ference) in the Y-axis. There are several prominent publication channels
ability in each geographical location (Balla et al., 2022). However, use of
(i.e., digital libraries) that publish articles on the water quality research,
these sources for human consumption requires great care in its’ treat
with Springer, ScienceDirect and IEEE being the front runner. Of these
ment and conditioning, especially water obtained from rivers and
collected articles, 123 articles (96.8%) are identified as journals,
streams (Adimalla and Qian 2019; Chigor et al. 2012). Fisheries or
whereas 4 papers (3.2%) are published in the designated conferences.
pisciculture is the science and associated engineering process to produce
The identification of journal articles are significantly higher than that of
fish and other aquatic resources for the purpose of providing human
the conferences. The probable reason might be the strict adherence to
food (Amiri et al. 2021). This process often adopts or build custom ca
the quality assessment criteria dictated in Table 4 and the extensiveness
nals, ponds, reservoirs, or lakes to provide ideal environment for fish
of the reported results.
culture (Low et al., 2016). For instance, in Manitoba, Canada, the lake
The selected articles are published over a span of 23 years, between
Winnipeg and lake Manitoba are managed using quotas, mesh size of gill
1999 and 2022. Fig. 2b presents a trend chart revealing the frequency of
nets, seasonal regulation of fishing (Lumb et al., 2006; Amiri et al.,
publications in every two years of intervals, starting from 1999. It can be
2021). On the other hand, in Bangladesh, inland closed water that
observed that most of the articles are published in recent years starting
constitutes 794,361ha of land in the form of ponds, seasonal cultured
from 2011 to 2012 (approx. 70%) and there is a growing tendency in
water body, lakes, Shrimp/Prawn farm and pen Culture, are allocated to
research interest in terms of the number of publications. Additionally,
produce approximately 2,060,408 tons of fish every year (Shamsuzza
this line curve is best fitted with the exponential trendline (as shown in
man et al. 2017). The Undertaking defines a non-consumptive way of
Fig. 2b) that affirm the rise in number of publications and research in
water usage in which the water is still available afterwards for other
terest over the years.
uses. For example, recreational use, food production and transportation
of people and goods (Breen et al., 2018). From a water transport point of
4. The landscape of surface water sources, usage and pollution view, vessels of all kinds, of many different forms and makes, have
navigated on rivers, streams and lakes throughout the ages and become
Surface water quality profiling is one of the high-priority mandates the economically viable primary mode of freightage for businesses. For
in almost all countries, especially in the developing world (Nguyen and instance, in Ohio, USA, the Monongahela and Allegheny rivers are used
Huynh 2022; Khan et al. 2021; Ustaoğlu et al. 2021). In line with this to transport over sixty million tons of raw materials and finished prod
directive, several organizations, and research communities, worldwide, ucts annually (Stickle, 1919).
get affiliated with the process of measurement, monitoring and super Unfortunately, nearly all of these consumer sectors are the cardinal
vision of the surface water quality. The starting point of this process is to source of surface water contamination (Low et al., 2016; Karami et al.,
get a holistic understanding of the surface water with respect to their 2012; Prathumratana et al., 2008). For instance, most of the industrial
sources, intended use and the instigation of pollution (Karami et al., complexes are built along the riverbanks that discharge industrial ef
fluents directly into the rivers either without or partial treatment
Table 5 (D’Agostino et al., 2020; Sarker et al., 2021). Many a times, the
Statistics of the article selection process. magnitude of contamination is severe to the extent that it is becoming a
Digital Library Automated Manual Reference Total serious threat to the environment, aquatic life and outbreak of water
Search Search Check Articles borne diseases (e.g., cholera, diarrhoea, dysentery, hepatitis A) (Orga
SpringerLink 92 26 7 33 nization 2022). For reference, statistical evaluation on the
ScienceDirect 65 16 3 19 physio-chemical properties of water in Turag River, Dhaka, Bangladesh
IEEE Digital 36 14 1 15 reveals that Biochemical Oxygen Demand (BOD), Chemical Oxygen
Library
MDPI 22 11 1 12
Demand (COD), Electric Conductivity (EC), Chlorine (Cl-), Dissolved
Google Scholar 101 39 4 43 Oxygen (DO), Turbidity, and Total Dissolved Solids (TDS) are mainly
(Other) responsible for the pollution, and are caused by the substantial amount
ACM 19 5 0 5 of industrial effluent and toxicological compound discharge (Rahman
Total 335 111 16 127
et al., 2021). Alongside, the production of hydro-electric power impact
5
Fig. 2. Demography of the primary articles.
Fig. 3. Categorization of the surface water in terms of their sources, intended use and the instigation of pollution.
on the environment, particularly interrupts the spawning movements of Alongside, having no or limited management for domestic effluent and
fish in the dam (Roje-Bonacci and Bonacci 2013). sediments, these pollutants often contribute heavily to the pollution of
The agricultural runoffs, excessive and uncontrolled use of pesticides the lake and river water (Gerecke et al. 2002).
and fertilizers often get deposited in the nearby ponds, lakes and wet Due to the causal relationship between the usage and the source of
lands (de Souza et al., 2020). Often, heavy metals and pesticide residues the surface water that leads to its’ contamination, this study further
are traced around the agricultural lands and vegetables in categorizes the research articles along these dimensions. The intention is
sewage-irrigated areas as well (Parvin et al., 2022; Islam et al., 2018). to better understand whether the reported research is analogous to the
Consequently, aquatic organisms and fish species living in these water most frequently used sources and their management. The outcomes are
bodies are contaminated by the heavy metals (Kawser Ahmed, Baki, plotted in Fig. 4a and b, where the earlier shows the frequency of articles
Kundu, Islam, Islam, Hossain et al., 2016; F. Islam et al., 2016). that investigate a specific source of surface water, and the latter, lists the
Fig. 4. Categorization of research articles.
6
frequency of articles that explore a particular usage type and pollution (DO), Potential of Hydroge (pH), Nitrate (NO3− ), Phosphate (PO34− ),
pattern. The findings confirm that river water is studied the most (in 74 Ammonia (NH3), Chloride (CL− ), Hardness, Heavy Metals (HM)) (Uddin
articles), followed by the lakes (in 44 articles), reservoirs (in 17 articles), et al., 2021; Boyacioglu 2010).
ponds (in 15 articles), and streams (in 12 articles). In relation to the The selection of water quality parameters is the most crucial step of
usage of the water sources, it is noted that the agricultural, domestic, the WQI measurement process (Uddin et al., 2021; Boyacioglu 2010).
and industrial sectors are the dominating consumers, with article counts There are no less than 69 quality parameters to choose from, and their
of 79, 67 and 53, respectively. selection is often underpinned by several decisive factors, e.g., (a) the
physical, chemical and biological characteristics of the water to be
5. Water quality parameters and their classification measured, (b) the purpose for which the water is to be used, (c) the
extent to which the quality and purity to be ensured, (d) the environ
Management of surface water quality requires the collection and mental significance of a water quality parameter, and (e) the WQI model
analysis of a large number of water quality parameters. A range of selected and the reasons for selecting it (Bartram and Ballance 1996; of
methods and tools have been developed to determine, measure, evaluate European Communities 2000; Uddin et al., 2021; Boyacioglu 2010).
and synthesize these parameters for the water quality measurement In this section a three-dimensional classification of the water quality
(Uddin et al., 2021). Among these, the Water Quality Index (WQI) model parameters is derived. This classification is analogous to the first three
is the highly appreciated one. This model is one of the 25 environmental factors presented above, namely, according to the natural properties (e.
performance indicators of the holistic Environmental Performance Index g., physical, chemical, and biological characteristics), based on the
(EPI) (Boyacioglu 2010). This index offers a simple, concise and water usage pattern, and according to the water quality indicators.
easy-to-understand method to express the quality of water bodies for
varied water sources, usage and pollution. There are several WQI models
that rely on aggregation functions to express water quality as a single 5.1. Classification based on natural properties
number through the measurement and analysis of large temporally and
spatially-varying water quality parameters (e.g., Dissolved Oxygen Natural properties or factors of surface water define the character
istics of the water and its’ suitability for a specific use, e.g., domestic,
Fig. 5. Taxonomy of the 69 water quality parameters along their Natural factors, e.g., physical, chemical, biological and bacteriological.
7
agricultural, industrial, fisheries and others (Shamsuzzaman et al., 2017). Acidic water contains extra Hydrogen ions (H+), whereas basic
2017). There are approximately 69 distinct parameters that are directly water contains extra Hydroxyl (OH− ) ions (Omer 2019). The pH values
associated with the quality assessment of the surface water and are being range from 0 to 14, within which a value less than 7 defines acidic water,
used by different Water Quality Index (WQI) and Pollution Index (PI) a value of 7 defines pure water and a value greater than 7 indicates base
models. Depending on their natural properties, these 69 parameters can solution (Beutler et al., 2014; Arora 2017). The Alkalinity of water is
be classified along three categories, namely, physical, chemical, and mainly caused by the presence of Hydroxide ions (OH− ), Bicarbonate
biological. Measurement of the parameters along this taxonomy allow ions (HCO 3− ), and Carbonate ions (CO23− ), or a mixture of these two
surface water to be assessed on its’ specific quality aspect (Drasovean ions in the water. The high level of either acidity or alkalinity in water
and Murariu 2021). For example, chemical and physical parameters are may be an indication of industrial or chemical pollution (Omer 2019).
important in the rapid determination of the water quality, while bio Chloride ions (Cl− ) occur naturally in the surface water and are
logical parameters provide an exhaustive and complex analysis of the usually not harmful to human health, except for causing an unpleasant
associated environment (Uddin et al., 2021; Drasovean and Murariu salty taste if found in high concentration (Omer 2019). Alongside,
2021). Fig. 5 demonstrates this taxonomy, where 6 parameters are relatively high Chloride concentration in freshwater (about 250 mg/L or
needed to characterize the biological properties, 10 parameters are more) may indicate pollution due to chloride-containing rock, agricul
required to measure the physical properties, and 53 parameters are tural runoff, and wastewater (Chatterjee 1996).
needed to assess the chemical characteristics of surface water. Nitrogen (N2) is the basic source of nutrients for water inhabitants, e.
g., fish, and smaller organisms. The Organic Nitrogen, Ammonia Nitrogen,
5.1.1. Physical parameters Nitrite Nitrogen, and Nitrate Nitrogen are the four forms of Nitrogen found
Physical parameters of water signify the appearance and physical in the water and wastewater (Beutler et al., 2014). Higher concentration
characteristics of the water. Referring to Fig. 5, there are 10 distinct of Organic and Ammonia nitrogen indicates water contamination due to
parameters in this category, namely, Temperature, Turbidity, Color, Taste, sewage, whereas rapid growth of the algae that degrades the water
and Odor, among others. These parameters can be observed and quality is due to high concentration of Nitrate (Tchobanoglus, Burton
measured without changing the chemical composition of the substance and Stensel 2003). Additionally, drinking water having Nitrate con
(Gorde and Jadhav 2013a; Hussen et al. 2018). Imbalance in physical centration over 10 mg/L cause immediate and severe health risk for
parameters often lead to impurities that are offensive to the sense of infants (Peavy et al., 1985).
sight, taste, or smell, and make the water inappropriate for use (Omer Dissolved oxygen (DO) is a direct indicator of the surface water
2019; Hussen et al., 2018). For example, Turbidity designates the pres quality, and is therefore, one of the key parameters for measuring water
ence of suspended materials such as clay, slit, finely divided organic pollution (Beutler et al., 2014). In general, the higher the concentration
material, plankton, and other inorganic materials in the water (Omer of DO, the better the water quality. Oxygen is slightly soluble in water
2019). Low Turbidity water is clear, while high Turbidity water is cloudy and the actual amount varies depending on pressure, temperature, and
or murky. salinity of the water (Tchobanoglus et al. 2003). DO has no direct effect
Temperature is an influential physical parameter that controls the on public health, but drinking water with very little or no oxygen tastes
palatability, viscosity, solubility, odors, and chemical reactions of the unpalatable to some people. However, fishes and other living organisms
water (Omer 2019). For instance, water at a temperature between 10 (e.g., bacteria, microorganisms) metabolize organic material through
and 15 ◦ C is most palatable for human beings. Alongside, other the consumption of DO (Beutler et al., 2014; Tchobanoglus et al. 2003).
bio-chemical process and properties of water, such as, the sedimentation This process discharges CO2 in the water and reduces the concentration
and chlorination processes, the biosorption process of the dissolved of DO. Therefore, oxygen has to be continuously replaced by natural or
heavy metals, and Biological Oxygen Demand (BOD) are affected by artificial means in the water to maintain the level of BOD
Temperature (Omer 2019; Arora 2017). Inorganic toxic substances (e.g., Metallic and Nonmetallic com
Other important physical parameters relating to the potability of pounds) found in surface water even in trace amounts, pose severe
water is the Taste and Odor (Omer 2019). Bad Odor and Taste are caused health risks (Davis 2010). These substances often occur in water due to
by the foreign substances (e.g., organic or inorganic materials, com industrial discharges, and improper management of sediments and
pounds, and dissolved gasses) that are often discharged by domestic, or hazardous waste (Tchobanoglus et al. 2003). Among the metallic com
agricultural sources (Lin et al., 2018). Finally, the Electrical Conductivity pounds, Arsenic (As) and Chromium (Cr6+) are the acute fatal poisons,
(EC) of water measures the ion concentration in the water body and and Cadmium (Cd), Lead (Pb), Mercury (Hg), and Thallium (T1) may
defines its’ suitability for irrigation and firefighting (Arora 2017; Omer cause chronic diseases (DeZuane 1997; Campanella, Onor, D’Ulivo,
2019). Giannecchini, D’Orazio, Petrini and Bramanti 2016; Das, Dutta, Cervera
and de la Guardia 2007; Lasheen et al., 1990; Organization et al., 1079
5.1.2. Chemical parameters 2020). Within nonmetallic compounds, Nitrates (NO3− ) and Cyanides
Water reacts with several chemical substances to change its’ mo (CN− ) cause chronic effects on the central nervous system and thyroid
lecular structure and form a new compound substance (Beutler et al., (Dojlido and Best, 1993).
2014; Chormey et al. 2018). Consequently, there are several chemical Among other prominent chemical parameters for surface water, the
parameters that play a pivotal role in defining the quality of surface F− ions, Fe, Mn, Cu and Zn are nontoxic. These parameters are often
water. This review identifies 53 chemical parameters as listed in Fig. 5, beneficial for human health and for the growth of plants and animals, if
among which, the pH, Alkalinity, DO, BOD, Chlorine (Cl), Inorganic found in permissible quantities (Beutler et al., 2014; Organization et al.,
Toxic Substances, Fluoride (F− ), Iron (Fe), Manganese (Mn), Copper 1996). For instance, a moderate amount of F− ions in the drinking water
(Cu), Nitrogen (N2) and Zinc (Zn), are the dominating ones (Omer is good for preventing tooth decay (Beutler et al., 2014; Peavy et al.,
2019). The chemical characteristics of the surface water are affected by 1985).
several sources, e.g., through soils and rocks with which thewater has
been in contact, by agricultural and urban runoffs, wastewater dis 5.1.3. Biological parameters
patched by the municipal and industrial waste management system, or Biological factors of water are measured by the presence of pollution
through microbial and chemical transformations (Hussen et al., 2018). indicators of organisms, e.g., Total Germ (e.g., Total Bacteria, Viruses,
Chemical contamination occurring through this process might cause Salmonella spp.), Coliforms (both Fecal and Total), Protozoa and Algae
severe health concerns (Akter et al., 2016). (Wilhm and Dorris 1968; Holcomb and Stewart 2020). These are
One of the most important chemical parameters for water quality is important parameters of water potability. The determination of bio
the pH, which is a measure of how acidic/basic the water is (Arora logical quality follows a microbiological analytic procedure that
8
analyses the samples of water and determines the concentration level genetic information necessary for their reproduction (Peavy et al.,
(Barrell et al., 2000; Champa and Kabir 2018). 1985). Many deadly diseases including infectious hepatitis and polio
Bacteria are the single-celled plants that occur in three basic cell myelitis are due to waterborne viral pathogens (Peavy et al., 1985; Or
shapes and have rapid reproductivity (Beutler et al., 2014; Weibe 2021). ganization et al., 1996; Weibe 2021). Finally, Algae are the microscopic
Several life-threatening waterborne diseases, namely, typhoid and plants with photosynthetic pigments (Weibe 2021; Mara and Horan
paratyphoid fever, leptospirosis, tularemia, shigellosis, and cholera are 2003). They often create taste and odor problems for potable water and
caused by bacteria (Peavy et al., 1985). A special group of bacteria is the cause serious environmental and public health issues (Alley 2007;
coliforms, which is a very important biological and pollution indicator of Viessman and Hammer 1993; Weibe 2021; Mara and Horan 2003).
surface water (Mara and Horan 2003). Due to their presence in the
human intestinal system (e.g., pathogenic coliforms), they are often 5.2. Classification based on intended use
excreted with body wastes to water and sewage (Weibe 2021; Peavy
et al., 1985). Coliform bacteria are aggressive organisms and survive in It is advised that measurement of the surface water quality should
the water longer than most pathogens (Beutler et al., 2014; Weibe depend on the use of the water, the geo-location and the type of water
2021). (Murariu et al. 2019). Thus, potable water must not contain chemicals or
The smallest biological structures are the Viruses that possess all micro-organisms which affect human health (Drasovean and Murariu
Fig. 6. A comprehensive classification of the 69 parameters along 5 fundamental usages of the surface water.
9
2021), whereas water used for agriculture/irrigation purposes should be morphological, physical and chemical properties of (European Com
free from large amount of sodium ions, high concentrations of nitrates munities 2000). Fig. 7 categorically list the parameters that contribute to
and other contaminants (Drasovean and Murariu 2021; Bartram and the estimation of a specific quality indicator cited by the standard
Ballance 1996). The scientific community, therefore, recommends that a frameworks (Bartram and Ballance 1996). A pertinent discussion on this
specific set of parameters should be selected that are subjected to assess classification is presented below.
the quality for a particular use of the water (Gorde and Jadhav 2013b;
Murariu et al., 2019). 5.3.1. Basic parameters
Consequently, this study carried out an in-depth classification of the Water quality parameters include physical, chemical and bacterio
69 parameters along the five fundamental usages of the surface water logical properties, and they are measured based on the desired water
(ref to Section 4 for the usage classification). A visual representation of usage and quality concern. However, according to the experts, there are
this classification is presented in Fig. 6, where each of the parameters is 16 quality parameters, which can be considered as the basic set of in
mapped to one of the usage categories according to its’ dominance. For dicators for ensuring water quality in general. Therefore, these param
example, the parameter pH influences all the usage categories, whereas eters are often applied to the domestic, agriculture, industrial, fisheries
Coliform only contributes to the quality measurement of the domestic and any other type of water quality measurement.
and freshwater (undertaking). Additionally, this figure materializes the
list of parameters required for an exhaustive assessment of the water 5.3.2. Toxic inorganic substances and persistent pollutant
quality in each usage category. For example, 28 parameters are needed To measure the toxicity and intensity of pollution, the inorganic
for Fisheries, for Agricultural purposes 30 parameters should be exploited, substances and persistent pollutant need to be measured (Sarkar et al.,
47 parameters are affiliated with the undertaking and assessment of 44 2019; Hasan et al., 2019). There are 19 parameters classified in metallic
parameters can provide comprehensive analysis of the domestic water, and nonmetallic categories that can be used for this purpose. Inorganic
and finally, 26 parameters can be evaluated for Industrial Water contaminants typically result from the leaching of a contaminated
management. source zone into the surface water, such as waste disposal, industrial
However, in practice, only the basic set of parameters are evaluated effluent disposals, and mine-tailing sites (Gerecke et al., 2002; Tcho
by most of the WQI and PI models. This parameter set includes Tem banoglus et al. 2003). If these substances are found in the water even in
perature, Turbidity, pH, Suspended Solids (SS), Total Dissolved Solids trace amount, they can be a danger to public health (D’Agostino et al.,
(TDS), Faecal Coliforms (FC), Dissolved Oxygen (DO), Biochemical 2020; Water and Organization, 2009).
Oxygen Demand (BOD) and Nitrate Nitrogen (NH3–N) (Omer 2019). The Metallic compounds include some toxic heavy metals, namely,
Usually, the number of parameters varies between 4 and 26, depending Cadmium (Cd), Chromium (Cr), Lead (Pb), Mercury (Hg), Silver (Ag),
on the WQI and PI model selected, or expert opinion on deciding the Arsenic (As), Barium (Ba), Thallium (Tl), and Selenium (Se) (Tchoba
parameter significance, and above all, subject to the availability of data. noglus et al. 2003; Järup 2003). Their toxicity varies from being acute
fatal poisons (e.g., As and Cr6+), to the source of chronic diseases (e.g.,
Cd, Hg, Pb, and Tl) (DeZuane 1997; Campanella et al., 2016; Das et al.,
5.3. Classification based on indicative properties
2007; Lasheen et al., 1990; Organization et al., 1079 2020). The con
centration of these heavy metals can be determined by atomic absorp
Alongside the above categorization, other directives suggest that
tion photometers, spectrophotometers, or inductively coupled plasma
parameters should be classified based on the quality indicators that they
(ICP) for very low concentrations.
pointed to. This classification supports the operational monitoring that
The Nonmetallic compounds includes Nitrates (NO3) and Cyanides
is often based on the measurement of relevant biological, hydro-
Fig. 7. Indicative classification of the water quality parameters.
10
(CN). Cyanide is a rapidly acting and potentially deadly chemical that 6.1. Water quality index (WQI) models
causes oxygen deprivation by binding the hemoglobin sites and prevents
the red blood cell from carrying the oxygen (Davis 2010). This causes a The water quality index (WQI) model produces a number as the in
blue skin color syndrome called cyanosis. It also causes chronic effects dicator of water quality for a given location over time and are based on
on the central nervous system and thyroid (Dojlido and Best, 1993). several water quality parameters that are supplied as input to the model
Nonmetallic compounds can be measured by colorimetric, titrimetric, or (Drasovean and Murariu 2021). Therefore, the expression of a WQI
electrometric methods (Beutler et al., 2014). model transforms a large number of complex water quality parameter
measurements into an easy-to-comprehend information for people to
5.3.3. Oxygen condition understand and act upon (Drasovean and Murariu 2021; Drasovean and
The measurement of Oxygen condition is an indicative measure that Murariu, 2021; Drasovean et al. 2018).
ensures the level of free oxygen present in the water (Rahman et al., The use of such water quality indices for profiling the quality can be
2021; Chapra et al., 2021). The categorical measurement of oxygen level traced back to the mid-1800s (Abbasi and Abbasi 2012). However, WQI
can be evaluated with 4 parameters as listed in Fig. 7. This measurement models have only been developed over the last 60 years with Horton
is often crucial for the survival of fish and other aquatic organisms proposing the first WQI model in the 1960s that uses 10 water quality
(Amiri et al., 2021). parameters (Uddin et al., 2021; Poonam et al., 2013). His model proved
significant for quality measurement for most of the waterbodies (Poo
5.3.4. Nutrients that contribute to eutrophication nam et al., 2013). Later, Brown developed a rigorous version of Hortons’
Nutrients are chemical elements that the plants and animals need to model, named NSF-WQI, with support from the National Sanitation
grow and survive. Two of the most important and abundant nutrients are Foundation (Noori et al. 2019). A panel of 142 water quality experts
the Nitrogen and phosphorus. An overabundance of nutrients, primarily defined the process of parameter selection and weighting for this model
nitrogen and phosphorus in the water body is often harmful, as it starts a (Abbasi and Abbasi 2012). Since then, several other WQI models have
process called eutrophication. This process excels the algal blooms that been developed and put into practice. Table 6 presents an exhaustive list
turn the water green and block the sunlight. The decomposition of dead of the WQI models that are currently in use, worldwide. In this table, a
algae causes bacteria to consume the dissolved oxygen, creating dead total of 23 WQI models are listed along with the executive summary of
zones for fish and other aquatic inhabitants (Pham et al., 2022). their primary properties, e.g., the list of parameters used, the origin and
evaluation criteria, the water body where applied, and the reference for
5.3.5. Pesticides detail implementation of the models. In general, the structure of a WQI
Contamination of surface water due to pesticides (e.g., Chlorpyrifos, model consists of four main steps to be executed in sequence, namely, (a)
Diazinon, Carbofuran, Carbaryl, Malathion, Diazinon, Methoxychlor, selection of the water quality parameters, (b) produce the parameter
DDT, and others) is caused by the persistent chemical erosion from the sub-indices, (c) selection of weights for the parameters and finally, (d)
pesticide production factories, agricultural activities, or from the urban use of an aggregation function to calculate the WQI (Abbasi and Abbasi
use (de Souza et al., 2020). Effects of pesticides on human health varies 2012; Uddin et al., 2021; Poonam et al., 2013; Lumb et al., 2011; Gorde
from mild to chronic, depending on the level of exposure to pesticides and Jadhav 2013a).
and age (Syafrudin et al. 2021). For instance, health effects may range
from mild stinging eyes, rashes, nausea, dizziness to chronic effects, such (Step 1) Select the water quality parameters:
as, cancers, birth defects, reproductive harm, and immunotoxicity (de
Souza et al., 2020). The number of selected parameters for WQI models varies between 4
and 26, with most of the models waged 8 to 11 parameters (Ferreira
5.3.6. Salinity indicators et al., 2011; Lumb et al., 2006; Said et al., 2004; Lumb et al., 2011). As
Salinity defines the dissolved salt content in a water body, identified detailed in Section 5, parameter selection for a WQI model is often
by 5 distinct parameters, e.g., Chlorides (Cl− ), Sulphates (SO24− ), Cal influenced by several factors. Such as, the geo-location, the
cium (Ca2+), Magnesium (Mg2+), and Sodium (Na+). Salinity directly physio-chemical properties of water to be evaluated, eutrophication,
affects the agricultural production, water quality, ecological health of health concerns, oxygen availability, dissolved constituents, the inten
streams, terrestrial biodiversity, soil erosion, flood risk, and irrigation ded use of the water, environmental significance and data availability
(Jóźwiakowska et al. 2020). For instance, with the increase in the con (Debels et al., 2005). To assess these factors for the selection of appro
centration of Cl− ions, the plants get poisoned and die. Also, a change in priate number of parameters, the Delphi Technique is predominantly used
salinity affects the quality of water for irrigation and drinking (e.g., for most of the WQI models (Abbasi and Abbasi 2012; Hsu and Sandford
change in Na+ and MgSO4 causes laxative effect) (Jóźwiakowska et al., 2007). In this technique, interviews or surveys are conducted with the
2020). experts who, based on their experience suggest the appropriate set of
parameters for a given water body to be assessed (House 1989). Apart
6. On the quality assessment models and methods from this approach, there is no other systematic technique developed to
formalize and standardize the parameter selection process for general
Water quality profiling is an effective and economical way to un adoption. Additionally, the WQI models often do not consider the haz
derstand the overall condition of the surface water and the health of an ardous parameters and toxic or radioactive constituents to evaluate the
ecosystem (Drasovean and Murariu 2021). Profiling allows to monitor water quality (Hernando et al. 2006; Parvin et al., 2022; Tripathee et al.
the surface water quality both in spatial and temporal regional varia 2016; Karami et al., 2012). The default list of parameters and their se
tions (Gorde and Jadhav 2013b). This quality profiling is underpinned lection process for each WQI model is summarized in Column 3 of
by the water usage pattern, water sources, geo-location, and the extent Table 6.
to which the quality and purity need to be measured (Balla et al., 2022).
Considering these factors, several quality assessment models are pro (Step 2) Produce parameter Sub-indices:
posed and put into practice (Schreiber et al., 2022; Parmar and Bhard
waj 2014). These models can be broadly classified along two axes, In this step, the measured values for the parameters are converted to
namely, the WQI models, and the PI models. Alongside the contempo dimensionless and unit less quantities, known as the parameter sub-
rary Statistical methods are also evaluated for predicting the changes in indices (Abbasi and Abbasi 2012). There are several procedures avail
water quality in space and in time. able for calculating the sub-indices, namely, the Parameter Concentra
tions (Abbasi and Abbasi 2012), Linear Interpolated Functions (Noori
11
Table 6 Table 6 (continued )

Water Quality Index (WQI) Models for Spatial and Temporal Water quality No WQI Model Name Parameters Origin Usage
profiling.
(25–50), very bad
No WQI Model Name Parameters Origin Usage (0–25).
1 Universal Water 13 Parameters: South Africa. Rivers and 7 West Java Index 13 parameters: Indonesia. River. Other
Quality Index pH, Turbidity, Additive other surface (WJI) (Sutadian Temperature, Adopted contaminated
(UWQI) (Banda NH3, Ca, Cl, Chl- version of water et al., 2018) Total solids, DO, from Storet sources.
and Kumarasamy a, EC, F, CaCO3, different reservoirs COD, CL-, Fecal Index and
2020; Low et al., Mg, Mn, NO3, WQI’s. Coliform, Total WPI.
2016) SO4. Parameter Phosphate, Modified
selection: Expert Nitrates, Mn, Cu, version of
opinion and use. Hg, Pb, Detergent. NSF.
2 (CCME) Canadian Minimum 4 Canada. River, lakes, Water Quality
Council of Parameters. (Any Modified Wells and Levels: Excellent
Ministers of the four). Parameter version of reservoir (90–100), Good
Environment ( Selection: Delphi. BCWQI. (90–75), Fair
Sharma and Water Quality (75–50), Marginal
Kansal 2011; Levels: Excellent (50–25), Poor
Uddin et al., (95–100), Good (25–5).
2017a; Balla et al., (80–94), Fair 8 Almeida Index (Qi 8 Parameters: pH, Argentina. River. Surface
2022) (65–79), Marginal et al., 2022) Turbidity, COD, water source.
(45–65), Poor Fecal Coliforms,
(0–44). Total Coliforms,
3 Universal Water 13 Parameters: South Africa. Rivers and Total Phosphate,
Quality Index pH, Turbidity, Additive other surface Total Nitrates,
(UWQI) (Banda NH3, Ca, Cl, Chl- version of water Detergent.
and Kumarasamy a, EC, F, CaCO3, different reservoirs Technique Used:
2020; Low et al., Mg, Mn, NO3, WQI’s. Delphi.
2016) SO4. Parameter Water Quality
selection: Expert Levels: Excellent
opinion and use. (91–100), good
4 Recreational 10 Parameters: Argentina. Recreational (81–90), medium
water quality pH, Turbidity, Water Source. (71–80), poor
index (RWQI) Detergents, NO3, (<25), poor
(Breen et al., COD, PO4, Total (<70).
2018; Poonam coliforms, Faecal 9 Malaysian Index ( 6 parameters: pH, Malaysia. River, Ponds
et al., 2013) coliforms, Koki et al. 2019) DO, BOD, COD, Most of the
Escherichia coli, Suspended Solids, other surface
Enterococci. NH3–N. water sources.
Technique Used: Water Quality
Delphi. Levels: Parameter
Water Quality based individual
Levels: Excellent rating scale used.
(91–100), Good 10 British Colombia Parameters not Canada. Most of the
(81–90), Medium Index (Lumb et al., specified. Surface water
(71–80), Poor 2011; Poonam Water Quality sources.
(<70). et al., 2013) Levels: excellent
5 Weighted 14 Parameters USA. Drinking Water (0–3), good
arithmetic Water (any parameters). Modified Source (4–17), fair
Quality Index Parameter version of (18–43),
(WQI) (Akter selection: Expert Horton and borderline
et al., 2016; Ewaid opinion and use. NSF Index. (44–59), poor
and Abed 2017) Water Quality (60–100).
Levels: Excellent 11 Smith Index ( Minimum 7 New Zealand. Rivers and
(0–25), Good Smith 1990; Parameters: DO, streams.
(26–50), Poor Poonam et al., BOD, Turbidity,
(51–75), Very 2013) Temperature,
Poor (76–100), Suspended Solids,
Unsuitable for NH3–N, Fecal
drinking (>100). Coliform.
6 National 9 Parameters: USA. River, Streams, Parameter
Sanitation dissolved oxygen Extended Canals, Lakes, Selection: Delphi
Foundation Water (DO), fecal Version of Wetland. technique.
Quality Index coliform, pH, Horton, Water Quality
(NSFWQI) (Noori biochemical Dinius Water Levels: No classes
et al., 2019; oxygen demand Quality Index specified.
Shokuhi et al., (BOD), (DWQI). 12 Horton index ( 8 Parameters: pH, USA. Most of the
2012; Poonam temperature, total Abbasi and Abbasi DO, Specific Con., other surface
et al., 2013) phosphate, 2012) Alkalinity, Cl-, water sources.
nitrate, turbidity, NH3–N, F.
total solids. Coliforms.
Water Quality Other parameters
Levels: Excellent added on expert
(90–100), Good opinion. Water
(70–90), Medium Quality Levels:
(50–70), bad Very good
(91–100) Good
(71–90), Poor
(continued on next page)
12
Table 6 (continued ) Table 6 (continued )

No WQI Model Name Parameters Origin Usage No WQI Model Name Parameters Origin Usage
(51–70), Bad piggery waste

(31–50), Very bad (0–19).
(0–30). 17 Oregon Index ( 10 Parameters: Northern Wetland, River.
13 Dojildo Index ( 19 parameters: Poland. River Water. Cude 2001; Temperature, SS, America. Most of the
Barbulescu et al., pH, DO, BOD, Poonam et al. pH, DO, BOD, Refined other sources.
2021) COD, Cl-, NH3–N, 2013) NH3–N, F. version of
Suspended Solids, Coliforms, T. NSF index.
Total Phosphate, Phosphate, T.
Total Sulfate, Nitrates, T.
Total Nitrates, Nitrogen.
Total Hardness, Technique Used:
Total Nitrogen, Delphi.
Cd, Mn, Zn, Cu, Water Quality
Hg, Pb, Phenols. Levels: Excellent
Used common (90–100), good
monitoring (85–89), fair
parameters. Water (80–84), poor
Quality Levels: (60–79), very
Very clean poor (<60).
(75–100), clean 18 Dinius Index ( 11 Parameters: USA. River.
(50–75), polluted Abbasi and Abbasi Temperature, Modified
(25–50), very 2012; Uddin et al., Color, pH, DO, version of
polluted (0–25). 2021) BOD, Specific NSF.
14 House index ( 9 Parameters: pH, UK. Refined Most of the Con., Alkalinity,
House 1989). DO, BOD, Cl-, version of water sources. Cl-, Fecal
NH3–N, Total SRDD index Coliforms, Total
Coliforms, Coliforms, Total
Temperature, Hardness.
Suspended Solids, Technique Used:
Total Nitrates. Delphi.
Parameter Water Quality
selection: Expert Levels:
opinion and use. Purification not
Water Quality required
Levels: high (90–100), minor
quality (71–100), purification
reasonable quality required (80–90),
(51–70), treatment
moderate quality required (50–80),
(31–50), polluted doutful (40–50).
(10–30). 19 Ross Index (Uddin 4 General USA. River.
15 Environmental 9 Parameters: Northern Most of the et al., 2021) Parameters: Modified
Quality Index ( Suspended Solids, America. other sources. Suspended Solids, version of
Katyal 2011) Color, Cl-, Fecal DO, BOD, NH3–N. NSF.
Coliforms, Total Technique Used:
Phosphate, cd, Zn, Delphi.
Cu, Hg. Classification of
Adopted Delphi result not
method. Water specified.
Quality Levels: 20 Bascaron Index ( 26 Parameters Spain. River, sea.
excellent Uddin et al., 2021) Suggested: Modified Other surface
(90–100), very Temperature, version of water sources.
good (80–89), Color, Turbidity, SRDD.
good (70–79), fair pH, DO, BOD,
(55–69), poor Specific Con., Cl-,
(<55). NH3–N, Total
16 SRDD Index ( 10 Parameters: Scotland. Surface water Coliforms, Total
Uddin et al., 2021) Temperature, SS, sources. Sulfate, T.
pH, DO, BOD, Nitrates, T.
Specific Con., Hardness,
NH3–N, Fecal Detergent, etc.
Coliforms, Total Used Delphi
Phosphate, Total Technique. Water
Nitrogen. Quality Levels:
Technique Used: Excellent
Delphi. (90–100), Good
Water Quality (70–90), Medium
Levels: clean (50–70), Bad
(90–100), good (25–50), Very bad
(80–89), good (0–25)
with treatment 21 Dalmatian Index ( 8 Parameters Southern Sea, Marine,
(70–79), tolerable Abbasi and Abbasi Suggested: Croatia. Coastal.
(40–69), polluted 2012; Uddin et al., Temperature, DO, Modified
(30–39), several 2021) BOD, T. version of
polluted (20–29), Coliforms, T. SRDD.
Phosphate, T.
(continued on next page)
13
Table 6 (continued ) 2013; Abbasi and Abbasi 2012) with some do not use the weighting
No WQI Model Name Parameters Origin Usage technique at all (Lumb et al., 2006; Smith 1990; Barbulescu et al., 2021).
Two methods are used to obtain the weight values, namely, The
Nitrogen.
Parameter
Analytic Hierarchy Process (AHP) method (Sutadian et al. 2018) and
selection: Expert The House Index Method (House 1989). It is advised that the weights for
opinion and use. the same WQI model should be adjusted depending on the model
Used Delphi application to improve the measurement accuracy (Poonam et al., 2013;
Technique.
Gorde and Jadhav, 2013a). Therefore, parameter weights should differ
Categories not
specified. for the same model applied for river and marine waterbodies, respec
22 Liou Index (Uddin 13 Parameters Taiwan. River. tively (Ewaid and Abed 2017). A suggestive guideline for parameter
et al., 2021) Used: weight selection can be found in (Akter et al., 2016). The selection of
Temperature, appropriate parameter weights has a deep influence in measuring the
Turbidity, SS, pH,
DO, BOD, Specific
final WQIvalue (Step 4), therefore, contributes highly to the models’
Con., NH3–N, F. robustness by reducing the uncertainty in the WQI model, and
Coliforms, Cd, Zn, improving the model integrity (Pham et al., 2022). On the contrary,
Cu, Pb. inappropriate weightings affect the model performance adversely (Dash
Parameter
and Kalamdhad 2021).
selection based on
environmental
and health (Step 4) Use of aggregation function to calculate WQI:
significance.
Technique Used: The aggregation function is applied to the weighted sub-indices to
Delphi. Categories
not specified.
produce a single water quality index score (Sutadian et al., 2018). The
23 Said Index (Abbasi 5 Parameters: USA. Sea, Coastal. index score is evaluated against a rating scale to categories/classify the
and Abbasi 2012; Turbidity, DO, water quality. The rating scale is specific to a WQI model and varies
Uddin et al., 2021) Specific Con., F. significantly from model-to-model (Poonam et al., 2013). There are
Coliforms, T.
several aggregation functions that are adopted by the WQI models,
Phosphate.
Parameter namely, Additive Functions, Multiplicative Functions, Combined Aggregating
selection based on Functions, Square Root of the Harmonic Mean Function, Minimum Operator
environmental Function, and Unique Linear/non-linear Aggregation Functions (Abbasi and
significance. Abbasi 2012; Poonam et al., 2013; Gorde and Jadhav 2013a).
Used Delphi
Adoption of these four steps may vary depending on the WQI Models.
Technique. Water
Quality Levels: Therefore, it is advisable to consult the specifications of a WQI Model for
Three the selection of parameters, and adoption of associated weighting pro
classification from cess, the sub-indexing method and the aggregation function (Debels
0 to 3, highest
et al., 2005; Abbasi and Abbasi 2012).
purity (3),
marginal quality
(<2). 6.1.1. Handling WQI model uncertainty
24 Hanh Index ( 8 Parameters: Vietnam. River. All WQI indices are derived based on a mathematical aggregation
Uddin et al., 2021) Turbidity,SS, DO, function and therefore, uncertainty in the model performance is un
BOD, COD, Cl-, T.
avoidable (Lowe et al., 2017). Researchers identified that this uncer
Coliforms.
Parameter tainty in due to all the four steps followed in measuring the WQI indices,
selection based on and the model eclipsing problem.
monitoring data The Eclipsing Problem arises while the WQI model output do not
availability.
reflect the true nature of the water quality parameters (Uddin et al.,
Water Quality
Levels: Excellent
2021). This problem, according to the researchers, is caused by inap
(91–100), good propriate selection of sub-indexing rules or parameter weightings,
(76–90), fair which distort the true relative influences of the parameters, or using
(51–75), marginal inappropriate aggregation functions. Therefore, following actions
(26–50), poor
should be taken to minimize the WQI model uncertainty and the eclipse
(<25).
problem, (a) select the justifiable number of parameters, (b) ensure the
quality of parameter values (Ma et al. 2020), (c) collect comprehensive
et al., 2019; Hossain and Patra 2020; Lobato et al., 2015) and Rating set of data (Ongley and Booty 1999) that is required for the model
Curve Functions (Fulazzaky et al., 2010; Othman and Alaa Eldin 2012). performance, (d) carefully select the sub-indexing rules and weighting
Most of the WQI models use these procedures (Lumb et al. 2000; Abbasi factors that do not conceal the parameter’s importance/influence
and Abbasi 2012; Sutadian et al. 2018), however, few models skip this (Swamee and Tyagi 2000), and finally, (e) select a minimum operator
conversion process, e.g., the CCME model and the Dojildo model aggregation function (Abbasi and Abbasi 2012).
(Glozier et al., 2004; Barbulescu et al., 2021).
6.2. Pollution index (PI) models
(Step 3) Selection of weights for the parameters:
PI models, unlike the WQI models, measure the water quality in
Each parameter of the WQI model is assigned a weight relative to its’ terms of the degree or level of pollution of the surface water bodies
importance and in according to the water quality measurement guide (Karami et al., 2012; Prathumratana et al., 2008; Sarkar et al., 2019). In
lines (Sarkar and Abbasi 2006). Most of the WQI models adopted un general, pollution indices are generated by evaluating the Heavy Metal
equal weighting techniques,meaning, a parameter weight is assigned (HM) concentration in the water bodies or in the sediments (Islam et al.,
based on its’ relative importance in measuring the WQI. However, few 2018). Among the HMs, the Lead (Pb), Iron (Fe), Copper (Cu), Zinc (Zn),
other WQI models use an equal weighting approach (Poonam et al., Manganese (Mn), Nickel (Ni), Cadmium (Cd), Arsenic (As), Cobalt (Co),
14
Mercury (Hg) and Molybdenum (Mo) are the most measured parameters Table 7
for surface water pollution indexing (Parsad and Bose 2001; Kumar PI Models for Spatial and Temporal Water quality profiling.
et al., 2019). HMs are often characterized by their long persistence, No PI Model Parameters Assessment Usage
bioaccumulation and biomagnification in the aquatic compartments. Criteria
Therefore, they have particular significance in causing toxic effects at 1 Heavy metal 10 Parameters: Fe, Five Classes: River,
points far from the source of pollution (Tripathee et al., 2016; Tian et al. pollution index Cu, Zn, Cr, Mn, Co, Excellent sewerage.
2015; Wepener 2012). Being non-degradable, they constantly bio (HPI) Ni, Cd, As and Hg. (<25), Good industrial
accumulate in freshwater bodies (Chowdhury, Jahan, Islam, Monir (Parsad and Bose Parameter (26–50), waste and
2001; Kurnaz selection Depends Poor effluents,
uzzaman, Alam, Zaman, Karim and Gan 2012a) and eventually being et al. on expert opinion (51–75), hospital
adsorbed onto the sediments, causing benthic organisms to devour at 2016; Kumar and use. Very Poor wastes,
varying degree, and subsequently to the food chain (Kawser Ahmed et al., 2019) (76–100), municipal
et al., 2016; Adimalla and Qian 2019). Unsuitable waste and
for drinking recreational
Therefore, PI Models and associated standard guidelines are devel
(>100). operations.
oped primarily for assessing the ecological consequences of the surface 2 Heavy metal Parameters are not Three River.
water pollution, and sediment pollution due to the HMs (Banu et al. evaluation index fixed. HM Classes: Low Contaminated
2013). These PI models can be classified along two broad categories, the (HEI) parameters are (<10), water sources.
Heavy Metal Pollution indices and Sediment Pollution indices. To measure (Kumar et al., selected based on moderate
2019; Kabir et al., need. (10–20) and
the heavy metal pollution, models like Heavy Metal Pollution index 2020) Parameter highly
(HPI), Heavy Metal Evaluation index (HEI), Degree of Contamination selection depends polluted
(Cd) and Toxicity Load, are used. On the other hand, for the sediment on expert opinion (>20)
pollution measurement, the models like, Enrichment Factor (EF), the and use.
3 Contamination Parameters are not Three River.
Geoaccumulation Index (Igeo) and the Anthropogenic Enrichment
Index (Cd) (Kumar fixed. HM Classes: Low Contaminated
Assessment (IAP), are used. These models often calculate the anthro et al., 2019; Kabir parameters are (<1), water sources.
pogenic fraction of HMs in the water bodies (Tian et al., 2015). In et al., 2020) selected based on moderate
Table 7 an encyclopedic documentation of the PI models and Sediment need. (1–3) and
Index models is presented. Parameter highly
selection depends pollution
PI models generate a single index value as a pollution indicator,
on expert opinion (>3)
following the identical approach of WQI models (Poonam et al., 2013). and use.
This index value is then assessed against the standards and ideal values 4 Water Pollution 19 water quality Four Classes: River and
to classify it within a range of highly pollutant to low pollutant water Index (WPI) ( parameters: pH, Excellent Ground Water
Hossain and Patra EC, TDS, Na+, K+, quality
(Parsad and Bose 2001; Kumar et al., 2019). Analogous to the 4 Step
2020) Mg2+, Ca2+, F− , (WPI< 0.5),
WQI model (as discussed above), the first step of this process is to HCO3− , Cl− , good (0.5 >
determine the number of parameters to be evaluated based on specific NO3− , SO2 4, WPI <0.75)
type of pollution to be measured. The second and third step is to calculate Zn2+, Cd2+, P and
the sub-indices and relative weighting factor for each HM parameter, b2+, Cu2+, Ni2+, moderately
Co2+, Total Fe polluted
respectively, based on their relative importance in measuring the
(Fe2++Fe3+). (0.75 > WPI
pollution index. Finally, apply an aggregation function over the indices <1), highly
and weights to calculate the pollution index value (Abbasi and Abbasi polluted
2012; Swamee and Tyagi 2000). The sub-indices and the weight factor (WPI >1).
5 Overall Index of Parameter Five Classes: River.
calculation in the second and third steps are optional and depends on the
Pollution (OIP) ( selection Depends Excellent -
pollution index model under consideration Boyacioglu (2006). For Shukla et al., on expert opinion Class C1
example, the HPI model calculates the sub-indices for each pollution 2017; Sargaonkar and use. (0–1),
parameter and assigns a rating as weight to each of them on their rela and Deshpande Acceptable -
tive significance for calculating the HPI (Ustaoğlu et al., 2021). How 2003) Class C2
(1–2),
ever, other indexing models, such as HEI (Kabir et al. 2020), and (Kumar
Slightly
et al., 2019) use the parameter values directly in the aggregation func polluted -
tion, omitting steps two and three. Class C3
(2–4),
Polluted -
6.2.1. Handling PI model uncertainty
Class C4
Results produced by a PI model might not reflect the actual pollution (4–8),
status due to several considerations. For instance, use of absolute values Heavily
of heavy metal concentrations with equal severity in terms of their polluted -
biological consequences, would lead to model eclipsing problem (Uddin Class C5
(8–16).
et al., 2021). Therefore, to derive conclusive evaluation of the pollution
6 Aquatic Toxicity 14 parameters: pH, Three River and
status when applying a PI model, each HM concentration should be Index (ATI) ( DO, Mn, Ni, F, Cr, P Classes: other water
considered according to its’ expected toxicity (i.e., sub-indexing) and Wepener 2012) b, NH+ 4, Cu, Zn, Suitable for sources for
the overall pollution load that an area is experiencing (through index Orthophosphates, all fish fish.
weighting) (Poonam et al., 2013). K, Turbidity, Total species
Dissolved Salts. (60–100),
Alongside, use of different pollution index models for the same water Suitable only
body could lead to different pollution status and thus ended up with for tolerant
different conclusions (Sutadian, Muttil, Yilmaz and Perera 2016a). This fish species
differentiation is due to the formulation of the specific indices and the (51–59),
Totally
use of distinct aggregation functions. For instance, the aggregation
unsuitable
function for HPI model uses the maximum acceptable value, unit for normal
weightage, the standard permissible value and highest desirable value (continued on next page)
for each HM (Reza and Singh 2010) as an input, whereas the aggregation
15
Table 7 (continued ) Table 7 (continued )

No PI Model Parameters Assessment Usage No PI Model Parameters Assessment Usage
Criteria Criteria
fish life Polluted (4<

(0–50). mCD <8)
7 Nutrient pollution Parameters are not Three River and Strongly
index (NPI) ( fixed. Classes: No Other drinking Polluted (8<
Ustaoğlu et al., pollution water source. mCD <16)
2021) (<1), from Strongly
moderate Polluted to
polluted Extremely
(1–3), polluted
considerable (16< mCD
polluted <32)
(3–6), very Extremely
high polluted Polluted
(>6) (mCD >32)
8 Enrichment factor Parameters are not Seven Generic to all
(EF) (Karaouzas fixed. However, Classes: sources.
et al. 2021) according to some Unpolluted function for the Cd model uses the division between monitored value
guidelines (<1) Slightly and the maximum acceptable value, and includes an additional sub
following HMs are Polluted (1 <
used, Cu, Ni, Cd, EF < 3)
traction of unity (Bhuiyan et al. 2016). Furthermore, pollution index
Cr, P b, As, Hg, Zn, Moderately evaluation criteria vary from model to model with different standards
Al, Fe, Organic Polluted (3 < and ideal value ranges (Kabir et al., 2020). Therefore, the Cd model
Carbon, Li. EF < 5) from appears stricter than HPI and HEI models, and a preferable choice for the
Moderately
evaluation of a study area (Kumar et al., 2019).
to heavily
Polluted (5 <
EF < 10)
6.3. Statistical methods for model optimization
Strongly
Polluted (10
< EF < 25) It is the necessity to apply appropriate statistical methods when
from Strongly analyzing water quality data. These statistical methods provide robust
Polluted to
scientific inference to draw a global vision and pattern of changes in
Extremely
polluted (25
water quality in space and in time (Drasovean and Murariu 2021; Dra
< EF < 50) sovean et al., 2018). This time-series evaluation can assist in
Extremely evidence-based decision making for the water regulatory bodies, pro
Polluted (EF vide actionable advice regarding water management and to draw valid
> 50)
conclusions (Drasovean et al., 953 2019; Nguyen and Huynh 2022). In
9 Geoaccumulation Parameters are not Seven Generic to all
Index (Igeo) ( fixed. Classes: sources. the related research, a wide range of statistical methods are applied for
Karaouzas et al., Unpolluted time series profiling and prediction of the water quality. This includes,
2021) (<0) Slightly for example, Cluster Analysis (CA), Principal Component Analysis
Polluted (PCA), Multivariate Statistical Analysis, Entropy Weighted methods
(0–1)
Moderately
(Nguyen and Huynh 2022), Pearson and Spearman Correlation Co
Polluted efficients (Drasovean and Murariu 2021), Generalized Least Squares,
(1–2) from Linear Mixed and Generalized Linear Mixed-effect model and Bayesian
Moderately Techniques (Parmar and Bhardwaj 2014). Table 8 provides an elabo
to heavily
rated listing of these methods along with the tools and techniques uti
Polluted
(2–3) lized in measuring them for a specific water quality assessment.
Strongly However, the selection of appropriate statistical methods, their
Polluted experimental design and selection of tools are pivotal to achieve accu
(3–4) from racy in the data analytical processes (Iticescu et al. 2019; Drasovean and
Strongly
Polluted to
Murariu 2021). For example, CA can be conducted in order to find the
Extremely associations among the parameters, whereas PCA can assist reducing
polluted certain variables to determine the indices which can describe the vari
(4–5) ation in the water quality data with minimal loss of information (Noori
Extremely
et al. 2010). This process of parameter reduction using PCA is especially
Polluted (>5)
10 Anthropogenic Parameters are not Seven Generic to all useful for those countries where resources for operational water quality
enrichment fixed. Classes: sources. modelling are scarce (Ustaoğlu et al., 2021). Alongside, to draw a linear
assessment (IAP) ( Unpolluted relationship between two parameters of water samples, Pearson Corre
Karaouzas et al., (<1.5) lation Coefficient can be used, while, for non-linear correlation the
2021) Slightly
Polluted
Spearman Coefficient is appropriate (Iticescu et al., 2019). Nevertheless,
(1.5< mCD this selection must accommodate the uncertainty, inconsistency and
<2) variability of the environmental data, and the temporal and spatial de
Moderately pendency structures (Drasovean et al., 2018; Tyagi et al. 2013).
Polluted (2<
Considering the data variability, in (Parmar and Bhardwaj 2014)
mCD <4)
from Generalized Least Squares, Linear Mixed and Generalized Linear
Moderately Mixed-effect models, and Bayesian Techniques are used to achieve
to heavily better accuracy in prediction. Finally, applied models can be validated
using different standard methods, that includes, Root Mean Square Error
16
Table 8 Table 8 (continued )

Multivariate statistical analysis methods. No Methods Tools Used Techniques Research Purpose
No Methods Tools Used Techniques Research Purpose
factors are
1 Cluster IBM SPSS 26. Mean, lower and Change in sewage rotated.
Analysis ( Shapiro–Wilk upper quartile, network over the 5 Regression Not Specified. Calculate the Modeling and
Nguyen and test. Wilcoxon mode, median time. analysis ( correlation and analyzing the
Huynh 2022; signed-ranks and standard Parmar and regression variables.
Ustaoğlu et al., test. deviation. Bhardwaj coefficients. Understanding
2021) Boxplot 2013) Standard the variation in
diagrams. deviation of value of the
2 Principal Not Specified. Five major steps: For reducing variables. dependent
component 1) coding the complexity of Find the expected variable.
analysis ( variables (X1, X2, input variables. value and
Noori et al., …, Xp) to have determine the
2010; zero means and regression line by
Ustaoğlu et al., unit variance; 2) variables
2021) calculate the (dependent and
correlation independent)
matrix R; 3) find 6 Entropy Not Specified. Four steps: Step Correlation with
the eigenvalues Weighted 1: Formulation of the observed
λ1, λ2,…, λp and Method (Dash dataset matrix. dataset and their
the and Step 2: uncertainties of
corresponding Kalamdhad, Normalization of occurrence.
eigenvectors a1, 2021) the dataset
a2,…, ap; 4) matrix. Step 3:
discard any Computation of
components with information
a small entropy and
proportion of the entropy weights.
variation in data Step 4:
sets; and (5) Estimation of
develop the Result.
factor loading
matrix and
perform a (RMSE), R-Squared, Mean Absolute Error (MAE), Maximum Absolute
Varimax rotation Error, Mean Absolute Percentage Error, Maximum Absolute Percentage
on it.
Error, Normalized Bayesian Information Criterion, Ljung–Box Analysis,
3 Canonical Not Specified. Form two An exploratory
correlation canonical tool, used as a Predicted value and Confidence limits (Parmar and Bhardwaj 2014;
analysis ( variables (U = X data reduction Drasovean et al., 953 2019; 2018; Drasovean and Murariu 2021).
Noori et al., and V– –Y), a method.
2010) correlation
7. Pave to the future research
matrix (p + q) ×
(p + q) between
the variables This section opens a broader discussion on the potential future
((X1, X2,…, Xp research directions pertaining to surface water quality measurement and
and Y1, Y2,…, monitoring. In doing so, the reported results, and the suggestions for
Yq)), Calculate
future research are categorically assessed and documented from the
eigenvalue (1 N2
N … Nr) problem reviewed articles. This categorization identified two plausible research
equation from directions, namely, (a) design and development of a tech-savvy surface
the matrix. water quality monitoring, and profiling systems, and (b) amelioration of
Coefficients of
the WQI and PI models in relation to their uncertainty and eclipsing
the canonical
variates are for
issues.
standardized X
and Y variables. 7.1. On the Water Quality Monitoring System (WQMS)
4 Factor Finding Three stages of Explain the
Analysis ( different Factor analysis: correlations
Boyacioglu factors. For all the between the Realizing the severity of water contamination, governments
2006; Kumar variables a observations in throughout the world have issued definite directives for acquiring
et al., 2019) correlation terms of the adequate information about surface water for exhaustive water quality
matrix is underlying assessment and management (Dong et al. 2015; Uddin et al., 2017b;
generated. factors, which are
Factors are not directly
Samad et al., 2013). However, the current process of measuring surface
extracted from observable. water quality mostly requires laboratory-based manual sample collec
the correlation tion and measurement, where samples are collected from the
matrix based on pre-defined points in the rivers and from the effluent treatment plants.
the correlation
This approach is identified as tardy, time-consuming, expensive, ad-hoc,
coefficients of the
variables. error-prone, and untraceable (Bo et al., 2022). It hinders timely mea
To maximize the surement, assessment, decision-making, and long-term planning for
relationship water quality assurance (Geetha and Gouthami 2016). Alongside, this
between some of approach requires an ample amount of expert workforce with logistic
the factors and
variables, the
support, which many of the governments cannot afford and maintain
(Gholizadeh et al., 2016).
Therefore, the design and development of a tech-intensive, autono
mous Water Quality Monitoring System (WQMS) is essential to reflect on
17
the vacuum in current processes and practices. The system should enable integral part of remote sensing (Glasgow et al., 2004; Koparan et al.,
autonomous, safe, timely, and comprehensive water quality data 2020). The WQMS system should have the remote sensing capabilities to
collection through a self-operated process with minimum human su operate in real-time. This enables simultaneous deployment of the sys
pervision and intervention (Kamienski et al., 2019). It should also tem in several strategic hot spots where water quality parameters need
maximize the autonomous Remote Sensing (RS) of geo-tagged sensor to be measured automatically in regular intervals (Karim et al. 2021).
data (Balla et al., 2022; Glasgow et al. 2004) for periodic or real-time The deployed system should be equipped with adequate sensors that are
sensor data logging to the remote cloud server for further processing. configurable remotely for data acquisition (Islam et al., 2020; Kamienski
The server should integrate the statistical methods, AI-driven big data et al., 2019). The sensors should be configured, and polled remotely with
processing, reasoning, prediction and decision support system (DSS) for graphical visualization and assistance, showing all the critical parameter
automated/semi-automated data post-processing, visualization, and set by the organization and providing real-time/periodic baseline and
decision-making (Lowe et al., 2022; Chen et al., 2020). Therefore, this trend analysis. Remote management of physical elements and operation
system can effectively serve the purpose of every concerned stakeholder, of basic units should be part of this vigilance (Balla et al., 2022; Ali and
namely, the government policy makers, the domain experts, the re Qamar 2013). Based on the system settings (e.g., frequency of sensor
searchers, and the management, alike (Lumb et al., 2006). Fig. 8 re polling and data logging, permissible ranges for measured parameters),
capitulates these requirements in terms of system specifications that the system should automatically monitor and generate alarm/notifica
should define the terrain of functionalities and features required for a tion (Islam et al., 2020; Geetha and Gouthami 2016).
smart-integrated WQMS system. An explicit denotation and interpreta
tion of these characteristics are given below which will open the pave for 7.1.3. Remote Data Logging in the cloud server
a tech-intensive water quality profiling research and system All data associated with the water quality parameter measurement
development. (e.g., sensor data, laboratory-tested data) must be logged into the cloud
data server, either in real-time or in regular intervals (Islam et al., 2020).
7.1.1. Geo-Tagged Parameter Measurement This server should collect and store the data and send commands and
Geo-Tagging refers to the amendment of geo-spatial/geographical in configuration to the field data loggers for controlling the sensors. It must
formation with the remote sensing data, e.g., videos/images taken by a expose required APIs (Application Protocol Interfaces) for information
drone, data communication over the smartphone, or data collected retrieval, processing, manipulation, exchange, and modification for
through sensors from a given location, among others. Geo-spatial in monitoring, remote control and decision-making. This integrated system
formation often includes the latitude and longitude coordinates of the model that connects the sensors and the cloud system, is an efficient
geographic location along with the data associated with it (Koparan solution to minimize the computational overhead on the sensors and the
et al. 2020). If needed, geo-tagging may include geocoding. This means micro-controllers (Islam et al., 2020; Kamienski et al., 2019). This
having a text-based description of a location (i.e., street addresses, arrangement increases the power and processing efficiency, extends the
towns, postal zip codes) along with geo-spatial information for better lifespan of the sensors and associated components, and lowers the
comprehending the location. maintenance overhead (Lowe et al., 2022).
Geo-Tagging of the data related to water quality parameter mea
surement and its’ logging to the server, plays a pivotal role in making a 7.1.4. Water quality profiling, Statistical Data Analysis and Prediction
WQMS system highly useable (Bo et al., 2022). It can enhance the sus Water quality profiling is an effective and economical way to un
tainable water resources management and development process through derstand the overall health of an ecosystem and the condition of the
location-based water quality forecasting (Parmar and Bhardwaj 2014; surface water (Drasovean and Murariu 2021). Profiling allows to
Ali and Qamar 2013). For example, based on the geo-location the most monitor of the surface water quality both in spatial and temporal
optimum WQI model, and appropriate set of parameters (their standard regional variations (Gorde and Jadhav 2013b). To assess the water
values, sub-indices and weights) can be selected. Additionally, tailored quality, several Water Quality Index (WQI) models, Pollution Index
monitoring, assessment and supervision can be done based on the need models, and numerous Statistical methods are deployed for periodic and
for a geographic location, and a countrywide taxonomy of water quality time-series analysis of the water quality (Schreiber et al., 2022; Parmar
profiling can be drawn (Balla et al., 2022; Koparan et al., 2020). and Bhardwaj 2014). An exhaustive list of these models and methods is
presented in Section 6.1. The WQMS system should possess a
7.1.2. Remote sensing, and real-time Supervision and Control high-performance cloud server system that should implement all these
Remote Sensing (RS) is the process of data acquisition about an ob models and methods to operate over the logged parameters as per the
ject or phenomenon without making physical contact with the object or requirements (Parvin et al., 2022; Balla et al., 2018, 2022).
on-site observation (Bo et al., 2022; Kamienski et al., 2019). In current
days, the use of sensor technologies and satellite imaging become an 7.1.5. AI Integration for Profiling and Autonomous Decision Making
Recent research identified that the deployment of artificial neural
networks helps in reducing the uncertainty (e.g., eclipse problem)
resulting from the final aggregation process of the WQI and PI models
(Lowe et al., 2022; Altalak et al., 2022). Therefore, AI tools and tech
niques should be pursued to reduce model uncertainties and increase the
accuracy of the final computed indices.
Alongside, integration of Machine Learning (ML) for AI-driven data-
intensive decision-making and controlling is a valuable component for
the WQMS system (Lowe et al., 2022; Altalak et al., 2022). Due to
consistent and continuous logging of water quality data in the cloud
server, a complete and comprehensive database is in place for devel
oping diverse ML algorithms. Therefore, the cloud server should provide
integration of all the necessary tools, libraries and frameworks (e.g., R,
Python Libraries) and allow access to both historical and real-time data
for the development, test and run of the ML algorithms (Chen et al.,
Fig. 8. System Specifications for a smart integrated Water Quality Manage 2020; Wang et al., 2017).
ment System. Additionally, there should be an integrated interface for external
18
users (e.g., researchers, academicians, and policymakers) to deploy to an inappropriate set of parameter selection, or erroneous
scripts and run them periodically to get necessary observations and data sub-indexing rules and parameter weightings (Abbasi and Abbasi 2012),
analytics. The cloud server should offer this interoperability under a or the use of improper aggregation function (Smith 1990). To date, the
given configurations, legislation and collaboration. For instance, Delphi technique is used to define each stage of the WQI model (Sutadian,
exposing authenticated APIs to access data, or exchange data with other Muttil, Yilmaz and Perera 2016b). This technique relies heavily on the
authorities under a given model of collaboration, e.g., with the gov survey-based expert panel opinion for parameter selection, development
ernment departments and research organizations managing GIS of sub-indexing rules and weighting for the parameters. This process is
(Geographic Information Systems) database (Drasovean and Murariu subject to human biases that might suppress the true relative influences
2021). of parameters in calculating the WQI index. Therefore, further research
on this track must define a statistically validated evidence-based
7.1.6. Integration to GEMS approach for each of these stages so that the model uncertainty can be
Water quality assessment data should be logged into the Global minimized.
Environmental Monitoring System (GEMS) (Gwynne 1982) database as a Finally, almost no WQI model consider using the toxicological
benevolent contribution towards the better understanding, management components (e.g., heavy metals and nutrients) as part of their aggre
and protection of the Earth’s environment. Governments all over the gation function for measuring the quality indices, as can be verified from
world are taking serious initiatives towards this direction. Therefore, the Column 3 of Table 6. Therefore, the PI models are being evolved (ref to
WQMS system should expose and implement necessary APIs, authenti Section 6.2) to measure the water quality in terms of the degree/level of
cation mechanism, and protocols that adheres to GEMS guideline for pollution of the surface water (Karami et al., 2012; Hossain and Patra
water quality data preparation and periodic logging to the GEMS server 2020). However, these models need fine-tuning and verification, like the
(Gwynne 1982). WQI models to ensure their consistency and compatibility.
7.1.7. One System Serves All 8. Validation of the study

There is a wide range of stakeholders that require access to the water
quality data and associated analytics for their specific needs. The Gov Carrying out a literature review is mostly a manual task, and subject
ernment regulatory and monitoring organizations (e.g., the Department to interpretation (Kitchenham et al., 2010). Therefore, researcher bias is
of Environment (DoE), Water Development Board, and Public Health the most likely threat to the validity of the reported results (Robinson
Engineering Department) require a comprehensive transcript of water et al., 2021). To minimize this concern, the biblical process for con
quality assessment for consistent monitoring and supervision of the ducting a SLR is consulted and assumed in detail before commencing
same to ensure the needs of potable water, irrigation, health, domestic, this study. In section 3, every step taken for conducting this review is
fisheries and industries. There are other research and educational in documented with justification. For instance, the research questions are
stitutes (e.g., Agricultural Universities, Engineering Universities, the defined based on the review objective, whereas the article inclusion
Agricultural Research Institute, and Fisheries Research Institute, among criteria and search keywords are defined based on the research ques
others) that also require access to these data and associated analytic for tions. Adhering to the inclusion criteria, articles were collected in a
in-depth research and development in the concerned domains. The three-step process: automated keyword search in the digital libraries,
WQMS system must support configurable system architecture that manual selection of the shortlisted articles, and finally, checking the
should meet the needs of these distinct organizations. references for the manually selected ones. This process ensured that the
selected articles were both representative, complete and free form
7.2. On the quality assessment models reviewer bias (Kitchenham and Charters, 2007; Keele et al., 2007).
Furthermore, the review of the selected articles is subjective and
Almost all the research to date, either developed or utilized the WQI susceptible to researcher bias. This might restrain from reporting the
models based on a specific usage, sources or pollution pattern of the true nature of the research results (Kitchenham et al., 2010; Kitchenham
surface water, as detailed in Section 4, and Column 5 of Table 6. 2004). To mitigate this issue, the domain experts cross-checked the
Alongside, from Column 4 of Table 6 it is evident that the model ap obtained data against the reviewed papers and the research questions to
plications are mostly region/site-specific, even though most models are guarantee proper interpretation and presentation (Kitchenham et al.,
theoretically generic such that they are easily transferable to other sites 2010).
(Uddin et al. 2017a, 2017b). Therefore, research should be conducted to
determine which model suits best for which type of water usage and 9. Discussion and conclusion
sources, and how to increase the certainty of model performance in
relation to its’ applicability to the overall water ecosystem and associ Being the ubiquitous source for the majority of water needs the
ated domain in concrete terms (Abbasi and Abbasi 2012). surface water becomes susceptible to significant contamination and
Furthermore, the application of different WQI and PI models on the pollution. Rapid urbanization and industrialization, inadequate sanita
same water body in a given geo-location may produce significantly tion, overuse and inconsistent monitoring, exacerbate the situation. As a
different results, leading to the concern of interpretation, compatibility, result, the topic of surface water quality testing, monitoring, and man
and generalizability of the produced quality indices (Abdul Hameed M agement has received major academic interest in recent decades,
Jawad et al., 2010). Related studies noted that while most models have resulting in an abundance of research. This study therefore, summarized
broadly similar structures (e.g., the 4-step model detailed in Section and analyzed the existing research on the concerned domain using the
6.1), there is very little uniformity among them at the implementation bibliographic SLR approach. This review strived to obtain a compre
level, causing inconsistencies in the produced results (Lowe et al., 2017). hensive understanding of the intellectual structure of the water quality
Therefore, compatibility among the quality indices produced by assessment research. In summary, this study offers several key contri
different models for the same set of parameters and for a given water butions by (a) identifying and discussing the landscape of surface water
source should be a core concern of future research. sources, usage and pollution pattern, and their interdependency, (b)
Several studies highlighted that all four stages of the WQI model (ref derive a detail taxonomy of the water quality parameters concerning
to Section 6.1 for detail) can contribute to eclipsing problem and model their physio-chemical properties and impact on the water sources and
uncertainty. The eclipsing problem arises when the WQI model output usages, (c) revealing comprehensive knowledge clusters on the Water
does not reflect the true nature of the water quality parameters and leads Quality Index (WQI) models, Pollution Index (PI) models and the Sta
to wrong conclusion (Uddin et al., 2021). This situation might occur due tistical methods used for water quality measurement and monitoring,
19
and finally, (d) suggesting the future research directions. Below, the determine their weights. Incorporation of international guideline values
main findings of the review on the surface water quality research is (e.g., WHO, EU WFD or similar) may also help to improve the process.
outlined. The use of fuzzy interface systems and AI-based models can also help in
A detailed examination of the fresh surface water landscape reveals reducing the uncertainty of the aggregate function.
13 distinct water sources (e.g., rivers, wetlands, ponds, lakes and so on) This study also highlights the limitations and practical usability is
that are mostly utilized by five sectors (e.g., agricultural, industrial, sues of the current manual water quality measurement approaches and
domestic, etc.). Practically all of these sectors are the primary source of argues that the design and development of a technologically advanced,
surface water contamination as industrial effluents, agricultural runoffs, autonomous Water Quality Monitoring System (WQMS) can overcome
pesticides and fertilizers, and domestic sewage often get deposited in the these limitations. Correspondingly, this study proposes a set of seven
nearby water sources. system requirements for the development of the same, namely, Geo-
Surface water quality and contamination severity are defined by the Tagged Parameter Measurement; Remote Sensing, and Real-time Su
assessment of 69 key water quality parameters. However, the selection pervision and Control; Remote Data Logging in the Cloud Server, Water
of these parameters is frequently influenced by a number of critical Quality Profiling, Statistical Data Analysis and Prediction; AI Integration
factors, including the natural properties of the parameters, the purpose for Profiling and Autonomous Decision Making; Integration to GEMS
for which the water is to be used, the extent to which quality is to be and One System Serves All. This system should enable safe, timely and
ensured, and the environmental significance of a water quality param comprehensive water quality data collection through self-operated
eter. As a result, this study carried out a detailed classification of the process with minimum human monitoring and intervention. The sys
parameters along these three axes. tem should effectively satisfy the needs of many stakeholders including
Furthermore, this study developed a comprehensive documentation Government policy makers, domain experts and researchers and the
of the water quality assessment models and statistical approaches in management.
relation to parameters, water sources, usage patterns, and model per
formance. The majority of WQI and PI models have four step evaluation
process, e.g., selecting the water quality parameters, determining Declaration of competing interest
parameter sub-indices, and assigning weights to the parameters, and
finally, applying an aggregation function to compute the overall water The authors declare that they have no known competing financial
quality or pollution index. Although most of the models are generic in interests or personal relationships that could have appeared to influence
terms of portability to other regions or sites, model applications are the work reported in this paper.
quite region-specific. The two main issues that affect the accuracy of the
model output are the eclipsing problem and model uncertainty. All four Data availability
stages of the WQI/PI models contribute to these issues. For example, the
classical Delphi Technique that is used for the parameter selection and No data was used for the research described in the article.
weightings, often introduces uncertainty and eclipsing effects into the
models. Furthermore, as the number of operators in the aggregation Acknowledgement
function increases, so does the model’s uncertainty. Therefore, future
research should investigate the role of statistical methods such as prin This research is partially funed by the RIoT Research Center, Inde
cipal component analysis and cluster analysis to select parameters and pendent University, Bangladesh.
Appendix
Table 9
Reference to the list of primary articles reviewed in this study.
Article Total Reference

Type Count
Journal 123 Nguyen and Huynh (2022) Lobato et al. (2015) Sánchez et al. (2007) Chowdhury et al. (2012a) Chowdhury et al. (2012b) Syeed et al. (2020) Mama et al.
(2021) Carvalho et al. (2011) Debels et al. (2005) Ortega et al. (2016) Pham et al. (2022) Uddin et al. (2021) Drasovean and Murariu (2021) Akter et al.
(2016) Lumb et al. (2011) Parsad and Bose (2001) Kurnaz et al. (2016) Kumar et al. (2019) Reza and Singh (2010) Karami et al. (2012) Parmar and
Bhardwaj (2013) Balla et al. (2022) Schreiber et al. (2022) Gopaul et al. (2009) Parmar and Bhardwaj (2014) Bartram et al. (2001) Bo et al. (2022)
Prathumratana et al. (2008) Islam et al. (2018) Hasan et al. (2019) Lowe et al. (2022) Chen et al. (2020) Altalak et al. (2022) Khan et al. (2021) Ustaoğlu
et al. (2021) Tripathee et al. (2016) Chigor et al. (2012) Organization (2022) Griffiths et al. (2012) Lumb et al. (2006) Low et al. (2016) Whitehead et al.
(2009) Zhao et al. (2013) Fallah and Zamani-Ahmadmahmoodi (2017) Acharya et al. (2020) Karim et al. (2021) Gholizadeh et al. (2016) Amiri et al.
(2021) Shamsuzzaman et al. (2017) Breen et al. (2018) D’Agostino et al. (2020) Rahman et al. (2021) Roje-Bonacci and Bonacci (2013) de Souza et al.
(2020) Gerecke et al. (2002) Boyacioglu (2010) Gorde and Jadhav (2013a) Omer (2019) Arora (2017) Lin et al. (2018) Beutler et al. (2014) Chormey
et al. (2018) Campanella et al. (2016) Das et al. (2007) Lasheen et al. (1990) Holcomb and Stewart (2020) Champa and Kabir (2018) Barrell et al. (2000)
Murariu et al. (2019) Gorde and Jadhav (2013b) Water and Organization (2009) Järup (2003) Chapra et al. (2021) Syafrudin et al. (2021)
Jóźwiakowska et al. (2020) Drasovean et al., 953 2019 Noori et al. (2019) Sharma and Kansal (2011) Ferreira et al. (2011) Said et al. (2004) Hsu and
Sandford (2007) House (1989) Hernando et al. (2006) Barbulescu et al. (2021) Fulazzaky et al. (2010) Othman and Alaa Eldin (2012) Sarkar and Abbasi
(2006) Smith (1990) Abdul Hameed M Jawad et al. (2010) Sutadian et al. (2018) Katyal (2011) Ewaid and Abed (2017) Dash and Kalamdhad (2021)
Uddin et al. (2017a) Banda and Kumarasamy (2020) Shokuhi et al. (2012) Qi et al. (2022) Koki et al. (2019) Cude (2001) Lowe et al. (2017) Ma et al.
(2020) Ongley and Booty (1999) Swamee and Tyagi (2000) Banu et al. (2013) Tian et al. (2015) Boyacioglu (2006) Kabir et al. (2020) Sutadian et al.
(2016a) Tyagi et al. (2013) Iticescu et al. (2019) Noori et al. (2010) Dong et al. (2015) Samad et al. (2013) Geetha and Gouthami (2016) Kamienski et al.
(2019) Hossain and Patra (2020) Shukla et al. (2017) Sargaonkar and Deshpande (2003) Karaouzas et al. (2021) Wepener (2012) Koparan et al. (2020)
Wang et al. (2017) Sutadian et al. (2016b)
Conferences 4 Islam et al. (2020); Ali and Qamar (2013); Hussen et al. (2018); Drasovean et al. (2018)
20
References Chapra, S.C., Camacho, L.A., McBride, G.B., 2021. Impact of global warming on dissolved
oxygen and bod assimilative capacity of the world’s rivers: modeling analysis. Water
13, 2408.
Abbasi, T., Abbasi, S.A., 2012. Water Quality Indices. Elsevier.
Chatterjee, A., 1996. Water Supply Waste Disposal and Environmental Pollution
Abdul Hameed M Jawad, A., Haider, S.A., Bahram, K.M., 2010. Application of water
Engineering (Including Odour, Noise and Air Pollution and its Control. Khanna
quality index for assessment of dokan lake ecosystem,kurdistan region, Iraq. J. Water
Publishers.
Resour. Protect. 2010.
Chen, Y., Song, L., Liu, Y., Yang, L., Li, D., 2020. A review of the artificial neural network
Acharya, A., Sharma, M.L., Bishwakarma, K., Dahal, P., Chaudhari, S.K., Adhikari, B.,
models for water quality prediction. Appl. Sci. 10, 5776.
Neupane, S., Pokhrel, B.N., Pant, R.R., 2020. Chemical characteristics of the
Chigor, V.N., Umoh, V.J., Okuofu, C.A., Ameh, J.B., Igbinosa, E.O., Okoh, A.I., 2012.
karmanasha river water and its appropriateness for irrigational usage. J. Nepal
Water quality assessment: surface water sources used for drinking and irrigation in
Chem. Soc. 41, 94–102.
zaria, Nigeria are a public health hazard. Environ. Monit. Assess. 184, 3389–3400.
Adimalla, N., Qian, H., 2019. Groundwater quality evaluation using water quality index
Chormey, D.S., Bakirdere, S., Turan, N.B., Engin, G.O., 2018. Fundamentals of Quorum
(wqi) for drinking purposes and human health risk (hhr) assessment in an
Sensing, Analytical Methods and Applications in Membrane Bioreactors. Elsevier.
agricultural region of nanganur, south India. Ecotoxicol. Environ. Saf. 176, 153–161.
Chowdhury, A.Z., Jahan, S.A., Islam, M.N., Moniruzzaman, M., Alam, M.K., Zaman, M.
Ahmed, D.S., 2016. River water quality report, department of environment, river water
A., Karim, N., Gan, S.H., 2012a. Occurrence of organophosphorus and carbamate
quality report. department of environment. https://doe.portal.gov.bd. (Accessed 3
pesticide residues in surface water samples from the rangpur district of Bangladesh.
March 2023). Accessed.
Bull. Environ. Contam. Toxicol. 89, 202–207.
Akter, T., Jhohura, F.T., Akter, F., Chowdhury, T.R., Mistry, S.K., Dey, D., Barua, M.K.,
Chowdhury, M.A.Z., Banik, S., Uddin, B., Moniruzzaman, M., Karim, N., Gan, S.H.,
Islam, M.A., Rahman, M., 2016. Water quality index for measuring drinking water
2012b. Organophosphorus and carbamate pesticide residues detected in water
quality in rural Bangladesh: a cross-sectional study. J. Health Popul. Nutr. 35, 1–12.
samples collected from paddy and vegetable fields of the savar and dhamrai upazilas
Ali, M., Qamar, A.M., 2013. Data analysis, quality indexing and prediction of water
in Bangladesh. Int. J. Environ. Res. Publ. Health 9, 3318–3329.
quality for the management of rawal watershed in Pakistan. In: Eighth International
Cornelissen, B., Zaidman, A., Van Deursen, A., Moonen, L., Koschke, R., 2009.
Conference on Digital Information Management (ICDIM 2013). IEEE, pp. 108–113.
A systematic survey of program comprehension through dynamic analysis. IEEE
Alley, E.R., 2007. Water Quality Control Handbook. McGraw-Hill Education.
Trans. Software Eng. 35, 684–702.
Altalak, M., Ammad uddin, M., Alajmi, A., Rizg, A., 2022. Smart agriculture applications
Cude, C.G., 2001. Oregon water quality index a tool for evaluating water quality
using deep learning technologies: a survey. Appl. Sci. 12, 5919.
management effectiveness 1. JAWRA Journal of the American Water Resources
Amiri, H., Hadizadeh, B., Mooselu, M.G., Azadi, S., Sayyahzadeh, A.H., 2021. Evaluating
Association 37, 125–137.
the water quality index in dam lake for cold water fish farming. Environmental
Das, A.K., Dutta, M., Cervera, M.L., de la Guardia, M., 2007. Determination of thallium in
Challenges 5, 100378.
water samples. Microchem. J. 86, 2–8.
Arora, P., 2017. Physical, chemical and biological characteristics of water (e content
Dash, S., Kalamdhad, A.S., 2021. Discussion on the existing methodology of entropy-
module). Cent. Univ. Punjab 1–16, 2018.
weights in water quality indexing and proposal for a modification of the expected
Ball, P., 2015. H2O: A Biography of Water. Hachette UK.
conflicts. Environ. Sci. Pollut. Control Ser. 28, 53983–54001.
Balla, D., Zichar, M., Kiss, E., Szabó, G., Mester, T., 2022. Possibilities for assessment and
Davis, M.L., 2010. Water and Wastewater Engineering: Design Principles and Practice.
geovisualization of spatial and temporal water quality data using a webgis
McGraw-Hill Education.
application. ISPRS Int. J. Geo-Inf. 11, 108.
de Souza, R.M., Seibert, D., Quesada, H.B., de Jesus Bassetti, F., Fagundes-Klen, M.R.,
Banda, T.D., Kumarasamy, M., 2020. Development of a universal water quality index
Bergamasco, R., 2020. Occurrence, impacts and general aspects of pesticides in
(uwqi) for south african river catchments. Water 12, 1534.
surface water: a review. Process Saf. Environ. Protect. 135, 22–37.
Banu, Z., Chowdhury, M.S.A., Hossain, M.D., Nakagami, K., 2013. Contamination and
Debels, P., Figueroa, R., Urrutia, R., Barra, R., Niell, X., 2005. Evaluation of water quality
Ecological Risk Assessment of Heavy Metal in the Sediment of Turag River,
in the chillán river (central Chile) using physicochemical parameters and a modified
bangladesh: an Index Analysis Approach.
water quality index. Environ. Monit. Assess. 110, 301–322.
Barbulescu, A., Barbes, L., Dumitriu, C.S., 2021. Assessing the water pollution of the
DeZuane, J., 1997. Handbook of Drinking Water Quality. John Wiley & Sons.
brahmaputra river using water quality indexes. Toxics 9, 297.
Dojlido, J., Best, G.A., et al., 1993. Chemistry of Water and Water Pollution. Ellis
Barrell, R., Hunter, P., Nichols, G., et al., 2000. Microbiological standards for water and
Horwood Limited.
their relationship to health risk. Comm. Dis. Publ. Health 3, 8–13.
Dong, J., Wang, G., Yan, H., Xu, J., Zhang, X., 2015. A survey of smart water quality
Bartram, J., Ballance, R., 1996. Water Quality Monitoring: a Practical Guide to the
monitoring system. Environ. Sci. Pollut. Control Ser. 22, 4893–4906.
Design and Implementation of Freshwater Quality Studies and Monitoring
Dooge, J.C., 2009. Fresh Surface Water. Eolss Publishers Company Limited.
Programmes. CRC Press.
Drasovean, R., Murariu, G., 2021. Water Quality Parameters and Monitoring Soft Surface
Bartram, J.K., Fewtrell, L., Stenström, T.A., 2001. Harmonised 891 Assessment of Risk
Water Quality Using Statistical Approaches. Promising Techniques for Wastewater
and Risk Management for Water-Related Infectious Disease : an Overview.
Treatment and Water Quality Assessment, p. 217.
Beecham, S., Baddoo, N., Hall, T., Robinson, H., Sharp, H., 2008. Motivation in software
Drasovean, A., Murariu, M., Condurache-Bota, F., Constantinescu, G., 2018. Studies on
engineering: a systematic literature review. Inf. Software Technol. 50, 860–878.
thewater qualityof the siret river, near galati city. In: 18th International
Behmel, S., Damour, M., Ludwig, R., Rodriguez, M., 2016. Water quality monitoring
Multidisciplinary Scientific GeoConferences SGEM 2018, pp. 687–694.
strategies—a review and future perspectives. Sci. Total Environ. 571, 1312–1329.
Drasovean, R., Murariu, G., Constantinescu, G., Circiumaru, A., 953 2019. Assessment of
Berner, K., Berner, R., 1996. Global Enviromental: Water, Air and Geochemical Cycles,
surface water quality of danube river in terms of usual parameters and correlation
vol. 376. Practice Hall, new jersey, usa.
analyses. Rev. Chim. (Bucharest) 70, 392–406.
Beutler, M., Wiltshire, K., Meyer, B., Moldaenke, C., Luring, C., Meyerhofer, M.,
D’Agostino, F., Bellante, A., Quinci, E., Gherardi, S., Placenti, F., Sabatino, N., Buffa, G.,
Hansen, U., 2014. Apha (2005), standard methods for the examination of water and
Avellone, G., Di Stefano, V., Del Core, M., 2020. Persistent and emerging organic
wastewater, Washington dc: American public health association. ahmad, sr, and dm
pollutants in themarine coastal environment of the gulf ofmilazzo (southern Italy):
Reynolds (1999), monitoring of water quality using fluorescence technique: prospect
human health risk assessment. Front. Environ. Sci. 8, 117 of European Communities,
of on-line process control. Dissolved Oxyg. Dyn. Model. Case Study A Subtrop.
C., 2000. Directive on establishing a framework for community action in the field of
Shallow Lake 217, 95.
water policy (2000/60/ec). Official Journal 327.
Bhateria, R., Jain, D., 2016. Water quality assessment of lake water: a review.
Ewaid, S.H., Abed, S.A., 2017. Water quality index for al-gharraf river, southern Iraq.
Sustainable Water Resources Management 2, 161–173.
Egypt. J. Aquat. Res. 43, 117–122.
Bhuiyan, M.A.H., Bodrud-Doza, M., Islam, A.T., Rakib, M., Rahman, M.S.,
Fallah, M., Zamani-Ahmadmahmoodi, R., 2017. Assessment of water quality in Iran’s
Ramanathan, A., 2016. Assessment of groundwater quality of lakshimpur district of
anzali wetland, using qualitative indices from 1985, 2007, and 2014. Wetl. Ecol.
Bangladesh using water quality indices, geostatistical methods, and multivariate
Manag. 25, 597–605.
analysis. Environ. Earth Sci. 75, 1–23.
Ferreira, W., Paiva, L., Callisto, M., 2011. Development of a benthic multimetric index
Bo, L., Liu, Y., Zhang, Z., Zhu, D., Wang, Y., 2022. Research on an online monitoring
for biomonitoring of a neotropical watershed. Braz. J. Biol. 71, 15–25.
system for efficient and accurate monitoring of mine water. IEEE Access 10,
Firoz, D.A., 2007. Water Environment of River Basins.
18743–18756.
Fulazzaky, M.A., Seong, T.W., Masirin, M.I.M., 2010. Assessment of water quality status
Boyacioglu, H., 2006. Surface water quality assessment using factor analysis. WaterSA
for the selangor river in Malaysia. Water Air Soil Pollut. 205, 63–77.
32, 389–393.
Geetha, S., Gouthami, S., 2016. Internet of things enabled real time water quality
Boyacioglu, H., 2010. Utilization of the water quality indexmethod as a classification
monitoring system. Smart Water 2, 1–19.
tool. Environ. Monit. Assess. 167, 115–124.
Gerecke, A.C., Schärer, M., Singer, H.P., Müller, S.R., Schwarzenbach, R.P., Sägesser, M.,
Breen, B., Curtis, J., Hynes, S., 2018. Water quality and recreational use of public
Ochsenbein, U., Popow, G., 2002. Sources of pesticides in surface waters in
waterways. J. Environ. Econ. Pol. 7, 911 1–91115.
Switzerland: pesticide load through waste water treatment plants—-current situation
Campanella, B., Onor, M., D’Ulivo, A., Giannecchini, R., D’Orazio, M., Petrini, R.,
and reduction potential. Chemosphere 48, 307–315.
Bramanti, E., 2016. Human exposure to thallium through tap water: a study from
Gholizadeh, M.H., Melesse, A.M., Reddi, L., 2016. A comprehensive review on water
valdicastello carducci and pietrasanta (northern tuscany, Italy). Sci. Total Environ.
quality parameters estimation using remote sensing techniques. Sensors 16, 1298.
548, 33–42.
Glasgow, H.B., Burkholder, J.M., Reed, R.E., Lewitus, A.J., Kleinman, J.E., 2004. Real-
Carvalho, L., Cortes, R., Bordalo, A.A., 2011. Evaluation of the ecological status of an
time remote monitoring of water quality: a review of current applications, and
impaired watershed by using a multi-index approach. Environ. Monit. Assess. 174,
advancements in sensor, telemetry, and computing technologies. J. Exp. Mar. Biol.
493–508.
Ecol. 300, 409–448.
Champa, H., Kabir, S.L., 2018. Microbial analysis of tap water collected from selected
Glozier, N.E., Prairie, C.E.C., Division, N.R.E.S., 2004. Water Quality Characteristics and
upazillas of jamalpur, tangail, kishoreganj and netrokona districts of Bangladesh.
Trends for Banff and Jasper National Parks: 1973-2002. Environment Canada.
Asian. J. Med. Biol. Res. 4, 193–200.
21
Gopaul, P.R., Nowbuth, M.D., Baguant-Moonshiram, Y., 2009. Water Quality Indexing metal content in surface water bodies: a meta-analysis using heavy metal pollution
for Predicting Variation of Water Quality over Time, vol. 15. University of Mauritius indices and multivariate statistical analyses. Chemosphere 236, 124364.
Research Journal, pp. 186–199. Kurnaz, A., Mutlu, E., Uncumusaoğlu, A.A., 2016. Determination of water quality
Gorde, S., Jadhav, M., 2013a. Assessment of water quality parameters: a review. J Eng parameters and heavy metal content in surface water of çiğdem pond (kastamonu/
Res Appl 3, 2029–2035. Turkey). Turk. J. Agric. Food Sci. Technol. 4, 907–913.
Gorde, S., Jadhav, M., 2013b. Assessment of water quality parameters: a review. J Eng Lasheen, M.R., Shehata, S.A., Ali, G.H., 1990. Effect of cadmium, copper and chromium
Res Appl 3, 2029–2035. (vi) on the growth of nile water algae. Water Air Soil Pollut. 50, 19–30.
Griffiths, C., Klemick, H., Massey, M., Moore, C., Newbold, S., Simpson, D., Walsh, P., Lin, T.F., Watson, S., Suffet, I.M., 2018. Taste and Odour in Source and Drinking Water:
Wheeler, W., 2012. Us environmental protection agency valuation of surface water Causes, Controls, and Consequences. IWA Publishing.
quality improvements. Rev. Environ. Econ. Pol. Lobato, T., Hauser-Davis, R., Oliveira, T., Silveira, A., Silva, H., Tavares, M., Saraiva, A.,
Gwynne, M.D., 1982. The global environment monitoring system (gems) of unep. 2015. Construction of a novel water quality index and quality indicator for reservoir
Environ. Conserv. 9, 35–41. water quality evaluation: a case study in the amazon region. J. Hydrol. 522,
Hasan, M.K., Shahriar, A., Jim, K.U., 2019. Water pollution in Bangladesh and its impact 674–683.
on public health. Heliyon 5, e02145. Low, K.H., Koki, I.B., Juahir, H., Azid, A., Behkami, S., Ikram, R., Mohammed, H.A.,
Hernando, M.D., Mezcua, M., Fernández-Alba, A.R., Barceló, D., 2006. Environmental Zain, S.M., 2016. Evaluation of water quality variation in lakes, rivers, and ex-
risk assessment of pharmaceutical residues in wastewater effluents, surface waters mining ponds in Malaysia. Desalination Water Treat. 57, 28215–28239.
and sediments. Talanta 69, 334–342. Lowe, L., Szemis, J., Webb, J.A., 2017. Uncertainty and environmental water. In: Water
Holcomb, D.A., Stewart, J.R., 2020. Microbial indicators of fecal pollution: recent for the Environment. Elsevier, pp. 317–344.
progress and challenges in assessing water quality. Current environmental health Lowe, M., Qin, R., Mao, X., 2022. A review on machine learning, artificial intelligence,
reports 7, 311–324. and smart technology in water treatment and monitoring. Water 14, 1384.
Hossain, M., Patra, P.K., 2020. Water pollution index–a new integrated approach to rank Lumb, A., Halliwell, D., Sharma, T., 2006. Application of ccme water quality index to
water quality. Ecol. Indicat. 117, 106668. monitor water quality: a case study of the mackenzie river basin, Canada. Environ.
House, M.A., 1989. A water quality index for river management. Water Environ. J. 3, Monit. Assess. 113, 411–429.
336–344. Lumb, A., Sharma, T., Bibeault, J.F., 2011. A review of genesis and evolution of water
Hsu, C.C., Sandford, B.A., 2007. The delphi technique: making sense of consensus. quality index (wqi) and some future directions. Water Quality, Exposure and Health
Practical Assess. Res. Eval. 12, 10. 3, 11–24.
Hussen, A.M.E.A., Retnaningdyah, C., Hakim, L., Soemarno, S., 2018. The variations of Ly, K., Larsen H, D.N., 2015. Lower mekong regional water quality monitoring report.
physical and chemical water quality in coban rondo waterfall, malang Indonesia. In: vientiane: Mekong river commission.
AIP Conference Proceedings. AIP Publishing LLC, 050011. Ma, Z., Li, H., Ye, Z., Wen, J., Hu, Y., Liu, Y., 2020. Application of modified water quality
Islam, F., Rahman, M., S.K.B.A.A., Halder, M., 2016. Heavy metals in water, sediment index (wqi) in the assessment of coastal water quality in main aquaculture areas of
and some fishes of karnofuly river, Bangladesh. Int. J. Environ. Res. 32, 321–332. dalian, China. Mar. Pollut. Bull. 157, 111285.
Islam, M.M., Karim, M.R., Zheng, X., Li, X., 2018. Heavy metal and metalloid pollution of Magesh, N., Krishnakumar, S., Chandrasekar, N., Soundranayagam, J.P., 2013.
soil, water and foods in Bangladesh: a critical review. Int. J. Environ. Res. Publ. Groundwater quality assessment using wqi and gis techniques, dindigul district,
Health 15, 2825. Tamil nadu, India. Arabian J. Geosci. 6, 4179–4189.
Islam, M.A., Khan, R.H., Syeed, M., 2020. A smart and integrated surface water monitor Mahmood, A., 2018. Evaluation of raw water quality in wassit governorate by canadian
system architecture: Bangladesh perspective. In: Proceedings of the International water quality index. In: MATECWeb of Conferences, EDP Sciences, 05020.
Conference on Computing Advancements, pp. 1–6. Mama, A.C., Bodo, W.K.A., Ghepdeu, G.F.Y., Ajonina, G.N., Ndam, J.R.N., 2021.
Iticescu, C., Georgescu, L.P., Murariu, G., Topa, C., Timofti, M., Pintilie, V., Arseni, M., Understanding seasonal and spatial variation of water quality parameters in
2019. Lower danube water quality quantified through wqi and multivariate analysis. mangrove estuary of the nyong river using multivariate analysis (Cameroon southern
Water 11, 1305. atlantic coast). Open J. Mar. Sci. 11, 103–128.
Järup, L., 2003. Hazards of heavy metal contamination. Br. Med. Bull. 68, 167–182. Mara, D., Horan, N.J., 2003. Handbook of Water and Wastewater Microbiology. Elsevier.
Jóźwiakowska, K., Brodowska, N., Wójcik, M., Listosz, A., Micek, A., Marzec, M., Murariu, G., Iticescu, C., Murariu, A., Rosu, B., Munteanu, D., Buruiana, D.L., 2019.
Pochwatka, P., 2020. The concentration of the salinity indicators in the water of the Assessment of water quality state dynamics using adaptive filtering methods and
bystrzyca river on the area of lublin city in Poland. J. Eng. Ecol. 21. neural networks approaching case study-danube river in galati area. Rev. Chem. 70,
Kabir, M.H., Islam, M.S., Tusher, T.R., Hoq, M.E., Al Mamun, S., 2020. Changes of heavy 1914–1919. https://doi.org/10.37358/RC.19.6.7246.
metal concentrations in shitalakhya river water of Bangladesh with seasons. Indones. Najafi Saleh, H., V.S.Z.A.e.a, ., 2020. Assessment of groundwater quality around
J. Sci. Technol. 5, 395–409. municipal solid waste landfill by using water quality index for groundwater
Kamienski, C., Soininen, J.P., Taumberger, M., Dantas, R., Toscano, A., Salmon resources and multivariate statistical technique: a case study of the landfill site,
Cinotti, T., Filev Maia, R., Torre Neto, A., 2019. Smart water management platform: qaem shahr city, Iran. Environ. Geochem. Health 1305–1319. https://doi.org/
iot-based precision irrigation for agriculture. Sensors 19, 276. 10.1007/s10653-019-00417-0.
Karami, J., Alimohammadi, A., Modabberi, S., 2012. Analysis of the spatio-temporal Nguyen, T.G., Huynh, T.H.N., 2022. Assessment of surface water quality and monitoring
patterns of water pollution and source contribution using the modis sensor products in southern vietnam using multicriteria statistical approaches. Sustain. Environ. Res.
and multivariate statistical techniques. IEEE J. Sel. Top. Appl. Earth Obs. Rem. Sens. 32, 1–12.
5, 1243–1255. Noori, R., Sabahi, M.S., Karbassi, A.R., Baghvand, A., Zadeh, H.T., 2010. Multivariate
Karaouzas, I., Kapetanaki, N., Mentzafou, A., Kanellopoulos, T.D., Skoulikidis, N., 2021. statistical analysis of surface water quality based on correlations and variations in
Heavymetal contamination status in Greek surface waters: a review with application the data set. Desalination 260, 129–136.
and evaluation of pollution indices. Chemosphere 263, 128192. Noori, R., Berndtsson, R., Hosseinzadeh, M., Adamowski, J.F., Abyaneh, M.R., 2019.
Karim, M.R., Chowdhury, P., Rahman, L., Kazary, S., 2021. 1016 an al-based security A critical review on the application of the national sanitation foundation water
system using computer vision and nlp conversion system. In: 2021 3rd International quality index. Environ. Pollut. 244, 575–587.
Conference on Sustainable Technologies for Industry 4.0 (STI). IEEE, pp. 1–5. Omer, N.H., 2019. Water quality parameters. Water quality-science, assessments and
Katyal, D., 2011. Water quality indices used for surface water vulnerability assessment. policy 18, 1–34.
Int. J. Environ. Sci. 2. Ongley, E.D., Booty, W.G., 1999. Pollution remediation planning in developing
Kawser Ahmed, M., Baki, M.A., Kundu, G.K., Islam, S., Islam, M., Hossain, M., et al., countries: conventional modelling versus knowledge-based prediction. Water Int. 24,
2016. Human health risks from heavy metals in fish of buriganga river, Bangladesh. 31–38.
SpringerPlus 5, 1–12. Organization, W.H., 2022. Guidelines for drinking-water quality. https://www.who.int/t
Keele, S., et al., 2007. Guidelines for Performing Systematic Literature Reviews in eams/environment-climate-change-and-health/water-sanitation-and-health/water-
Software Engineering. safety-and-quality/publications.
Khan, R., Saxena, A., Shukla, S., Sekar, S., Senapathi, V., Wu, J., 2021. Environmental who/hep/ech/wsh/2020.3 Organization, W.H., 2020. Chromium in Drinking-Water,
contamination by heavy metals and associated human health risk assessment: a case 1079. World Health Organization.
study of surface water in gomti river basin, India. Environ. Sci. Pollut. Control Ser. Organization, W.H., 1996. Cancer Pain Relief: with a Guide to Opioid Availability. World
28, 56105–56116. Health Organization.
Kitchenham, B., 2004. Procedures for Performing Systematic Reviews, vol. 33. Keele Ortega, D.J., Pérez, D.A., Américo, J.H., De Carvalho, S.L., Segovia, J.A., 2016.
University, Keele, UK, pp. 1–26. Development of index of resilience for surface water in watersheds. J. Urban
Kitchenham, B., Charters, S., et al., 2007. Guidelines for Performing Systematic Environ. Eng. 10, 72–82.
Literature Reviews in Software Engineering. Othman, F., Alaa Eldin, M.E., 2012. Assessment of the klang river quality using the water
Kitchenham, B., Pretorius, R., Budgen, D., Brereton, O.P., Turner, M., Niazi, M., quality indices. In: Advanced Materials Research. Trans Tech Publ, pp. 237–240.
Linkman, S., 2010. Systematic literature reviews in software engineering–a tertiary Parmar, K.S., Bhardwaj, R., 2013. Water quality index and fractal dimension analysis of
study. Inf. Software Technol. 52, 792–805. water parameters. Int. J. Environ. Sci. Technol. 10, 151–164.
Koki, I.B., Zain, S.M., Low, K.H., Azid, A., Juahir, H., Abdul Zali, M., 2019. Development Parmar, K.S., Bhardwaj, R., 2014. Water quality management using statistical analysis
of water quality index of ex-mining ponds in Malaysia. Mal. J. Fund. Appl. Sci 15, and time-series prediction model. Appl. Water Sci. 4, 425–434.
54–60. Parsad, B., Bose, J., 2001. Evaluation of heavymetal pollution index for surface and
Koparan, C., Koc, A.B., Sawyer, C., Privette, C., 2020. Temperature profiling of spring water near a limestonemining area of the lower himalayas. Environ. Geol. 41,
waterbodies with a uav-integrated sensor subsystem. Drones 4, 35. 8. https://doi.org/10.1007/s002540100380.
Kumar, V., Parihar, R.D., Sharma, A., Bakshi, P., Sidhu, G.P.S., Bali, A.S., Karaouzas, I., Parvin, F., Haque, M.M., Tareq, S.M., 2022. Recent status of water quality in Bangladesh:
Bhardwaj, R., Thukral, A.K., Gyasi-Agyei, Y., et al., 2019. Global evaluation of heavy a systematic review, meta-analysis and health risk assessment. Environmental
Challenges 6, 100416. https://doi.org/10.1016/j.envc.2021.100416.
22
Patil, P., Sawant, D., Deshmukh, R., 2012. Physico-chemical parameters for testing of Sutadian, A.D., Muttil, N., Yilmaz, A.G., Perera, B., 2016b. Development of river water
water–a review. Int. J. Environ. Sci. 3, 1194–1207. quality indices—a review. Environ. Monit. Assess. 188, 1–29.
Peavy, H.S., Rowe, D.R., Tchobanoglous, G., 1985. Environmental Engineering, ume Sutadian, A.D., Muttil, N., Yilmaz, A.G., Perera, B., 2018. Development of a water quality
2985. McGraw-Hill, New York. index for rivers in west java province, Indonesia. Ecol. Indicat. 85, 966–982.
Petticrew, M., Roberts, H., 2008. Systematic Reviews in the Social Sciences: A Practical Swamee, P.K., Tyagi, A., 2000. Describing water quality with aggregate index.
Guide. John Wiley & Sons. J. Environ. Eng. 126, 451–455.
Pham, T.L., Tran, T.H.Y., Tran, T.T., Ngo, X.Q., Nguyen, X.D., 2022. Assessment of Syafrudin, M., Kristanti, R.A., Yuniarto, A., Hadibarata, T., Rhee, J., Al-Onazi, W.A.,
surface water quality in a drinking water supply reservoir in vietnam: a combination Algarni, T.S., Almarri, A.H., Al-Mohaimeed, A.M., 2021. Pesticides in drinking
of different indicators. Rendiconti Lincei. Sci. Fis. Nat. 33, 653–662. https://doi.org/ water—a review. Int. J. Environ. Res. Publ. Health 18, 468.
10.1007/s12210-022-01086-5. Syeed, M.M., Hammouda, I., Systä, T., 2013. J. Softw. Evolution of Open Source Software
Poonam, T., Tanushree, B., Sukalyan, C., 2013. Water quality indices-important tools for Projects: A Systematic Literature Review, vol. 8, pp. 2815–2829. Tchobanoglus, G.,
water quality assessment: a review. International Journal of Advances in chemistry Burton, F., Stensel, H.D., 2003. Wastewater engineering: treatment and reuse.
1, 15–28. American Water Works Association. Journal 95, 201.
Prathumratana, L., Sthiannopkao, S., Kim, K.W., 2008. The relationship of climatic and Syeed, M., Islam, M.A., Fatema, K., 2020. Precision agriculture in Bangladesh: need and
hydrological parameters to surface water quality in the lower mekong river. Environ. opportunities. Precis. Agric. 29, 6782–6800.
Int. 34, 860–866. Tian, H., Zhu, C., Gao, J., Cheng, K., Hao, J., Wang, K., Hua, S., Wang, Y., Zhou, J., 2015.
Qi, J., Yang, L., Liu, E., 2022. A holistic framework of water quality evaluation using Quantitative assessment of atmospheric emissions of toxic heavy metals from
water quality index (wqi) in the yihe river (China). Environ. Sci. Pollut. Control Ser. anthropogenic sources in China: historical trend, spatial distribution, uncertainties,
29, 80937–80951. and control policies. Atmos. Chem. Phys. 15, 10127–10147.
Rahman, A., Jahanara, I., Jolly, Y.N., 2021. Assessment of physicochemical properties of Tripathee, L., Kang, S., Sharma, C.M., Rupakheti, D., Paudyal, R., Huang, J.,
water and their seasonal variation in an urban river in Bangladesh. Water Sci. Eng. Sillanpää, M., 2016. Preliminary health risk assessment of potentially toxic metals in
14, 139–148. surface water of the himalayan rivers, Nepal. Bull. Environ. Contam. Toxicol. 97,
Reza, R., Singh, G., 2010. Heavy metal contamination and its indexing approach for river 855–862.
water. Int. J. Environ. Sci. Technol. 7, 785–792. Turner, S., Pangare, G., Mather, R., 2009. Water Governance: A Situational Analysis of
Robinson, K.A., Brunnhuber, K., Ciliska, D., Juhl, C.B., Christensen, R., Lund, H., et al., cambodia, Lao Pdr and viet nam. Mekong Region Water Dialogue Publication. No 2.
2021. Evidence-based research series-paper 1: what evidence-based research is and Tyagi, S., Sharma, B., Singh, P., Dobhal, R., 2013. Water quality assessment in terms of
why is it important? J. Clin. Epidemiol. 129, 151–157. water quality index. J. Am. Water Resour. 1, 34–38.
Roje-Bonacci, T., Bonacci, O., 2013. The possible negative consequences of underground Uddin, M.G., Moniruzzaman, M., Khan, M., 2017a. Evaluation of groundwater quality
dam and reservoir construction and operation in coastal karst areas: an example of using ccme water quality index in the rooppur nuclear power plant area, ishwardi,
the hydro-electric power plant (hepp) ombla near dubrovnik (Croatia). Nat. Hazards pabna, Bangladesh. Am. J. Environ. Protect. 5, 33–43.
Earth Syst. Sci. 13, 2041–2052. Uddin, M.G., Moniruzzaman, M., Khan, M., 2017b. Evaluation of groundwater quality
Said, A., Stevens, D.K., Sehlke, G., 2004. An innovative index for evaluating water using ccme water quality index in the rooppur nuclear power plant area, ishwardi,
quality in streams. Environ. Manag. 34, 406–414. pabna, Bangladesh. Am. J. Environ. Protect. 5, 33–43.
Samad, M., Hossain, M., Rahman, B., 2013. Present status of broodstock management at Uddin, M.G., Nash, S., Olbert, A.I., 2021. A review of water quality index models and
carp hatcheries in jessore. J. Bangladesh Agric. Univ. 11, 349, 35. their use for assessing surface water quality. Ecol. Indicat. 122, 107218 https://doi.
Sánchez, E., Colmenarejo, M.F., Vicente, J., Rubio, A., García, M.G., Travieso, L., org/10.1016/j.ecolind.2020.107218.
Borja, R., 2007. Use of the water quality index and dissolved oxygen deficit as simple Ustaoğlu, F., Taş, B., Tepe, Y., Topaldemir, H., 2021. Comprehensive assessment of water
indicators of watersheds pollution. Ecol. Indicat. 7, 315–328. quality and associated health risk by using physicochemical quality indices and
Sargaonkar, A., Deshpande, V., 2003. Development of an overall index of pollution for multivariate analysis in terme river, Turkey. Environ. Sci. Pollut. Res. 28,
surface water based on a general classification scheme in indian context. Environ. 62736–62754.
Monit. Assess. 89. Viessman, W., Hammer, M.J., 1993. Water Supply and Pollution Control.
Sarkar, C., Abbasi, S., 2006. Qualidex: a virtual instrument for continuous monitoring of Wang, X., Zhang, F., Ding, J., 2017. Evaluation of water quality based on a machine
water quality indices. Environ. Monit. Assess. 119, 201–231. learning algorithm and water quality index for the ebinur lake watershed, China. Sci.
Sarkar, A.M., Rahman, A., Samad, A., Bhowmick, A.C., Islam, J.B., et al., 2019. Surface Rep. 7, 12858.
and ground water pollution in Bangladesh: a review. Asian Review of Environmental Wanyama, S.B., McQuaid, R.W., Kittler, M., 2022. Where you search determines what
and Earth Sciences 6, 47–69. you find: the effects of bibliographic databases on systematic reviews. Int. J. Soc.
Sarker, B., Keya, K.N., Mahir, F.I., Nahiun, K.M., Shahida, S., Khan, R.A., et al., 2021. Res. Methodol. 25, 409–422.
Surface and ground water pollution: causes and effects of urbanization and Water, S., Organization, W.H., et al., 2009. Risk Assessment of Cryptosporidium in
industrialization in south asia. Sci. Rev. 7, 32–41. Drinking Water. World Health Organization. Technical Report.
Schreiber, S.G., Schreiber, S., Tanna, R.N., Roberts, D.R., Arciszewski, T.J., 2022. Weibe, S.M.J., 2021. Assessment of Drinking Water Quality for Residents of Kuria West
Statistical tools for water quality assessment and monitoring in river ecosystems–a in Migori County. Ph.D. thesis. University of Nairobi.
scoping review and recommendations for data analysis. Water Qual. Res. J. 57, Wepener, V., 2012. Metal Ecotoxicology of the Olifants River in the Kruger National Park
40–57. and the Effect Thereof on Fish Haematology. University of Johannesburg (South
Shamsuzzaman, M.M., Islam, M.M., Tania, N.J., Al-Mamun, M.A., Barman, P.P., Xu, X., Africa).
2017. Fisheries resources of Bangladesh: present status and future direction. Fish. Whitehead, P.G., Wilby, R.L., Battarbee, R.W., Kernan, M., Wade, A.J., 2009. A review of
Aquacult. 2, 145–156. the potential impacts of climate change on surface water quality. Hydrol. Sci. J. 54,
Sharma, D., Kansal, A., 2011. Water quality analysis of river yamuna using water quality 101–123.
index in the national capital territory, India (2000–2009). Appl. Water Sci. 1, Wilhm, J.L., Dorris, T.C., 1968. Biological parameters for water quality criteria.
147–157. Bioscience 477–481.
Shokuhi, R., Hosinzadeh, E., Roshanaei, G., Alipour, M., Hoseinzadeh, S., 2012. Yousefi, M., Saleh, H.N., Mohammadi, A.A., Mahvi, A.H., Ghadrpoori, M., Suleimani, H.,
Evaluation of aydughmush dam reservoir water quality by national sanitation 2017. Data on water quality index for the groundwater in rural area neyshabur
foundation water quality index (nsf-wqi) and water quality parameter changes. Iran. county, razavi province, Iran. Data Brief 15, 901. https://doi.org/10.1016/j.
J. Health Environ. 4. dib.2017.10.052. –907. URL: https://www.sciencedirect.com/science/article/pii/
Shukla, A.K., Ojha, C., Garg, R., 2017. Application of overall index of pollution (oip) for S2352340917305735.
the assessment of the surface water quality in the upper ganga river basin, India. In: Yousefi, M., Ghoochani, M., Hossein Mahvi, A., 2018. Health risk assessment to fluoride
Development of Water Resources in India. Springer, pp. 135–149. in drinking water of rural residents living in the poldasht city, northwest of Iran.
Smith, D.G., 1990. A better water quality indexing system for rivers and streams. Water Ecotoxicol. Environ. Saf. 148, 426–430. https://doi.org/10.1016/j.
Res. 24, 1237–1244. ecoenv.2017.10.057. https://www.sciencedirect.com/science/article/pii/S0147651
Stickle, H.W., 1919. Monongahela River Navigation, vol. 11. Professional Memoirs, 317307273.
Corps of Engineers, United States Army, and Engineer Department at Large, Zhao, P., Tang, X., Tang, J., Wang, C., 2013. Assessing water quality of three gorges
pp. 695–713. reservoir, China, over a five-year period from 2006 to 2011. Water Resour. Manag.
Sutadian, A.D., Muttil, N., Yilmaz, A.G., Perera, B., 2016a. Development of river 1142 27, 4545–4558.
water quality indices—a review. Environ. Monit. Assess. 188, 1–29.
23

SYEED Et Al, 2023

Uploaded by

Copyright:

Available Formats

SYEED Et Al, 2023

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SYEED Et Al, 2023

Uploaded by

Copyright:

Available Formats

Environmental and Sustainability Indicators 18 (2023) 100247

Contents lists available at ScienceDirect

Environmental and Sustainability Indicators

Fig. 1. Overview of the SLR methodology used in this research.

3.2.3. Inclusion criteria

Table 3 3.2.5. Reference checking and final selection

Fig. 2. Demography of the primary articles.

Fig. 4. Categorization of research articles.

Fig. 7. Indicative classification of the water quality parameters.

Table 6 Table 6 (continued )

Table 6 (continued ) Table 6 (continued )

(51–70), Bad piggery waste

Table 7 (continued ) Table 7 (continued )

fish life Polluted (4<

Table 8 Table 8 (continued )

7.1.7. One System Serves All 8. Validation of the study

Article Total Reference

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.