Exploring the recent trends, progresses, and challenges in the application of Artificial intelligence in water quality assessment and monitoring in Nigeria: A systematic review

Exploring the recent trends, progresses, and challenges in the application of Artificial intelligence in water quality assessment and monitoring in Nigeria: A systematic review | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Exploring the recent trends, progresses, and challenges in the application of Artificial intelligence in water quality assessment and monitoring in Nigeria: A systematic review Michael E. Omeka This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4429965/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In recent decades, machine learning (ML) artificial intelligence has found wide application in water quality monitoring and prediction due to the increasing complexity of water quality data. This complexity has been attributed to the global upsurge in anthropogenic activities and climatic variations. It is therefore critical to identify the most accurate and suitable ML model for water quality prediction. In this study, a systematic literature review (SLR) was carried out to explore the trend and progress in the application of ML models in water quality monitoring and prediction in Nigeria from 2003–2024. A comprehensive review of the effectiveness of advanced ML models as well as the gaps in their application in the area of water quality assessment and monitoring was also carried out using the PRISMA-P meta-analysis technique. Forty publications were used to perform bibliographic analysis and visualization using the VOS viewer software. The study found that globally, the use of hybrid ML models in water quality prediction has not been well explored; a majority of the prediction has been based on the use of artificial neural networks (ANN). Among the ANN algorithms, the adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models are the most accurate in prediction; with temperature, dissolved oxygen (DO), pH, conductivity (EC), and total dissolved solids (TDS) among the most frequently predicted parameters. Nigeria is grossly lagging in the application of ML in water quality prediction. This limitation is largely attributed to inadequate data on environmental monitoring. It is critical therefore for future water quality monitoring and prediction studies in Nigeria to take advantage of the rapidly evolving field of machine learning; with more emphasis placed on the hybridized machine learning algorithms Machine learning Pollution Artificial intelligence Drinking water quality Nigeria Figures Figure 1 Highlights Globally, the use of hybrid ML models in water quality prediction has not been well explored; a majority of the prediction has been based on the use of artificial neural networks (ANN). The adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models are the most accurate in prediction. Nigeria is grossly lagging in the application of ML in water quality prediction. This limitation is largely attributed to inadequate data on environmental monitoring. 1. Introduction Approximately 75% of the earth's crust is made up of water. About 68.7 and 30.3% of the total water mass, respectively, are encased in ice caps or glaciers and ocean, respectively. Fewer than 1% are regarded as fresh (Al Aani et al. 2019 ). The quality of water supplies has been declining dramatically in recent decades. This reflects both the increase in human activity and the variability of the global climate (Ayejoto et al. 2023 ; Agbasi et al. 2023 ). Natural hydrological, hydrogeologic, and geogenic processes also influence water quality because they alter the water's chemistry (Xiao et al. 2021 ). The predominant pollution precursors that are unique to the location, either locally or regionally, will have a major influence on how quickly the water quality is altered, or polluted. Furthermore, the mechanisms and origins of declining water quality are typically numerous and intricate (Omeka, 2023). Examples of natural process variations in aquifer characteristics include water-rock interaction, chemical speciation, changes in stream dynamics in response to pollutant input, and variations in rainfall intensity. Both point (produce from mine and industrial effluents) and non-point (runoff from crop fields, effluents from town waste dumps seeping into groundwater, runoff from open defecation, etc.) sources can contribute to anthropogenic processes (Xiao et al. 2021 ; Okamkpa et al. 2022 ). Because of this intricacy, evaluating the quality of the water usually involves difficult and thorough data gathering, laboratory investigation, and data analysis. Both conventional and non-traditional methods have been used to monitor and assess the quality of the water in recent decades. The non-conventional approach uses multiple quantitative metrics, parametric and non-parametric models of statistics, health risk assessment mathematical models, and spatiotemporal models (by employing GIS) for quality checks, whereas traditional methods depend on the application of local and international benchmark standards. However, the disadvantage of using numerous numerical models is that it usually takes a lot of effort and expert knowledge to compute the sub-indices for model output. This will therefore lead to inaccuracies in the output of the results and consequently affect the decision-making process, which could have a significant effect on the economy (Kouadri and Samir, 2021 ). With a population density of more than 1.9 million, Nigeria has the largest human population of all the black African nations (Ighalo et al., 2020 ). A startling 66.3 million people, however, lack access to portable, safe drinking water (Ighalo et al., 2020 ; Okamkpa et al. 2022 ). Additionally, the accessible water sources are under a lot of stress because of the high population density and rising socioeconomic activity. Additionally, it has been established that seasonal variations in temperature and rainfall intensity brought on by climate change significantly alter the quality of water resources (Kouadri and Samir, 2021 ; Omeka et al., 2024 ). In this regard, there will be an indirect rise in the variability of climate change as a result of the growing anthropogenic activities in the area (such as hydrocarbon exploration, industrialization, land usage, and population growth). The frequency of aquifer recharge/discharge rates will therefore be impacted by this (Omeka and Egbueri 2022 ; Ricolf et al., 2020). Therefore, efforts and focus should be directed towards developing a relationship between the various water quality precursors unique to the area in order to provide effective and sustainable water management and preservation. This will require the development of innovative, economical methods for comprehensive, reliable assessment and monitoring of water quality. Artificial Intelligence (AI) can be utilised to achieve this. In this regard, there will be an indirect rise in the variability of climate change as a result of the growing anthropogenic activities in the area (such as hydrocarbon exploration, industrialization, land usage, and population growth). The frequency of aquifer recharge/discharge rates will therefore be impacted by this (Omeka and Egbueri 2022 ; Ricolf et al., 2020). Therefore, efforts and focus should be directed towards developing a relationship between the various water quality precursors unique to the area in order to provide effective and sustainable water management and preservation. This will require the development of innovative, economical methods for comprehensive, reliable assessment and monitoring of water quality. Artificial Intelligence (AI) can be utilised to achieve this. Globally, there has been some progress from the conventional water quality assessment and monitoring techniques towards artificial intelligence (AI) and Internet of Things (IoT) technologies (Geetha and Gouthami 2016 ). AI was first introduced in the early 1950s and has since gone through significant modifications and gained applications in various fields in data simulation and forecasting of events outcomes (Al Aani et al. 2019 ). The use of AI in water quality assessment and monitoring started in the early 1990s and has since then gained the attention of many researchers and environmental stakeholders (Hasan et al. 2011 ). As a result of the complexity, ubiquity, and nonlinearity usually associated with water quality datasets, a challenge may arise in the prediction of water quality through the use of conventional approaches. This is attributed to increasing sources of anthropogenic pollution sources and varying climatic changes. Therefore, developing a model that can efficiently predict and monitor water quality remains crucial, especially in the present Anthropocene age where there seems to be no sight of solution to the impending global pollution crisis. As part of her industrial revolution plan, Nigeria projects to increase its industrial output from 4–10% of GDP by 2020 (NIRP 2014). This will mean that more pressure will be put on the environment and its valued components (such as water, air, and soil). The big question remains how the country will maintain a balance between its industrial revolution projections and the protection of valued environmental components. It is feared that with the continuous increase in population and impeding industrial revolution, Nigeria will be put under greater strain in terms of climatic variations. Additionally, the quantity and quality of water resources will experience a further decline and continuous shifts (Ayejoto et al. 2023 ). It becomes imperative therefore for more conscious action plans to be put toward the assessment and prediction of the quality of water resources for informed decision-making and amelioration of contaminated sites through the use of cutting-edge approaches. As a response, in recent times, some studies have tilted towards the use of AI in water quality prediction in Nigeria, with some recorded progress. On a global scale, Ighalo et al. ( 2020 ) have carried out a comprehensive review of the application of different AI models in the assessment and monitoring of surface water in the last decade (2011–2020). Their study found that two AI models- Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) were the most applied models for water quality assessment and monitoring within the period. The study also showed that globally, Southeast Asia and Iran are among the countries with the most studies on the use of these models in water quality prediction. Studies in Nigeria on the application of AI in water quality prediction are very few in the literature. Among these studies, most have focused on the use of standalone machine learning (ML) algorithms (Egbueri 2022 ; Egbueri and Agbasi 2022 ; Omeka 2023) while some have integrated ML and AHP in water quality prediction and prioritization (Ifediegwu 2022 ; Omeka et al. 2023). For instance, Egbueri ( 2022 ) has integrated a multilayer perceptron artificial neural network (MLP-ANN) and multiple linear regression (MLR) with corrosion and scaling potential indices for the prediction of industrial water quality in southeastern Nigeria. This study concluded that the incorporation of ML with indexical models can provide robustness in industrial water quality assessment and management. In another study in the region, MLR and MLP-ANN were incorporated to predict groundwater quality using two numerical models- overall index of pollution (OIP) and water quality index (WQI) as predicted or output variables. The MLP-ANN showed higher efficiency of prediction as lower modeling errors were observed in their validation metrics (Egbueri and Agbasi, 2022 ). In an agricultural mine district in southeastern Nigeria, a composite irrigation suitability zonation map was generated through the integration of a GIS-based analytical hierarchy process (AHP) and ML algorithms (Omeka et al. 2023). The model successfully identified Mg, HCO 3 , Cl, and SO 4 as the major contributors to the deterioration of the irrigation water quality in the region. In a similar study, simple kriging (SK) and inverse distance weighting (IDW) GIS-based spatiotemporal models have been used in combination with supervised ML algorithms for a composite assessment of the irrigation water quality for crop yield in a mining region in southeastern Nigeria (Omeka, 2023). This study revealed that soil is more susceptible to sodicity through the application of surface water for irrigation due to the higher pH of the surface water compared to groundwater. Indeed, some efforts have been put toward the use of AI in water quality monitoring in Nigeria, especially, in the southeastern parts. It is however recommended that more research be channeled toward this area by embracing the application of hybridized (ensemble) ML algorithms. Due to the nonlinearity and randomness of water quality data, and the impending climate variability, the need for the use of improved ML models is imperative. The relationship between water quality and its controlling factors is known to vary over time (due to variations in anthropogenic factors and seasonal fluctuations), hence, the use of a predictive model that functions under the framework of historical data becomes inefficient over time (Khan et al. 2021 ). The use of hybridization in other parts of the world has shown very promising results. For instance, in Pakistan, the forward-rolling artificial intelligence algorithm has been used to generate a model that will show resiliency to variability in climate change in a hydrologic setting (Khan et al. 2021 ). Two ML algorithms- the recursive feature elimination and support vector machine learning (RFESVM) have been combined to monitor the variability in rainfall in Malaysia (Pour et al. 2020 ). In Penang, Malaysia, the support vector machines (SVM), backpropagation neural network (BP-NN), and radial basis function neural network (RB-NN) have been incorporated to monitor the quality of surface water in constructed wetlands (Mohammadpour et al. 2015). Nayak ( 2020 ) combined and compared the use of multiple linear regression (MLP) with correlation models in assessing the trend in the surface water quality of the Brahmani River. The MLR models were found to show better results in water quality trends. In the semi-enclosed Tolo Harbour water bay in Hon Kong, random forest machine (RF) and ANN machine learning models were combined to assess the spatial variations in eutrophication in the water. The ensemble model identified suspended solids, dissolved nutrients, and organic pollutants as significant contributors to water quality (Deng et al. 2022 ). In Iran, Sun et al. ( 2021 ), have combined two hybrid ML models- the multivariate adaptive regression spline (MARS) and intrinsic time-scale decomposition (ITD) with the multi-step supervised-based machine learning approach (MSMLEA) for the prediction of the water quality of the Tajan River in Iran. From the results, the hybrid ML models showed higher performance accuracy in water quality prediction. Nigeria has seen serious incidences of cholera, diarrhoea, and other gastrointestinal ailments in the past few decades as a result of consuming polluted water (Martins et al. 2016 ). In the joint WHO/UNICEF (2012) report on monitoring programmes for sanitation and water supply, Nigeria comes in third place, right after China and India, among the nations without access to clean drinking water. Waterborne illnesses are thought to be the cause of approximately 130,000 child deaths in Nigeria each year (WHO/UNICEF 2012). If immediate action is not taken, it is anticipated that this number would rise by 2040 (WHO/UNICEF 2012). As per the 2016 report on cancer patterns in Nigeria by the Nigeria National System of Cancer Registries, around 80% of the country's cancer risks are attributable to pollution. Thirty percent of this are related to drinking tainted water (Morounke et al. 2017 ). This seems to support the 1996 Harvard report on cancer prevention, which found that of all the risk factors for cancer, the consumption of dissolved heavy metals (such as Ni, Cd, As, and Pb) in drinking water poses 2% of the world's cancer risks. Therefore, there is a need for an ongoing evaluation of innovative and state-of-the-art methods for comprehensive and sufficient water quality monitoring and prediction. In contrast to developing nations like Nigeria, data on seasonal variations and water quality is widely available in developed countries. Several factors, such as a lack of government funding and inaccessibility to laboratory facilities in Nigeria, can be connected to this. Consequently, a systematic literature review (SLR) covering the previous 20 years (2003–2024) on the use of machine learning models in water quality monitoring and assessment was conducted for this study. The trend in machine learning, a relatively recent field of study that has showed significant promise in the prediction of water quality, must be embraced given the critical role clean water plays in both sustainable development and the preservation of human health. Furthermore, a comprehensive evaluation of the advancement from current well-observed developments to more advanced and hypothetical machine learning model simulation and comparative analysis has also been presented. The first section considers common factors that affect Nigeria's water quality in relation to the sources of contaminants and provides information on potential new contaminants. The second part considers the implications for prospects as well as the past and anticipated future trends in Nigeria's methods for managing and monitoring water quality. A priority for the frequency of input variables and the precision of validation measures used to show the findings has also been placed on the deployment of different ML models in water quality monitoring. Future research potential and investigations on water quality have also been recommended. 2. Study area 2.1 Location Nigeria is bounded to the north by Niger, towards the east by Chad and Cameroon, towards the south by the Atlantic Ocean's Gulf of Guinea, and to the west by Benin (Encyclopædia Britannica 2023 ) (Fig. 1) 2.1 Relief Nigeria's terrain is characterized by lowlands in the north and south, which are punctuated by plateaus and mountains in the center of the nation. The Sokoto Plains are located in the country's northwest region, whereas the Borno Plains are located in the country's northeastern corner and extend to the Lake Chad basin. Porous, geologically young sedimentary strata underpin the Lake Chad basin and coastal regions, along with the Niger River delta and western sections of the Sokoto region in the extreme northwest (Encyclopædia Britannica, 2023 ; Ayejoto et al. 2023 ). The plateaus' distinctive features are high plains with vast, shallow valleys interspersed with several hills or solitary summits known as inselbergs; the base rocks are crystalline, with sandstones appearing in river basins. Some degraded features, including the Udi-Nsukka escarpment, rise suddenly above the plains at least 1,000 feet in height (300 meters) (Okamkpa et al. 2022 ). 2.2 Climate The climate of Nigeria is tropical, with varying rainy and dry seasons depending on location. The southeast is hot and humid most of the year, but dry in the southwestern and further inland. The west and north have a savanna climate with distinct dry and wet seasons, whereas the extreme north has a steppe environment with limited precipitation. The frequency of the rainy season generally reduces from south to north. The rainy season runs from March to November in the south, except for mid-May to September in the extreme north. The rains stop completely in the south during August, resulting in a brief dry season known as the "August break." Precipitation is greater in the south, particularly the southeastern, which experiences over 120 inches (3,000 mm) of rain every year, compared to around 70 inches (1,800 mm) in the southwest. Temperature and humidity stay largely consistent in the south throughout the year, but seasons fluctuate significantly in the north; and during the northern dry season, the average temperature range also increases. 3. Methodology 3.1 Systematic Literature Review The research method used in the present study was the systematic literature review. To carry out the research, the following research questions were raised: (1) what are the major sources of water contamination in Nigeria? (2) what are the most frequently used machine learning models in water quality analysis in the last two decades? (3) what are the most accurate machine learning models based on their model predict metrics (such as coefficient of determination [ R 2 ]) (4) what are the most frequently monitored water quality parameters (5) what are the implications of the use of ML models in water quality monitoring and their implication for future research prospects? The systematic literature review was comprised of published articles from Web of Science and Scopus databases. The search terms used for the search in the databases included the sources of contamination in the aquatic environment in Nigeria and the application of machine learning in water quality prediction In the Scopus database, each term was entered in the search directory including the abstract, title, and keywords. The filters used were the year of publication (within the date range of 2003–2024), the thematic area of environmental sciences, access type (open and none-open access articles), and the document type (article or review). In the Web of Science database, the search filter was based on Abstract, title, author keywords, and study within the date range of 2003–2024. Two thematic categories- machine learning in water quality analysis and environmental science. The articles were selected based on the following criteria: (1) articles that give a detailed explanation of the major sources of water pollution (2) articles that explain the use of hybridized or standalone machine learning algorithms in water quality monitoring (3) articles that explain the peculiarities of the different validation metrics (e.g. coefficient of determination, standard error of estimate, the sum of square errors and residual error). The exclusion was made on certain criteria such as: (1) articles that appear outside the scope of the study (2) articles that do not explicitly address the use of machine learning algorithms in water quality monitoring (3) articles concerning other valued environmental components outside of water. 3.2 Creating a bibliographic map using the VOSviewer software The VOSviewer was used in creating a bibliographic map for keywords in machine learning and water quality between the period of 2003–2024. Bibliographic data from Scopus and Web of Science databases were extracted with a filter in the area of environmental sciences. A total of 729 and 2,234 articles were respectively returned for the search terms for sources of contamination in the aquatic environment in Nigeria and the application of machine learning in water quality prediction in the Scopus database. For the Web of Science database, 521 and 1,672 articles respectively were returned. The articles were then exported from the database in CSV format to the VOSviewer. The bibliographic map was then created in the VOSviewer environment using keywords with a minimum of five occurrences selected (van Ech and Waltman 2020 ). 4. Results and discussion 4.1 Results from Databases Table 1 shows the number of analyzed articles considered in the review, while Fig. 2 shows the PRISMA 2000 flow diagram for systematic literature review (SLR) showing the results of the database searches and their registries. The variation in the number of articles using the two search items ( sources of contamination in aquatic environment in Nigeria and application of machine learning in water quality prediction ) in the two databases showed an upward increase in articles on the term application of machine learning in water quality prediction compared to the former. The Scopus database showed a higher number of publications compared to the Web of Science database. This is because the Scopus database has a greater number of indexed journals compared to the Web of Science database. Table 1 Number of articles analyzed and considered for the SLR Search terms Returned articles Databases Selected articles in the search Sources of contamination in the aquatic environment in Nigeria 729 Scopus 35 Application of machine learning in water quality prediction 2,234 Scopus 20 Sources of contamination in the aquatic environment in Nigeria 521 Web of Science 25 Application of machine learning in water quality prediction 1,672 Web of Science 18 The sum of articles considered from abstracts and titles 98 The sum of articles considered after full-text analysis 40 The number of articles returned with the use of the search term sources of contamination in the aquatic environment in Nigeria and application of machine learning in water quality prediction were 729 and 2,234 respectively in the Scopus database. For the Web of Science database, 521 and 1,672 articles respectively were returned from the search (Table 1 , Fig. 2). The number of articles considered for review was 35 and 20 for the terms sources of contamination in the aquatic environment in Nigeria and application of machine learning in water quality prediction in the Scopus database. While 25 and 18 were selected for the Web of Science database (Table 1 ). It is worth mentioning that due to a large number of returned articles for the Scopus database, the titles and abstracts of articles read were for the first 600. For inclusion and exclusion criteria, only articles that did not appear in the two databases were included for the review; as some articles appeared in both the Scopus and Web of Science databases. Therefore, a total of 98 articles were selected for reading of titles and abstracts while a total of 40 articles were selected for full-text analysis while others were discarded (Table 1 ). The discarded articles did not meet the established criteria. Based on the search term “ application of machine learning in water quality prediction” , out of the 257 publications that were considered, China occurred in 61, the United States in 57, India in 32, Malaysia in 15, south Korea in 14, Australia in 13, the United Kingdom and Canada in 10, Vietnam in 9, Iran in 8, Brazil and Iraq occurred in 7, Poland and Taiwan occurred in 6, Russian Federation, France, Germany, Italy, and Saudi Arabia occurred in 5; Morocco, Sweden, Bangladesh, Netherlands, and Hong Kong occurred in 4; Singapore, Spain, United Arabs Emirates, Norway, Palestine, Pakistan, and Greece occurred in 3; Croatia, Egypt, Pietro Florentine, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia occurred in 2; Dakota, Philippines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Israel, and Mexico occurred in 1 had one publication each (Table 2 ). It can therefore be deduced frequency of the application of ML in these countries may be due availability of experts in the use of neural networks in engineering and environmental studies. Although this may be speculative, the author hopes that this observation will spur the interest of researchers in the field for more discussion and enlightenment. Table 2 Number of articles per country occurring from the systematic literature review Country No. of Publication China 61 United States 57 India 32 Malaysia 15 South Korea 14 Australia 13 United Kingdom 10 Canada 10 Vietnam 9 Iran 8 Brazil, Iraq 7 Poland, Taiwan, Poland 6 Russian Federation, France, Germany, Italy, Saudi Arabia 5 Morocco, Sweden, Bangladesh, Netherlands, Hong Kong 4 Singapore, Spain, United Arab Emirates, Norway, Palestine, Pakistan, Greece 3 Croatia, Egypt, Pietro Florentine, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia 2 Dakota, Philippines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Israel, and Mexico. 1 4.2 Bibliographic mapping using VOSviewer The bibliographic map was created using the VOSviewer software. Co-occurrence was chosen as the type of analysis, while All keywords were chosen as the unit of analysis, with fractional counting selected as the preferred counting method. For the application of machine learning in the water quality prediction search file, the software returned 377 keywords with 186 meeting the threshold and occurring at least six times across all analyzed articles. For each of the 186 keywords, the total strength of the co-occurrence links with other keywords was calculated. The keyword with the greatest link strength was then selected for generating the bibliographic map. Links between keywords are an indication of the relationship between them (the link strength) represented by a numerical value. Implying that the higher the link strength, the stronger the association where the two terms occur together (van Eck and Waltman 2020). Terms were then grouped into clusters of a different colour; with each colour representing the strength of the association. For the application of machine learning in water quality prediction terms, analysis was made based on both the countries of study and the type of machine learning algorithm frequently used. The bibliographic map based on the overlay visualization (with the year of publication) and density visualization is shown in Fig. 3 based on the type of machine learning model. The graphical representation of the variation in the number of publications and the occurrence of the ML algorithms in water quality prediction is shown in Fig. 3. Artificial neural networks had the highest occurrence and strongest link strength in all analyzed publications followed by regression analysis and random forest , while Bayes theorem and the convolutional neural network had the lowest occurrence and weakest link strength (Fig. 3–4). From the bibliographic map, high density and link strengths seem to occur mostly with Artificial neural networks , support vector machines, regression analysis, decision trees, and random forests . However, fuzzy systems, long short-term memory, convolutional neural networks, and hybrid, and ensemble ML models had the weakest links and strength, with the number of publications on the subject seeming to be increasing since 2021. This result implies that the use of hybrid ML models in water quality prediction has not been well explored globally; a majority of the prediction has been based on the use of artificial neural networks. Based on the countries of study, China, the United States, and India had the highest number of publications and link strength concerning the term application of machine learning in water quality prediction (Fig. 5). The total number of publications involving the review developed by country was 61, 57, and 15 respectively for China, the United States, and India (Table 2 , Fig. 5). This implies that countries like Croatia, Egypt, Pietro Fiorentini, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia, Dakota, Philipines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Isareal and Mexico are grossly lacking behind in the use of ML in water quality monitoring and assessment. For the sources of contamination in the aquatic environment in Nigeria search term, analysis was made based on the relationship between anthropogenic and natural/geogenic sources of contamination and the number of publications within the last two decades (2003–2024). Based on the relationship between pollutant type, heavy metals, Lead, Cadmium, Copper, Chromium, and Zinc were reported in more publications and had the greatest link strength among others. Meanwhile, Antibiotic agents, aromatic compounds, bitumen, Cl, combustion, dumpsite, EC, mineralization, municipal solid wastes, organic carbon, phosphates, rainwater, saline intrusion, urbanization, and suspended particulate matter showed the lowest link strength and were reported in fewer publications (Fig. 6, Table 3 ). Through the analysis of extracted literature, from 2003 to 2024, a relationship was established between anthropogenic and natural sources of contamination among all the reported articles on water quality contamination in Nigeria. From the results, it was observed that most studies attributed the occurrence of contamination in water to mining followed by lithogenic and anthropogenic processes (Fig. 7). Table 3 number of publications based on sources of aquatic contamination in Nigeria Pollutant Occurrences per publication Total link strength Heavy metal 67 62 Heavy metals 59 52 Lead 48 47 Cadmium 40 39 Copper 36 36 Chromium 35 35 Zinc 34 34 Escherichia coli 30 26 Geologic sediments 25 24 Polycyclic aromatic hydrocarbons 25 21 Manganese, Nickel 23 23 Arsenic, Iron 22 22 Nitrate 21 20 PAHs, pH 20 20 Electrical conductivity 19 16 Hydrocarbons 18 17 Coliform bacterium, sewage 14 13 Bacterium contamination, seasonal variation 13 13 Microbial contamination 12 10 Effluents, leachates, mercury, mining, petroleum, trace elements, trace metals, turbidity, weathering 11 11 Agriculture, Calcium, total dissolved solids 10 10 Aluminum, crude oil, feces, irrigation, sulfate, wastewater 9 9 Anti-infective agents, chlorine, fecal coliform, industrial waste, land use, NO 3 , Na, population density, wastewater 8 6 Chlorine compounds, DO, geology, industries, Mg, oil spill, K 7 7 Co, F, landfill, leachates, leaching, organic matter, runoff, sulfur compounds, hardness, domestic wastes 6 6 Antibiotic agents, aromatic compounds, bitumen, Cl, combustion, dumpsite, EC, mineralization, municipal solid wastes, organic carbon, phosphates, rainwater, saline intrusion, urbanization, suspended particulate matter 5 4 These results are in agreement with reports by Omeka & Egbueri ( 2022 ), that due to an upsurge in anthropogenic activities in the country such as mining, industrialization, and socio-economic activities, high concentrations of heavy metal(loids) and potentially toxic elements have been reported in drinking water in the country. As reported by Obasi, & Akudinobi ( 2020 ), there is an upward increase in the concentration of Cd, Pb, Cr, Zn, and cobalt in waters from the solid-mineral-rich southern Benue trough in southeastern Nigeria. This has been attributed to the open-cast mining method that is prevalent among most mining operations in the country (Okolo et al. 2018 ; Omeka & Igwe 2021 ). In southeastern Nigeria, Ajala et al. ( 2022 ), provided a critical review of potential sources of heavy metals in water and aquatic fish as well as the human health risks from their consumption. High concentrations of Cd, As, Cu, and Zn were reported in both the drinking water and the shell of fishes. In Nsukka, southeastern Nigeria, they have been a reported high concentration of As, Cd, Hg, and Cr in water; attributed to anthropogenic influxes and run-off from domestic solid and sewage wastes (Nnaji et al. 2023 ). Many studies in southwestern Nigeria have attributed the concentration of heavy metals in drinking water sources to industrialization. According to an extensive literature review carried out by Balogun et al. (2022) in the region, the long-standing sources of groundwater pollution cases have been attributed majorly to industrial effluents. In some parts of the Ibadan metropolis, southwest Nigeria, the concentration of heavy metals in shallow had-dug wells was observed to occur in the order of Zn > Fe > Pb > Cd > Mn; attributed to poor sanitation and industrialization (Ganiyu et al. 2021a ). In the same region, the water quality of shallow aquifers was investigated for their metal content and bacterial load. The water was reportedly polluted due to pathogens and heavy metals; with the heavy metals occurring in the order of Cd > Pb > Zn > Fe > Mn, attributed to varying anthropogenic activities in the area (Ganiyu et al. 2021b ). In the densely populated parts of the southwestern region, such as Lagos, water pollution has been associated with poor waste management practices, lowering of the water table (due to over-abstraction), and influx from industrial effluents (Ogundiran & Afolabi 2008 ). Some studies have also attributed the presence of heavy metals to both geogenic processes (such as weathering and leaching of subsurface geology) and anthropogenic sources (from industrial effluents). A study on the water quality assessment of Asa River in Ilorin, southwestern Nigeria revealed a high concentration of chromite (FeCr 3 O 4 ) and pyrite (FeS) in river sediments, attributed to weathering of subsurface geology. Heavy metals such as Zn, Fe, Cr, and Mn were also reportedly found in elevated concentration due to the influx of industrial effluents in sediments (Adekola & Elleta 2007). Although some studies in the region have reported the presence of heavy metals in water due to vehicular emissions and related repair products, domestic sewage, and effluents (Tijani et al. 2004 ), most have been mostly due to industrial and agricultural effluents (Olaojo et al. 2016; Emenike et al. 2020 ). In the northern and central parts of Nigeria, very few studies have reported water pollution due to industrial effluents Most reports on heavy metal concentration in water have been attributed to mining activities. In the Anka gold mining area of northwestern Nigeria, there have been reported cases of Pb poisoning found in the hairs and nails of children within the vicinity of the mine site; attributed to the ongoing mining activities within the region (Adebwumi 2020). Although a study by Lar et al. (2015) reported the concentration of heavy metals (such as Zn, V, Pb, Cu, Co, Be Cr, As, Cd, Sb, and Se) in drinking water wells due to volcanic eruption in the Panyam volcanic province, majority of the reports on metal concentration in the water have been attributed to mining and poor disposal of its effluents. Pb, Ni, and Hg poisoning have been reported in analyzed surface and groundwater samples from the Bagega gold mine province in Zamfara state, Northwestern Nigeria. The high elevated concentration of carcinogenic elements was attributed to artisanal gold mining (Nuhu et al. 2014 ). In southeastern Nigeria, the sources of water pollution are highly ubiquitous-ranging from a wide range of anthropogenic sources (e.g., mining, poor waste disposal activities, poor hygiene, industries, population increase, land use, etc.) and geogenic sources (precipitation, chemical weathering, weathering, and rock-water interaction) (Omeka et al. 2023; Aghamelu et al. 2022 ). Although most of the reports on contamination have been from both anthropogenic and geogenic activities (Nnorom et al. 2019 ; Edet et al. 2003 ), the majority have been from mining and poor waste disposal activities. Reports from mining have been majorly from the solid mineral-rich zones of the lower Benue trough; with high concentrations of ore minerals occurring in association with Pb-Zn. (Omeka & Igwe 2021 ; Adamu et al. 2015 ). This has been mainly reported from Ebonyi, Enugu, and Cross River states. A review carried out by Umeoguaju et al. ( 2022 ) for two decades (2000–2020) revealed that the concentration of heavy metals in most water sources in southeastern Nigeria is attributed majorly to anthropogenic influences from mining and oil exploration. This is in agreement with a study by Opuene & Agbozu 2008 ) on the heavy metal assessment in fish from Taylor Creek, southern Nigeria. 4.3 Application of neural networks in Water Quality Monitoring: implications for prospects This section addresses the efficacy of artificial neural networks (ANNs) in water quality assessment, the most frequently used ANN models, the input or predictor variables, and the model accuracy based on the coefficient of determination (R 2 ). To achieve this, a literature search was conducted on the application of neural networks in water quality modeling and prediction. The search was conducted on papers from the Scopus and Web of Science databases between 2003–2024, using the keywords “water quality and artificial intelligence” with emphasis on both surface and groundwater. The results from the search were compiled and presented in Table 4 , and plots were generated based on the results in Table 4 . Table 4 Dataset of selected articles from the Web of Science and Scopus databases used for SLR Year Location ANN model Input parameters Highest R 2 (during the testing stage) References 2023 1. Surface and groundwater, Ojoto suburb, southeastern Nigeria 2. Groundwater samples from Osisioma, southeastern Nigeria 3. Groundwater samples from Egbema, southeastern Nigeria 4. surface and groundwater from Okurumutet-Iyamitet mine province, southeastern Nigeria MLP-ANN MLP-ANN MLP-ANN MLP-ANN TH, T, pH, TDS EC, Cl, Ca, SO 4 , Pb, HCO 3 , Zn Fe. Cu, Pb, Fe As, Cr, Benzene, Ethylbenzene m-Xylene, Toluene, and o-Xylene Fe, Zn, Ni, Cd, Cu and Pb HCO 3 − , SO 4 2− , NO 3 − , Cl − , Mg 2+ , Ca 2+ , K + , and Na + 0.878 0.896 0.966 0.861 (Egbueri 2023) (Akakuru et al. 2023 ) (Akakuru et al. 2023 b) (Omeka et al. 2023) 2022 1. Bouregreg watershed, Morocco. 2. Groundwater from Agartala municipality, India. 3. groundwater from El Kharga Oasis, Western Desert of Egypt. BPNN BPNN ANFIS pH and EC Na + , Mg 2+ , Ca 2+ , EC, HCO 3 − , and B EC, pH, T ◦ , TDS, Na + , Mg 2+ , K + , Ca 2+ , Cl − , CO 3 2 − , SO 4 2 − , NO 3 − and HCO 3 − , 0.87 0.990 0.997 (Bilali et al. 2022) (Mallik et al 2022 ) (Ibrahim 2022) 2021 (1) Groundwater wells from Illizi County, Algeria (2) groundwater from El Merk is an oil field in the SOUTHEAST of Algerian BP-NN MLP-ANN pH, TH, EC, TDS, Ca, Mg, Na, K, HCO 3, Cl, SO 4 , and NO 3 TH, NO 3 , and NO 2 0.8957 0.9967 (Kouadri et al. 2021 ) (Kouadri & Samir 2021 ) 2020 1. The Elbe River, Germany 2. Groundwater from the Gaza Strip Palestine 3. Yipin River, China Wavelet and BPNN (W-BPNN) MLP-NN MLP-NN Flow, pH, Fe, and DO Abstraction average rate (AVR), relative humidity (RH), depth from the surface to well screen (DSWS), aquifer thickness (AT), recharge rate (RR), Initial chloride concentration (ICC), and groundwater level (GWL) pH, TP, Temperature, EC, PI, NH 3 -N, COD and TN 0.780 0.9770 0.717 (Li et al. 2020) (Kassem 2020) (Zhu and Heddam 2020 ) 2019 Water samples from the Shivganga River basin, India lakes, Tezpur University, India BP-NN MLP-ANN EC, pH, TH, TDS, Mg, Ca, K, Na, HCO 3 , PO 4 NO 3, Cl, and SO 4 BOD and TSS 0.932 0.783 (Kadam et al. 2019 ) (Ahamad et al. 2019 ) 2018 1. Gorganrood Basin, Iran 2. Gorganrood Basin, Iran ANFIS ANFIS SAR EC 0.99 0.99 (Azad et al. 2018 ) (Azad et al. 2018 ) 2017 1. Hooghly River, India 2. Hooghly River, India 3. Saint John River, Canada 4. Langat River and Klang River, Malaysia NN-CS NN-GA BPNN BPNN Chlorides, turbidity, pH, TA, and residual chlorine TH Chlorides, pH, residual chlorine TA TH, and turbidity TSS BOD, DO, COD, Ph, NH3-N, and TSS - - 0.976 0.7267 (Chatterjee et al. 2017 ) (Chatterjee et al. 2017 ) (El Din and Zhang 2017 ) (Hameed et al. 2017 ) 2016 1.Aji-Chay River, Iran 2. Aji-Chay River, Iran W-ANN W-ANFIS Salinity (EC) Salinity (EC) 0.9960 0.9958 Barzegar et al. 2016 (Barzegar et al. 2016 ) 2015 1. Hilo Bay, Hawaii, USA 2. Hilo Bay, Hawaii, USA 3. Dahan River, Taiwan 4. Nadong River, South Korea MLP W-ANN BPNN W-ANN Salinity, DO, and Temperature Temperature, DO, and Salinity NH 3 –N Water level 0.80 0.967 0.979 (Alizadeh and Kavianpour 2015) (Alizadeh and Kavianpour 2015) (Chang et al. 2015 ) (Seo et al. 2015 ) 2014 Karoon River, Iran Karoon River, Iran MLP-NN MLP-NN DO COD 0.85 0.74 (Emamgholizadeh et al. 2014) (Emamgholizadeh et al. 2014) 2013 Johor River, Malaysia Jishan Lake, China Jishan Lake, China MLP-NN W-ANN ENN EC, TDS, and turbidity DO, Temperature, and pH DO, Temperature, and pH 0.799 - - (Najah et al. 2013 ) (Xu and Liu 2013 ) (Xu and Liu 2013 ) 2012 Kinta River, Malaysia MLP-NN 36 parameters 0.765 (Gazzaz et al. 2012 ) 2011 Nile River, Egypt MLP-NN 33 parameters - (Khalil et al. 2011 ) As observed earlier from the bibliographic coupling map in Fig. 2, artificial neural networks appear to be the most frequently used machine learning model in water quality monitoring and assessment globally, followed only by support vector machines and decision trees. The wide usage of ANN in water quality studies has been attributed to its efficiency and versatility in predictions even in systems with poor computational strength (Egbueri et al. 2023 ; Omeka et al. 2023). According to Ghavidel and Montazeri ( 2014 ), the unique ability of the ANN model to accurately match a broad range of nonlinear variables makes it a widely accepted ML model in most water quality studies. The ANN architecture is designed to mimic the human neural system; with the unique ability to quickly learn and send signals about a range of linear and nonlinear datasets through its interwoven parts known as “neurons” (Ozel et al. 2020 ). This makes it stand out among other water quality modeling techniques. Water quality data are usually nonstationary, random, nonlinear, and unpredictable (Beven, 2016 ). This means that the relationship between water quality and its controlling factors will tend to vary over time (as a result of varying anthropogenic and seasonal fluctuations), hence, the application of predictive models that work on historical data will become ineffective over time (Khan et al., 2021 ), hence more focus has been put on ANN due to its unique ability to process complex datasets. 4.3.1 Model utilization It can be observed from Fig. 8 that the multilayer perceptron artificial neural network (MLP-NN) and back-propagated neural network (BP-NN) are the most frequently used neural network algorithms in water quality monitoring and prediction within the period of 2003–2024. This was closely followed by the adaptive neuro-fuzzy inference system (ANFIS), and other neural networks such as the Wavelet and BPNN (WNN) hybrid model, Neural Network trained by Cuckoo Search (NNCS), Neural Network trained by Genetic Algorithm (NN-GA), Elman Neural Network (ENN) and Wavelet-Adaptive Neural Fuzzy Interference System (WANFIS) in decreasing order. The high frequency of the MLP-NN and BP-NN in water quality monitoring and prediction is in line with a study by Rajaee et al. ( 2020 ). The wide acceptance of these algorithms has also been attributed to their ability to attain high modeling accuracy with fewer input parameters compared to other white-box models (Yetilmezsoy et al. 2011 ; Ighalo et al. 2020 ; Omeka 2023). 4.3.2 Model validation accuracy In this assessment, only studies that used the coefficient of determination (R 2 ) as a validation metric were considered. The exclusion was made for the root mean square error (RMSE). The RMSE was excluded because of the difficulty of directly comparing the units to different parameters. This implies that its accuracy is dependent on the size (value) of the parameter being considered. This would mean that a parameter with a large numerical value will show a high RMSE value; resulting in inaccuracy (Ighalo et al. 2020 ). Unlike the RMSE, the R 2 is a measure of the extent of variation of a particular dataset, therefore giving better accuracy of parameters in a particular dataset (Adeniyi et al. 2019). An observation of Table 4 shows that the adaptive neuro-fuzzy inference system (ANFIS), Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS), and the Wavelet and BPNN (W-NN) hybrid models are the most accurate neural networks for both surface and groundwater monitoring and prediction. These findings are in agreement with the observation of Ighalo et al. ( 2020 ) on the frequently used neural networks for surface water prediction between 2011–2020. This study has revealed that although the MLP-NN and BP-NN seem to be the most popularly used algorithms in water quality prediction, however, they have low modeling accuracy compared to other hybrid models. The high accuracy of the W-ANFIS and W-NN hybrid models has been attributed to the in-cooperation of the wavelet decomposition of covariates (predictor variables) into detail and approximate components (Barzegar et al. 2016 ). On the other hand, the high accuracy of the ANFIS is because its architecture integrates the learning ability of both the Fuzzy Inference System and ANN (Emamgholizadeh et al. 2014 ). 4.3.3 Predicted water quality parameters In this section, the most frequently analyzed water quality parameters for both surface and groundwater have been discussed. An observation of Table 4 shows that temperature, dissolved oxygen (DO), pH, Electrical conductivity (EC; as a function of salinity), and total dissolved solids (TDS) are the most frequently investigated parameters in water quality prediction. A possible reason for their high prevalence is due to significance in the determination of the overall water quality in both surface and groundwater sources. Ighalo et al. ( 2020 ) have attributed the prevalence of DO, TDS, and pH to the high availability and low cost of measuring equipment. Moreover, these parameters are readily measured in situ before laboratory analysis because of their high susceptibility to surface environmental changes (Igwe and Omeka 2021 ). 5. Conclusion With the aid of meta-analysis from Scopus and Web of Science databases and bibliometric coupling for data extraction and visualization, the present study has explored the recent trends to more hypothetical and advanced model simulation in the application of machine learning (ML) models in water quality monitoring and prediction in Nigeria for the last two decades (2003–2024). Emphasis has been made on the type of ML frequently used, the accuracy of the model in water quality prediction, and the frequently monitored water quality parameters. It was revealed that the use of hybridized ML models in water quality prediction has not been well explored globally; a majority of the prediction has been based on the use of artificial neural networks (ANN). The adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models were observed to be the most accurate ANN algorithms, in prediction. Based on analysis of the literature, temperature, dissolved oxygen (DO), pH, conductivity (EC) and total dissolved solids (TDS) appear to be the most frequently predicted parameters. China and the United States had the highest number of publications concerning the application of ML in water quality prediction; while Nigeria appears among the countries grossly lacking behind. Based on a literature analysis of the sources of contaminants in aquatic systems in Nigeria, heavy metals (Pb, Cd, Cu, Cr, and Zinc) were reported in more publications; linked majorly to mining and industrial effluents. This result implies that the increasing rate of anthropogenic activities coupled with climatic variation will continue to render water quality data complex; making the use of conventional assessment methods and standalone ML models inefficient. Therefore, future water quality monitoring and prediction studies in developing countries, especially in Nigeria- where data on environmental monitoring is lacking- should take advantage of the rapidly evolving field of machine learning; with more emphasis placed on the hybridized machine learning algorithms. Although this study only considered the Scopus and Web of Science databases in its analysis, it is recommended that future studies should explore other databases such as Science Direct, PubMed, Cochrane Library, Lens, and Dimensions. Other bibliometric analytic techniques such as the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) should be explored as well. Abbreviations SAR Sodium Absorption Ratio RMSE Root Mean Square Error MLP-ANN Multi-layer perceptron neural network TP Total Phosphorus NN-CS Neural Network trained by Cuckoo Search Algorithm TS Total Solids TDS Total Dissolved Solids ANFIS Adaptive Neuro-Fuzzy Inference System BOD Biochemical Oxygen Demand TN Total Nitrogen TA Total alkalinity TH Total Hardness COD Chemical Oxygen Demand DO Dissolved Oxygen TSS Total Suspended Solids BPNN Back Propagation Neural Network R 2 Coefficient of Determination PI Permanganate Index SVM Support Vector Machine W-BPNN Wavelet and back propagation neural network W-ANFIS Wavelet and Adaptive Neuro-Fuzzy Inference System Declarations Author contribution Michael E. Omeka Michael E. Omeka conceived and designed the project, conducted bibliometric modeling, and wrote the manuscript. Manuscript review and editing were also carried out by the author. Data availability Not applicable. Consent for publication Not applicable. Consent to participate Not applicable. Competing interests The author declares no competing interests regarding this research Funding No funding was received for this research References Adamu CI, Nganje TN, Edet A (2015) Heavy metal contamination and health risk assessment associated with abandoned barite mines in Cross River State, southeastern Nigeria, vol 3. Environmental Nanotechnology, monitoring & management, pp 10–21 Adekola FA, Eletta OAA (2007) A study of heavy metal pollution of Asa River, Ilorin. Nigeria; trace metal monitoring and geochemistry. Environ Monit Assess 125:157–163 Adeniyi AG, Ighalo JO, Marques G (2020) Utilisation of machine learning algorithms for the prediction of syngas composition from biomass bio-oil steam reforming. Int J Sustain Energy. https://doi.org/10.1080/14786451.2020.1803862 Adewumi AJ (2020) Contamination, sources and risk assessments of metals in media from Anka artisanal gold mining area. Northwest Nigeria Sci Total Environ 718:137235 Agbasi JC, Egbueri JC, Ayejoto DA, Unigwe CO, Omeka ME, Nwazelibe VE, Ighalo JO, Pande CB, Fakoya AA (2023) The impact of seasonal changes on the trends of physicochemical, heavy metal and microbial loads in water resources of southeastern Nigeria: A critical review. In: Egbueri JC, Ighalo JO, Pande CB (eds) Climate change impacts on Nigeria. Springer, Cham. https://doi.org/10.1007/978-3-031-21007-5_25 Aghamelu OP, Omeka ME, Unigwe CO (2022) Modeling the vulnerability of groundwater to pollution in a fractured shale aquifer in SE Nigeria using information entropy theory, geospatial, and statistical modeling approaches. Model Earth Syst Environ, 1–22 Ahamad KU, Raj P, Barbhuiya NH, Deep A (2019) Surface water Ajala OA, Oke MR, Ajibade TF, Ajibade FO, Adelodun B, Ighalo JO, Silva LF (2022) Concentrations, bioaccumulation, and health risk assessments of heavy metals in fishes from Nigeria’s freshwater: a general overview. Environ Sci Pollut Res 29(55):82660–82680 Akakuru OC, Adakwa CB, Ikoro DO, Eyankware MO, Opara AI, Njoku AO, Usman A (2023) Application of artificial neural network and multi-linear regression techniques in groundwater quality and health risk assessment around Egbema, Southeastern Nigeria. Environ Earth Sci 82(3):77 Akakuru OC, Njoku UB, Obinna-Akakuru AU, Akudinobi BE, Obasi PN, Aigbadon GO, Onyeanwuna UB (2023) Non-carcinogenic health risk assessment and predicting of pollution indexing of groundwater around Osisioma, Nigeria, using artificial neural networks and multi-linear modeling principles. Stochastic Environmental Research and Risk Assessment, pp 1–31 Al Aani S, Bonny T, Hasan SW, Hilal N (2019) Can machine language and artificial intelligence revolutionize process automation for water treatment and desalination? Desalination 458:84–96 Alizadeh MJ, Kavianpour MR (2015) Development of wavelet-ANN models to predict water quality parameters in Hilo Bay Pacific Ocean. Mar Pollut Bull 98:171–178 Ayejoto DA, Egbueri JC, Agbasi JC, Omeka ME, Unigwe CO, Nwazelibe VE, Ighalo JO, Pande CB (2023) Influence of Seasonal Changes on the Quality of Water Resources in Southwestern Nigeria: A Review. In: Egbueri JC, Ighalo JO, Pande CB (eds) Climate change impacts on Nigeria. Springer, Cham. https://doi.org/10.1007/978-3-031-21007-5_22 Azad A, Karami H, Farzin S, Saeedian A, Kashi H, Sayyahi F (2018) Prediction of water quality parameters using ANFIS optimized by intelligence algorithms (case study: Gorganrood River). KSCE J Civ Eng 22:2206–2213 Barzegar R, Adamowski J, Moghaddam AA (2016) Application of wavelet-artificial intelligence hybrid models for water quality prediction: a case study in Aji-Chay River, Iran. Stoch Environ Res Risk Assess 30:1797–1819 Beven K (2016) Advice to a young hydrologist. Hydrol Process 30(20):3578–3582. https://doi.org/10.1002/hyp.10879 Chang F-J, Tsai Y-H, Chen P-A, Coynel A, Vachaud G (2015) Modeling water quality in an urban river using hydrological factors– Data-driven approaches. J Environ Manage 151:87–96 Chatterjee S, Sarkar S, Dey N, Ashour AS, Sen S, Hassanien AE (2017) Application of cuckoo search in water quality prediction using artificial neural network. Int J Comput Intell Stud 6:229–244 Deng T, Duan HF, Keramat A (2022) Spatiotemporal characterization and forecasting of coastal water quality in the semi-enclosed Tolo Harbour based on machine learning and EKC analysis. Eng Appl Comput Fluid Mech 16(1):694–712. https://doi.org/10.1080/19942060.2022.2035257 Edet AE, Merkel BJ, Offiong OE (2003) Trace element hydrochemical assessment of the Calabar Coastal Plain Aquifer, southeastern Nigeria using statistical methods. Environ Geol 44:137–149 Egbueri JC, Agbasi JC (2022) Data-driven soft computing modeling of groundwater quality parameters in southeast Nigeria: comparing the performances of different algorithms. Environ Sci Pollut Res 29(25):38346–38373 Egbueri JC, Igwe O, Omeka ME, Agbasi JC (2023) Development of MLR and variedly optimized ANN models for forecasting the detachability and liquefaction potential index of erodible soils. Geosyst Geoenvironment 2(1):100104 Egbueri JC (2022) Incorporation of information entropy theory, artificial neural network, and soft computing models in the development of integrated industrial water quality index. Environ Monit Assess 194(10):693 El Bilali A, Taleb A (2020) Prediction of irrigation water quality parameters using machine learning models in a semi-arid environment. J Saudi Soc Agricultural Sci. https://doi.org/10.1016/j.jssas.2020.08.001 El Din ES, Zhang Y (2017) Improving the accuracy of extracting surface water quality levels (Swqls) using remote sensing and artificial neural network: a case study in the saint john river, Canada. XLII–4/W4:245–249International archives of the photogrammetry. Remote sensing & spatial information sciences Emamgholizadeh S, Kashi H, Marofpoor I, Zalaghi E (2014) Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models. Int J Environ Sci Technol 11:645–656 Emenike PC, Tenebe IT, Neris JB, Omole DO, Afolayan O, Okeke CU, Emenike IK (2020) An integrated assessment of land-use change impact, seasonal variation of pollution indices, and human health risk of selected toxic elements in sediments of River Atuwara. Nigeria Environ Pollution 265:114795 Encyclopædia, Britannica (2023) Inc Ganiyu SA, Mabunmi AA, Olurin OT, Adeyemi AA, Jegede OA, Okeh A (2021b) Assessment of microbial and heavy metal contamination in shallow hand-dug wells bordering Ona River, Southwest Nigeria. Environ Monit Assess 193:1–22 Ganiyu SA, Oyadeyi AT, Adeyemi AA (2021a) Assessment of heavy metals contamination and associated risks in shallow groundwater sources from three different residential areas within the Ibadan metropolis, southwest Nigeria. Appl Water Sci 11(5):81 Gazzaz NM, Yusof MK, Aris AZ, Juahir H, Ramli MF (2012) Artificial neural network modeling of the water quality index for Kinta River (Malaysia) using water quality variables as predictors. Mar Pollut Bull 64:2409–2420 Geetha S, Gouthami S (2016) Internet of things enabled real-time water quality monitoring system Smart. Water 2:1. https://doi.org/10.1186/s40713-017-0005-y Ghavidel SZZ, Montazeri M (2014) Application of different data-driven methods for the prediction of total dissolved solids in the Zarinehroud basin. Stoch Environ Res Risk Ass 28(8):2101–2118. https://doi.org/10.1007/s00477-014-0899-y Hameed M, Sharqi SS, Yaseen ZM, Afan HA, Hussain A, Elshafe A (2017) Application of artificial intelligence (AI) techniques in water quality index prediction: a case study in the tropical region, Malaysia. Neural Comput Appl 28:893–905 Hasan R, Raghav A, Mahmood S, Hasan MA (2011) Artificial Intelligence in River Quality Assessment. In: International Conference on Information Management, Innovation Management and Industrial Engineering, 2011. IEEE, pp 491–495 Ibrahim H, Yaseen ZM, Scholz M, Ali M, Gad M, Elsayed S, Khadr M, Hussein H, Ibrahim HH, Eid MH et al (2023) Evaluation and Prediction of Groundwater Quality for Irrigation Using an Integrated Water Quality Indices, Machine Learning Models and GIS Approaches: A Representative Case Study. Water 15:694. https://doi.org/10.3390/w15040694 Ifediegwu SI (2022) Assessment of groundwater potential zones using GIS and AHP techniques: a case study of the Lafa district, Nasarawa State, Nigeria. Appl Water Sci 12:10. https://doi.org/10.1007/s13201-021-01556-5 Ighalo JO, Adeniyi AG, Marques G (2020) Artificial intelligence for surface water quality monitoring and assessment A systematic literature analysis. Model Earth Syst Environ. https://doi.org/10.1007/s40808-020-01041-z Igwe O, Omeka ME (2021) Hydrogeochemical and pollution assessment of water resources within a mining area, SE Nigeria, using an integrated approach. Int J Energy Water Resour 1–22 Kadam AK, Wagh VM, Muley AA, Umrikar BN, Sankhua RN (2019) Prediction of water quality index using artificial neural network and multiple linear regression modeling approaches in Shivganga River basin. Model Earth Syst Environ India. https://doi.org/10.1007/s40808-019-00581-3 Khalil B, Ouarda T, St-Hilaire A (2011) Estimation of water quality characteristics at ungauged sites using artificial neural networks and canonical correlation analysis. J Hydrol 405:277–287 Khan N, Shahid S, Ismail TB, Behlil F (2021) Prediction of heat waves over Pakistan using support vector machine algorithm in the context of climate change. Stoch Environ Res Risk Assess 1–19. https://doi.org/10.1007/s00477-020-01963-1 Kouadri S, Elbeltagi A, Islam ARM, Kateb S (2021) Performance of machine learning methods in predicting water quality index based on the irregular data set: application on Illizi region (Algerian southeast). Appl Water Sci 11(12):1–20 Kouadri S, Samir K (2021) Hydro-chemical study with geospatial analysis of groundwater Quality Illizi Region, South-East of Algeria. Iran J Chem Chem Eng (IJCCE) 40(4):1315–1333. https://doi.org/10.30492/ijcce.2020.39800 Kouadri S, Samir K (2021) Hydro-chemical study with geospatial analysis of groundwater Quality Illizi Region, South-East of Algeria. Iran J Chem Chem Eng (IJCCE) 40(4):1315–1333. https://doi.org/10.30492/ijcce.2020.39800 Lar UA, Gusikit RB (2015) Environmental and health impact of potentially harmful elements distribution in the Panyam (Sura) volcanic province, Jos Plateau, Central Nigeria. Environ Earth Sci 74:1699–1710 Mallik S, Chakraborty A, Mishra U, Paul N (2022) Prediction of irrigation water suitability using geospatial computing approach: a case study of Agartala city, India. Environmental Science and Pollution Research, pp 1–16 Martins P, Forstinus NO, Ikechukwu NE (2016) Water and Waterborne Diseases: A Review Water and Waterborne Diseases: A Review. on. JANUARY Mohammadpour R, Shaharuddin S, Zakaria N, Ghani A, Vakili M, Chan N (2016) Prediction of water quality index in free surface constructed wetlands. Environ Earth Sci 75:1–12. https://doi.org/10.1007/s12665-015-4905-6 Morounke SG, Ayorinde JB, Benedict AO, Adedayo FF, Adewale FO, Oluwadamilare I, Sokunle SS, Benjamin A (2017) Epidemiology and Incidence of Common Cancers in Nigeria. J Cancer Biol Res 5(3):1105 Najah A, El-Shafe A, Karim OA, El-Shafe AH (2013) Application of artificial neural networks for water quality prediction. Neural Comput Appl 22:187–201 Nayak S (2020) Assessment of water quality of Brahmani river using correlation and regression analysis Nigerian National System of Cancer Registries (2013) Federal Ministry of Health of Nigeria, Federal Secretariat Complex Shehu Shagari Way, Garki, Abuja P.M.B. 083, GarkiAbuja ISBN: 978-978-955-030-2 Nnaji CC, Chibueze CV, Mama CN, Alum OL, Afangideh CB (2023) Impact of anthropogenic and environmental conditions on surface run-off quality: a case study of Nsukka, Eastern Nigeria. International Journal of Environmental Science and Technology, pp 1–12 Nnorom IC, Ewuzie U, Eze SO (2019) Multivariate statistical approach and water quality assessment of natural springs and other drinking water sources in Southeastern Nigeria. Heliyon, 5(1), e01123 Nuhu AA, Sallau MS, Majiya MH (2014) Heavy metal pollution: the environmental impact of artisanal gold mining on Bagega village of Zamfara state, Nigeria. Res J Pharm Biol Chem Sci 5(6):306–313 Obasi PN, Akudinobi BB (2020) Potential health risk and levels of heavy metals in water resources of lead–zinc mining communities of Abakaliki, southeast Nigeria. Appl Water Sci 10:184. https://doi.org/10.1007/s13201-020-01233-z Ogundiran OO, Afolabi TA (2008) Assessment of the physicochemical parameters and heavy metals toxicity of leachates from the municipal solid waste open dumpsite. Int J Environ Sci Technol 5:243–250 Okamkpa JR, Omeka ME, Igwe O, Iyiokwu MU (2022) An integrated geochemical and spatiotemporal assessment of groundwater resources within an industrial suburb, Southeastern Nigeria. Int J Energy Water Resour 1–20. https://doi.org/10.1007/s42108-022-00183-3 Okolo CC, Oyedotun TDT, Akamigbo FOR (2018) Open cast mining: a threat to water quality in the rural community of Enyigba in south-eastern Nigeria. Appl water Sci 8:1–11 Olaojo AA, Oladunjoye MA, Sanuade OA (2018) Geoelectrical assessment of polluted zone by sewage effluent in University of Ibadan campus southwestern Nigeria. Environ Monit Assess 190:1–12 Omeka ME, Egbueri JC (2022) Hydrogeochemical assessment and health-related risks due to toxic element ingestion and dermal contact within the Nnewi-Awka urban areas, Nigeria. Environ Geochem Health 1–29. https://doi.org/10.1007/s10653-022-01332-7 Omeka ME, Ezugwu AL, Agbasi JC, Egbueri JC, Abugu HO, Aralu CC, Ucheana IA (2024) A review of the status, challenges, trends, and prospects of groundwater quality assessment in Nigeria: an evidence-based meta-analysis approach. Environmental Science and Pollution Research, pp 1–24 Omeka ME, Egbueri JC, Unigwe CO (2022a) Investigating the hydrogeochemistry, corrosivity, and scaling tendencies of groundwater in an agrarian area (Nigeria) using graphical, indexical, and statistical modeling. Arab J Geosci 15(13):1–24. https://doi.org/10.1007/s12517-022-10514-7 Omeka ME (2022) Evaluation and prediction of irrigation water quality of an agricultural district, SE Nigeria: an integrated heuristic GIS-based and machine learning approach. Environ Sci Pollut Res. https://doi.org/10.1007/s11356-022-25119-6 Opuene K, Agbozu IE (2008) Relationships between heavy metals in shrimp (Macro brachium felicinum) and metal levels in the water column and sediments of Taylor Creek Ozel HU, Gemici BT, Gemici E, Ozel HB, Cetin M, Sevik H (2020) Application of artificial neural networks to predict the heavy metal contamination in the Bartin River. Environ Sci Pollut Res. https://doi.org/10.1007/s11356-020-10156-w Pour SH, Wahab AKA, Shahid S (2020) Physicalempirical models for prediction of seasonal rainfall extremes of Peninsular Malaysia. Atmos Res 233:104720. https://doi.org/10.1016/j.atmosres.2019 Kalamdhad A, Singh J, Dhamodharan K (eds) Advances in waste management. Springer, Singapore. https://doi.org/10.1007/978-981-13-0215-2_15 Rajaee T, Khani S, Ravansalar M (2020) Artificial intelligence-based single and hybrid models for prediction of water quality in rivers: a review. ChemometrIntell Lab Syst 200:103978 Ricolfi L, Barbieri M, Muteto PV, Nigro A, Sappa G, Vitale S (2020) Potential toxic elements in groundwater and their health risk assessment in drinking water of Limpopo National Park, Gaza Province, Southern Mozambique. Environmental Geochemistry and Health, pp 1–13 Seo Y, Kim S, Kisi O, Singh VP (2015) Daily water level forecasting using wavelet decomposition and artificial intelligence techniques. J Hydrol 520:224–243 Sun K, Rajabtabar M, Samadi S, Rezaie-Balf M, Ghaemi A, Band S, Mosavi A (2021) An integrated machine learning, noise suppression, and population-based algorithm to improve total dissolved solids prediction. Eng Appl Comput Fluid Mech 15(1):251–271. https://doi.org/10.1080/19942060.2020.1861987 Tijani MN, Jinno K, Hiroshiro Y (2004) Environmental impact of heavy metals distribution in water and sediments of Ogunpa River, Ibadan area, southwestern Nigeria. J Min Geol 40(1):73–83 Umeoguaju FU, Akaninwor JO, Essien EB, Amadi BA (2022) Heavy metal profile of surface and groundwater samples from the Niger Delta region of Nigeria: a systematic review and meta-analysis. Environ Monit Assess 194:1–37 van Ech NJ, Waltman L (2020) VOSviewer Manual. Universiteit Leiden, The Centre for Science and Technology Studies, Meaningful metrics. https://www.vosviewer.com/ . Accessed February 2023 World Health Organisation/United Nations Children Fund (WHO/UNICEF) 2010. Progress on drinking water and sanitation Xiao Y, Liu K, Hao Q, Li J, Zhang Y, Cui W, Pei Q (2021) Hydrogeochemical features and genesis of confined groundwater and health perspectives for sustainable development in urban Hengshui. North China Plain, J Chem Xu L, Liu S (2013) Study of short-term water quality prediction model based on wavelet neural network. Math Comput Model 58:807–813 Yetilmezsoy K, Ozkaya B, Cakmakci M (2011) Artificial intelligence-based prediction models for environmental engineering. Neural Netw World 21:193 Zhu S, Heddam S (2020) Prediction of dissolved oxygen in urban rivers at the Three Gorges Reservoir, China: extreme learning machines (ELM) versus artificial neural network (ANN). Water Qual Res J 55:106–118 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4429965","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":307370530,"identity":"ac4c004b-2261-4c1d-baf3-dbb1811d66c4","order_by":0,"name":"Michael E. Omeka","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYDACZgY2IGnBIAHifABiNnbitEiAtTDOAGlhJmwPQgszD8QQ/EC3nfnZgx8VEnKSM3KfSdv82ibPx8zA+OFjDm4tZofZzA17zkgYS0ukm0nn9t02bGNmYJacuQ2fFh42Cd42icR5Emls0rk9txmBWtiYeQlokfz7T6IerMWy57Y9UVqkeRskEqRBWhh+3E4kQgubmbTMMQnDmT3PmC17G24ntzEzNuP3y/nDzyTf1NjISxxPY7zx489t2/ntzQc/fMSjBRUwtoHJBmLVg8AfUhSPglEwCkbBSAEAGDtEfPUJYosAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0003-0405-8616","institution":"University of Calabar","correspondingAuthor":true,"prefix":"","firstName":"Michael","middleName":"E.","lastName":"Omeka","suffix":""}],"badges":[],"createdAt":"2024-05-16 09:25:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4429965/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4429965/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":58044695,"identity":"d7ac4ea3-1712-422e-91b8-4be1e6048998","added_by":"auto","created_at":"2024-06-10 11:18:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":393336,"visible":true,"origin":"","legend":"\u003cp\u003eLocation Map of Nigeria showing the different geographic locations (source: Encyclopædia Britannica, 2023)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4429965/v1/ed530578ad1340e0dbd0b22a.png"},{"id":59261905,"identity":"8c7a4555-ea9e-4fdc-8c5e-860e6c130522","added_by":"auto","created_at":"2024-06-28 09:56:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1229303,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4429965/v1/a831e5e4-70e8-490e-9f75-21968499b529.pdf"}],"financialInterests":"","formattedTitle":"Exploring the recent trends, progresses, and challenges in the application of Artificial intelligence in water quality assessment and monitoring in Nigeria: A systematic review","fulltext":[{"header":"Highlights","content":"\u003cul\u003e\n \u003cli\u003eGlobally, the use of hybrid ML models in water quality prediction has not been well explored; a majority of the prediction has been based on the use of artificial neural networks (ANN).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eThe adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models are the most accurate in prediction.\u003c/li\u003e\n \u003cli\u003eNigeria is grossly lagging in the application of ML in water quality prediction. This limitation is largely attributed to inadequate data on environmental monitoring.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eApproximately 75% of the earth's crust is made up of water. About 68.7 and 30.3% of the total water mass, respectively, are encased in ice caps or glaciers and ocean, respectively. Fewer than 1% are regarded as fresh (Al Aani et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The quality of water supplies has been declining dramatically in recent decades. This reflects both the increase in human activity and the variability of the global climate (Ayejoto et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Agbasi et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Natural hydrological, hydrogeologic, and geogenic processes also influence water quality because they alter the water's chemistry (Xiao et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The predominant pollution precursors that are unique to the location, either locally or regionally, will have a major influence on how quickly the water quality is altered, or polluted. Furthermore, the mechanisms and origins of declining water quality are typically numerous and intricate (Omeka, 2023). Examples of natural process variations in aquifer characteristics include water-rock interaction, chemical speciation, changes in stream dynamics in response to pollutant input, and variations in rainfall intensity. Both point (produce from mine and industrial effluents) and non-point (runoff from crop fields, effluents from town waste dumps seeping into groundwater, runoff from open defecation, etc.) sources can contribute to anthropogenic processes (Xiao et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Okamkpa et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Because of this intricacy, evaluating the quality of the water usually involves difficult and thorough data gathering, laboratory investigation, and data analysis. Both conventional and non-traditional methods have been used to monitor and assess the quality of the water in recent decades. The non-conventional approach uses multiple quantitative metrics, parametric and non-parametric models of statistics, health risk assessment mathematical models, and spatiotemporal models (by employing GIS) for quality checks, whereas traditional methods depend on the application of local and international benchmark standards. However, the disadvantage of using numerous numerical models is that it usually takes a lot of effort and expert knowledge to compute the sub-indices for model output. This will therefore lead to inaccuracies in the output of the results and consequently affect the decision-making process, which could have a significant effect on the economy (Kouadri and Samir, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWith a population density of more than 1.9\u0026nbsp;million, Nigeria has the largest human population of all the black African nations (Ighalo et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). A startling 66.3\u0026nbsp;million people, however, lack access to portable, safe drinking water (Ighalo et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Okamkpa et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Additionally, the accessible water sources are under a lot of stress because of the high population density and rising socioeconomic activity. Additionally, it has been established that seasonal variations in temperature and rainfall intensity brought on by climate change significantly alter the quality of water resources (Kouadri and Samir, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Omeka et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In this regard, there will be an indirect rise in the variability of climate change as a result of the growing anthropogenic activities in the area (such as hydrocarbon exploration, industrialization, land usage, and population growth). The frequency of aquifer recharge/discharge rates will therefore be impacted by this (Omeka and Egbueri \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ricolf et al., 2020). Therefore, efforts and focus should be directed towards developing a relationship between the various water quality precursors unique to the area in order to provide effective and sustainable water management and preservation. This will require the development of innovative, economical methods for comprehensive, reliable assessment and monitoring of water quality. Artificial Intelligence (AI) can be utilised to achieve this. In this regard, there will be an indirect rise in the variability of climate change as a result of the growing anthropogenic activities in the area (such as hydrocarbon exploration, industrialization, land usage, and population growth). The frequency of aquifer recharge/discharge rates will therefore be impacted by this (Omeka and Egbueri \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ricolf et al., 2020). Therefore, efforts and focus should be directed towards developing a relationship between the various water quality precursors unique to the area in order to provide effective and sustainable water management and preservation. This will require the development of innovative, economical methods for comprehensive, reliable assessment and monitoring of water quality. Artificial Intelligence (AI) can be utilised to achieve this.\u003c/p\u003e \u003cp\u003eGlobally, there has been some progress from the conventional water quality assessment and monitoring techniques towards artificial intelligence (AI) and Internet of Things (IoT) technologies (Geetha and Gouthami \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). AI was first introduced in the early 1950s and has since gone through significant modifications and gained applications in various fields in data simulation and forecasting of events outcomes (Al Aani et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The use of AI in water quality assessment and monitoring started in the early 1990s and has since then gained the attention of many researchers and environmental stakeholders (Hasan et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). As a result of the complexity, ubiquity, and nonlinearity usually associated with water quality datasets, a challenge may arise in the prediction of water quality through the use of conventional approaches. This is attributed to increasing sources of anthropogenic pollution sources and varying climatic changes. Therefore, developing a model that can efficiently predict and monitor water quality remains crucial, especially in the present Anthropocene age where there seems to be no sight of solution to the impending global pollution crisis.\u003c/p\u003e \u003cp\u003eAs part of her industrial revolution plan, Nigeria projects to increase its industrial output from 4\u0026ndash;10% of GDP by 2020 (NIRP 2014). This will mean that more pressure will be put on the environment and its valued components (such as water, air, and soil). The big question remains how the country will maintain a balance between its industrial revolution projections and the protection of valued environmental components. It is feared that with the continuous increase in population and impeding industrial revolution, Nigeria will be put under greater strain in terms of climatic variations. Additionally, the quantity and quality of water resources will experience a further decline and continuous shifts (Ayejoto et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). It becomes imperative therefore for more conscious action plans to be put toward the assessment and prediction of the quality of water resources for informed decision-making and amelioration of contaminated sites through the use of cutting-edge approaches. As a response, in recent times, some studies have tilted towards the use of AI in water quality prediction in Nigeria, with some recorded progress. On a global scale, Ighalo et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) have carried out a comprehensive review of the application of different AI models in the assessment and monitoring of surface water in the last decade (2011\u0026ndash;2020). Their study found that two AI models- Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) were the most applied models for water quality assessment and monitoring within the period. The study also showed that globally, Southeast Asia and Iran are among the countries with the most studies on the use of these models in water quality prediction.\u003c/p\u003e \u003cp\u003eStudies in Nigeria on the application of AI in water quality prediction are very few in the literature. Among these studies, most have focused on the use of standalone machine learning (ML) algorithms (Egbueri \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Egbueri and Agbasi \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Omeka 2023) while some have integrated ML and AHP in water quality prediction and prioritization (Ifediegwu \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Omeka et al. 2023). For instance, Egbueri (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) has integrated a multilayer perceptron artificial neural network (MLP-ANN) and multiple linear regression (MLR) with corrosion and scaling potential indices for the prediction of industrial water quality in southeastern Nigeria. This study concluded that the incorporation of ML with indexical models can provide robustness in industrial water quality assessment and management. In another study in the region, MLR and MLP-ANN were incorporated to predict groundwater quality using two numerical models- overall index of pollution (OIP) and water quality index (WQI) as predicted or output variables. The MLP-ANN showed higher efficiency of prediction as lower modeling errors were observed in their validation metrics (Egbueri and Agbasi, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In an agricultural mine district in southeastern Nigeria, a composite irrigation suitability zonation map was generated through the integration of a GIS-based analytical hierarchy process (AHP) and ML algorithms (Omeka et al. 2023). The model successfully identified Mg, HCO\u003csub\u003e3\u003c/sub\u003e, Cl, and SO\u003csub\u003e4\u003c/sub\u003e as the major contributors to the deterioration of the irrigation water quality in the region. In a similar study, simple kriging (SK) and inverse distance weighting (IDW) GIS-based spatiotemporal models have been used in combination with supervised ML algorithms for a composite assessment of the irrigation water quality for crop yield in a mining region in southeastern Nigeria (Omeka, 2023). This study revealed that soil is more susceptible to sodicity through the application of surface water for irrigation due to the higher pH of the surface water compared to groundwater.\u003c/p\u003e \u003cp\u003eIndeed, some efforts have been put toward the use of AI in water quality monitoring in Nigeria, especially, in the southeastern parts. It is however recommended that more research be channeled toward this area by embracing the application of hybridized (ensemble) ML algorithms. Due to the nonlinearity and randomness of water quality data, and the impending climate variability, the need for the use of improved ML models is imperative. The relationship between water quality and its controlling factors is known to vary over time (due to variations in anthropogenic factors and seasonal fluctuations), hence, the use of a predictive model that functions under the framework of historical data becomes inefficient over time (Khan et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The use of hybridization in other parts of the world has shown very promising results. For instance, in Pakistan, the forward-rolling artificial intelligence algorithm has been used to generate a model that will show resiliency to variability in climate change in a hydrologic setting (Khan et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Two ML algorithms- the recursive feature elimination and support vector machine learning (RFESVM) have been combined to monitor the variability in rainfall in Malaysia (Pour et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In Penang, Malaysia, the support vector machines (SVM), backpropagation neural network (BP-NN), and radial basis function neural network (RB-NN) have been incorporated to monitor the quality of surface water in constructed wetlands (Mohammadpour et al. 2015). Nayak (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) combined and compared the use of multiple linear regression (MLP) with correlation models in assessing the trend in the surface water quality of the Brahmani River. The MLR models were found to show better results in water quality trends. In the semi-enclosed Tolo Harbour water bay in Hon Kong, random forest machine (RF) and ANN machine learning models were combined to assess the spatial variations in eutrophication in the water. The ensemble model identified suspended solids, dissolved nutrients, and organic pollutants as significant contributors to water quality (Deng et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In Iran, Sun et al. (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), have combined two hybrid ML models- the multivariate adaptive regression spline (MARS) and intrinsic time-scale decomposition (ITD) with the multi-step supervised-based machine learning approach (MSMLEA) for the prediction of the water quality of the Tajan River in Iran. From the results, the hybrid ML models showed higher performance accuracy in water quality prediction.\u003c/p\u003e \u003cp\u003eNigeria has seen serious incidences of cholera, diarrhoea, and other gastrointestinal ailments in the past few decades as a result of consuming polluted water (Martins et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). In the joint WHO/UNICEF (2012) report on monitoring programmes for sanitation and water supply, Nigeria comes in third place, right after China and India, among the nations without access to clean drinking water. Waterborne illnesses are thought to be the cause of approximately 130,000 child deaths in Nigeria each year (WHO/UNICEF 2012). If immediate action is not taken, it is anticipated that this number would rise by 2040 (WHO/UNICEF 2012). As per the 2016 report on cancer patterns in Nigeria by the Nigeria National System of Cancer Registries, around 80% of the country's cancer risks are attributable to pollution. Thirty percent of this are related to drinking tainted water (Morounke et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). This seems to support the 1996 Harvard report on cancer prevention, which found that of all the risk factors for cancer, the consumption of dissolved heavy metals (such as Ni, Cd, As, and Pb) in drinking water poses 2% of the world's cancer risks. Therefore, there is a need for an ongoing evaluation of innovative and state-of-the-art methods for comprehensive and sufficient water quality monitoring and prediction. In contrast to developing nations like Nigeria, data on seasonal variations and water quality is widely available in developed countries. Several factors, such as a lack of government funding and inaccessibility to laboratory facilities in Nigeria, can be connected to this.\u003c/p\u003e \u003cp\u003eConsequently, a systematic literature review (SLR) covering the previous 20 years (2003\u0026ndash;2024) on the use of machine learning models in water quality monitoring and assessment was conducted for this study. The trend in machine learning, a relatively recent field of study that has showed significant promise in the prediction of water quality, must be embraced given the critical role clean water plays in both sustainable development and the preservation of human health. Furthermore, a comprehensive evaluation of the advancement from current well-observed developments to more advanced and hypothetical machine learning model simulation and comparative analysis has also been presented. The first section considers common factors that affect Nigeria's water quality in relation to the sources of contaminants and provides information on potential new contaminants. The second part considers the implications for prospects as well as the past and anticipated future trends in Nigeria's methods for managing and monitoring water quality. A priority for the frequency of input variables and the precision of validation measures used to show the findings has also been placed on the deployment of different ML models in water quality monitoring. Future research potential and investigations on water quality have also been recommended.\u003c/p\u003e"},{"header":"2. Study area","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Location\u003c/h2\u003e \u003cp\u003eNigeria is bounded to the north by Niger, towards the east by Chad and Cameroon, towards the south by the Atlantic Ocean's Gulf of Guinea, and to the west by Benin (Encyclop\u0026aelig;dia Britannica \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) (Fig.\u0026nbsp;1)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Relief\u003c/h2\u003e \u003cp\u003eNigeria's terrain is characterized by lowlands in the north and south, which are punctuated by plateaus and mountains in the center of the nation. The Sokoto Plains are located in the country's northwest region, whereas the Borno Plains are located in the country's northeastern corner and extend to the Lake Chad basin. Porous, geologically young sedimentary strata underpin the Lake Chad basin and coastal regions, along with the Niger River delta and western sections of the Sokoto region in the extreme northwest (Encyclop\u0026aelig;dia Britannica, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ayejoto et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The plateaus' distinctive features are high plains with vast, shallow valleys interspersed with several hills or solitary summits known as inselbergs; the base rocks are crystalline, with sandstones appearing in river basins. Some degraded features, including the Udi-Nsukka escarpment, rise suddenly above the plains at least 1,000 feet in height (300 meters) (Okamkpa et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Climate\u003c/h2\u003e \u003cp\u003eThe climate of Nigeria is tropical, with varying rainy and dry seasons depending on location. The southeast is hot and humid most of the year, but dry in the southwestern and further inland. The west and north have a savanna climate with distinct dry and wet seasons, whereas the extreme north has a steppe environment with limited precipitation.\u003c/p\u003e \u003cp\u003eThe frequency of the rainy season generally reduces from south to north. The rainy season runs from March to November in the south, except for mid-May to September in the extreme north. The rains stop completely in the south during August, resulting in a brief dry season known as the \"August break.\" Precipitation is greater in the south, particularly the southeastern, which experiences over 120 inches (3,000 mm) of rain every year, compared to around 70 inches (1,800 mm) in the southwest. Temperature and humidity stay largely consistent in the south throughout the year, but seasons fluctuate significantly in the north; and during the northern dry season, the average temperature range also increases.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Systematic Literature Review\u003c/h2\u003e \u003cp\u003eThe research method used in the present study was the systematic literature review. To carry out the research, the following research questions were raised:\u003c/p\u003e \u003cp\u003e(1) what are the major sources of water contamination in Nigeria?\u003c/p\u003e \u003cp\u003e(2) what are the most frequently used machine learning models in water quality analysis in the last two decades? (3) what are the most accurate machine learning models based on their model predict metrics (such as coefficient of determination [\u003cem\u003eR\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e])\u003c/p\u003e \u003cp\u003e(4) what are the most frequently monitored water quality parameters\u003c/p\u003e \u003cp\u003e(5) what are the implications of the use of ML models in water quality monitoring and their implication for future research prospects?\u003c/p\u003e \u003cp\u003eThe systematic literature review was comprised of published articles from Web of Science and Scopus databases. The search terms used for the search in the databases included the \u003cem\u003esources of contamination in the aquatic environment in Nigeria and the application of machine learning in water quality prediction\u003c/em\u003e\u003c/p\u003e \u003cp\u003eIn the Scopus database, each term was entered in the search directory including the abstract, title, and keywords. The filters used were the year of publication (within the date range of 2003\u0026ndash;2024), the thematic area of environmental sciences, access type (open and none-open access articles), and the document type (article or review). In the Web of Science database, the search filter was based on Abstract, title, author keywords, and study within the date range of 2003\u0026ndash;2024. Two thematic categories- machine learning in water quality analysis and environmental science.\u003c/p\u003e \u003cp\u003eThe articles were selected based on the following criteria:\u003c/p\u003e \u003cp\u003e(1) articles that give a detailed explanation of the major sources of water pollution\u003c/p\u003e \u003cp\u003e(2) articles that explain the use of hybridized or standalone machine learning algorithms in water quality monitoring\u003c/p\u003e \u003cp\u003e(3) articles that explain the peculiarities of the different validation metrics (e.g. coefficient of determination, standard error of estimate, the sum of square errors and residual error).\u003c/p\u003e \u003cp\u003eThe exclusion was made on certain criteria such as:\u003c/p\u003e \u003cp\u003e(1) articles that appear outside the scope of the study\u003c/p\u003e \u003cp\u003e(2) articles that do not explicitly address the use of machine learning algorithms in water quality monitoring\u003c/p\u003e \u003cp\u003e(3) articles concerning other valued environmental components outside of water.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Creating a bibliographic map using the VOSviewer software\u003c/h2\u003e \u003cp\u003eThe VOSviewer was used in creating a bibliographic map for keywords in machine learning and water quality between the period of 2003\u0026ndash;2024. Bibliographic data from Scopus and Web of Science databases were extracted with a filter in the area of environmental sciences. A total of 729 and 2,234 articles were respectively returned for the search terms \u003cem\u003efor sources of contamination in the aquatic environment in Nigeria and the application of machine learning in water quality prediction\u003c/em\u003e in the Scopus database. For the Web of Science database, 521 and 1,672 articles respectively were returned. The articles were then exported from the database in CSV format to the VOSviewer. The bibliographic map was then created in the VOSviewer environment using keywords with a minimum of five occurrences selected (van Ech and Waltman \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Results and discussion","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Results from Databases\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the number of analyzed articles considered in the review, while Fig.\u0026nbsp;2 shows the PRISMA 2000 flow diagram for systematic literature review (SLR) showing the results of the database searches and their registries. The variation in the number of articles using the two search items (\u003cem\u003esources of contamination in aquatic environment in Nigeria and application of machine learning in water quality prediction\u003c/em\u003e) in the two databases showed an upward increase in articles on the term \u003cem\u003eapplication of machine learning in water quality prediction\u003c/em\u003e compared to the former. The Scopus database showed a higher number of publications compared to the Web of Science database. This is because the Scopus database has a greater number of indexed journals compared to the Web of Science database.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNumber of articles analyzed and considered for the SLR\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSearch terms\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eReturned articles\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDatabases\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSelected articles in the search\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSources of contamination in the aquatic environment in Nigeria\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e729\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eApplication of machine learning in water quality prediction\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2,234\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSources of contamination in the aquatic environment in Nigeria\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e521\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWeb of Science\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eApplication of machine learning in water quality prediction\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,672\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWeb of Science\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThe sum of articles considered from abstracts and titles\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e98\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThe sum of articles considered after full-text analysis\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eThe number of articles returned with the use of the search term \u003cem\u003esources of contamination in the aquatic environment in Nigeria and application of machine learning in water quality prediction\u003c/em\u003e were 729 and 2,234 respectively in the Scopus database. For the Web of Science database, 521 and 1,672 articles respectively were returned from the search (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig.\u0026nbsp;2).\u003c/p\u003e \u003cp\u003eThe number of articles considered for review was 35 and 20 for the terms \u003cem\u003esources of contamination in the aquatic environment in Nigeria and application of machine learning in water quality prediction\u003c/em\u003e in the Scopus database. While 25 and 18 were selected for the Web of Science database (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). It is worth mentioning that due to a large number of returned articles for the Scopus database, the titles and abstracts of articles read were for the first 600.\u003c/p\u003e \u003cp\u003eFor inclusion and exclusion criteria, only articles that did not appear in the two databases were included for the review; as some articles appeared in both the Scopus and Web of Science databases. Therefore, a total of 98 articles were selected for reading of titles and abstracts while a total of 40 articles were selected for full-text analysis while others were discarded (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The discarded articles did not meet the established criteria.\u003c/p\u003e \u003cp\u003eBased on the search term “\u003cem\u003eapplication of machine learning in water quality prediction”\u003c/em\u003e, out of the 257 publications that were considered, China occurred in 61, the United States in 57, India in 32, Malaysia in 15, south Korea in 14, Australia in 13, the United Kingdom and Canada in 10, Vietnam in 9, Iran in 8, Brazil and Iraq occurred in 7, Poland and Taiwan occurred in 6, Russian Federation, France, Germany, Italy, and Saudi Arabia occurred in 5; Morocco, Sweden, Bangladesh, Netherlands, and Hong Kong occurred in 4; Singapore, Spain, United Arabs Emirates, Norway, Palestine, Pakistan, and Greece occurred in 3; Croatia, Egypt, Pietro Florentine, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia occurred in 2; Dakota, Philippines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Israel, and Mexico occurred in 1 had one publication each (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). It can therefore be deduced frequency of the application of ML in these countries may be due availability of experts in the use of neural networks in engineering and environmental studies. Although this may be speculative, the author hopes that this observation will spur the interest of researchers in the field for more discussion and enlightenment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNumber of articles per country occurring from the systematic literature review\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCountry\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo. of Publication\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChina\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e61\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnited States\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMalaysia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSouth Korea\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAustralia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnited Kingdom\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCanada\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVietnam\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIran\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBrazil, Iraq\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePoland, Taiwan, Poland\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRussian Federation, France, Germany, Italy, Saudi Arabia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMorocco, Sweden, Bangladesh, Netherlands, Hong Kong\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSingapore, Spain, United Arab Emirates, Norway, Palestine, Pakistan, Greece\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCroatia, Egypt, Pietro Florentine, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDakota, Philippines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Israel, and Mexico.\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Bibliographic mapping using VOSviewer\u003c/h2\u003e \u003cp\u003eThe bibliographic map was created using the VOSviewer software. Co-occurrence was chosen as the type of analysis, while All keywords were chosen as the unit of analysis, with fractional counting selected as the preferred counting method. For the \u003cem\u003eapplication of machine learning in the water quality prediction\u003c/em\u003e search file, the software returned 377 keywords with 186 meeting the threshold and occurring at least six times across all analyzed articles. For each of the 186 keywords, the total strength of the co-occurrence links with other keywords was calculated. The keyword with the greatest link strength was then selected for generating the bibliographic map. Links between keywords are an indication of the relationship between them (the link strength) represented by a numerical value. Implying that the higher the link strength, the stronger the association where the two terms occur together (van Eck and Waltman 2020). Terms were then grouped into clusters of a different colour; with each colour representing the strength of the association.\u003c/p\u003e \u003cp\u003eFor the \u003cem\u003eapplication of machine learning in water quality prediction\u003c/em\u003e terms, analysis was made based on both the countries of study and the type of machine learning algorithm frequently used. The bibliographic map based on the overlay visualization (with the year of publication) and density visualization is shown in Fig.\u0026nbsp;3 based on the type of machine learning model. The graphical representation of the variation in the number of publications and the occurrence of the ML algorithms in water quality prediction is shown in Fig.\u0026nbsp;3. \u003cem\u003eArtificial neural networks\u003c/em\u003e had the highest occurrence and strongest link strength in all analyzed publications followed by \u003cem\u003eregression analysis and random forest\u003c/em\u003e, while \u003cem\u003eBayes theorem\u003c/em\u003e and the \u003cem\u003econvolutional neural network\u003c/em\u003e had the lowest occurrence and weakest link strength (Fig.\u0026nbsp;3–4). From the bibliographic map, high density and link strengths seem to occur mostly with \u003cem\u003eArtificial neural networks\u003c/em\u003e, \u003cem\u003esupport vector machines, regression analysis, decision trees, and random forests\u003c/em\u003e. However, \u003cem\u003efuzzy systems, long short-term memory, convolutional neural networks, and hybrid, and ensemble\u003c/em\u003e ML models had the weakest links and strength, with the number of publications on the subject seeming to be increasing since 2021. This result implies that the use of hybrid ML models in water quality prediction has not been well explored globally; a majority of the prediction has been based on the use of artificial neural networks.\u003c/p\u003e \u003cp\u003eBased on the countries of study, China, the United States, and India had the highest number of publications and link strength concerning the term \u003cem\u003eapplication of machine learning in water quality prediction\u003c/em\u003e (Fig.\u0026nbsp;5). The total number of publications involving the review developed by country was 61, 57, and 15 respectively for China, the United States, and India (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;5). This implies that countries like Croatia, Egypt, Pietro Fiorentini, Serbia, Belgium, Algeria, Nigeria, South Africa, Turkey, Columbia, Dakota, Philipines, Tunisia, Austria, Uganda, Romania, Chile, Denmark, Czech Republic, Luxembourg, Switzerland, Finland, Georgia, Peru, Sri Lanka, Portugal, State of Libya, Isareal and Mexico are grossly lacking behind in the use of ML in water quality monitoring and assessment.\u003c/p\u003e \u003cp\u003eFor the \u003cem\u003esources of contamination in the aquatic environment in Nigeria\u003c/em\u003e search term, analysis was made based on the relationship between anthropogenic and natural/geogenic sources of contamination and the number of publications within the last two decades (2003–2024). Based on the relationship between pollutant type, heavy metals, Lead, Cadmium, Copper, Chromium, and Zinc were reported in more publications and had the greatest link strength among others. Meanwhile, Antibiotic agents, aromatic compounds, bitumen, Cl, combustion, dumpsite, EC, mineralization, municipal solid wastes, organic carbon, phosphates, rainwater, saline intrusion, urbanization, and suspended particulate matter showed the lowest link strength and were reported in fewer publications (Fig.\u0026nbsp;6, Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Through the analysis of extracted literature, from 2003 to 2024, a relationship was established between anthropogenic and natural sources of contamination among all the reported articles on water quality contamination in Nigeria. From the results, it was observed that most studies attributed the occurrence of contamination in water to mining followed by lithogenic and anthropogenic processes (Fig.\u0026nbsp;7).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003enumber of publications based on sources of aquatic contamination in Nigeria\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePollutant\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOccurrences per\u003c/p\u003e \u003cp\u003epublication\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTotal link strength\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeavy metal\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e67\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeavy metals\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e59\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e52\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLead\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e48\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e47\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCadmium\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e39\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCopper\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChromium\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZinc\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e34\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e34\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEscherichia coli\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGeologic sediments\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePolycyclic aromatic hydrocarbons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eManganese, Nickel\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArsenic, Iron\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrate\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAHs, pH\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectrical conductivity\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHydrocarbons\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eColiform bacterium, sewage\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBacterium contamination, seasonal variation\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMicrobial contamination\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEffluents, leachates, mercury, mining, petroleum, trace elements, trace metals, turbidity, weathering\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgriculture, Calcium, total dissolved solids\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAluminum, crude oil, feces, irrigation, sulfate, wastewater\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnti-infective agents, chlorine, fecal coliform, industrial waste, land use, NO\u003csub\u003e3\u003c/sub\u003e, Na, population density, wastewater\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChlorine compounds, DO, geology, industries, Mg, oil spill, K\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCo, F, landfill, leachates, leaching, organic matter, runoff, sulfur compounds, hardness, domestic wastes\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAntibiotic agents, aromatic compounds, bitumen, Cl, combustion, dumpsite, EC, mineralization, municipal solid wastes, organic carbon, phosphates, rainwater, saline intrusion, urbanization, suspended particulate matter\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eThese results are in agreement with reports by Omeka \u0026amp; Egbueri (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), that due to an upsurge in anthropogenic activities in the country such as mining, industrialization, and socio-economic activities, high concentrations of heavy metal(loids) and potentially toxic elements have been reported in drinking water in the country. As reported by Obasi, \u0026amp; Akudinobi (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), there is an upward increase in the concentration of Cd, Pb, Cr, Zn, and cobalt in waters from the solid-mineral-rich southern Benue trough in southeastern Nigeria. This has been attributed to the open-cast mining method that is prevalent among most mining operations in the country (Okolo et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Omeka \u0026amp; Igwe \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In southeastern Nigeria, Ajala et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), provided a critical review of potential sources of heavy metals in water and aquatic fish as well as the human health risks from their consumption. High concentrations of Cd, As, Cu, and Zn were reported in both the drinking water and the shell of fishes. In Nsukka, southeastern Nigeria, they have been a reported high concentration of As, Cd, Hg, and Cr in water; attributed to anthropogenic influxes and run-off from domestic solid and sewage wastes (Nnaji et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMany studies in southwestern Nigeria have attributed the concentration of heavy metals in drinking water sources to industrialization. According to an extensive literature review carried out by Balogun et al. (2022) in the region, the long-standing sources of groundwater pollution cases have been attributed majorly to industrial effluents. In some parts of the Ibadan metropolis, southwest Nigeria, the concentration of heavy metals in shallow had-dug wells was observed to occur in the order of Zn \u0026gt; Fe \u0026gt; Pb \u0026gt; Cd \u0026gt; Mn; attributed to poor sanitation and industrialization (Ganiyu et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). In the same region, the water quality of shallow aquifers was investigated for their metal content and bacterial load. The water was reportedly polluted due to pathogens and heavy metals; with the heavy metals occurring in the order of Cd \u0026gt; Pb \u0026gt; Zn \u0026gt; Fe \u0026gt; Mn, attributed to varying anthropogenic activities in the area (Ganiyu et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e). In the densely populated parts of the southwestern region, such as Lagos, water pollution has been associated with poor waste management practices, lowering of the water table (due to over-abstraction), and influx from industrial effluents (Ogundiran \u0026amp; Afolabi \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Some studies have also attributed the presence of heavy metals to both geogenic processes (such as weathering and leaching of subsurface geology) and anthropogenic sources (from industrial effluents). A study on the water quality assessment of Asa River in Ilorin, southwestern Nigeria revealed a high concentration of chromite (FeCr\u003csub\u003e3\u003c/sub\u003eO\u003csub\u003e4\u003c/sub\u003e) and pyrite (FeS) in river sediments, attributed to weathering of subsurface geology. Heavy metals such as Zn, Fe, Cr, and Mn were also reportedly found in elevated concentration due to the influx of industrial effluents in sediments (Adekola \u0026amp; Elleta 2007). Although some studies in the region have reported the presence of heavy metals in water due to vehicular emissions and related repair products, domestic sewage, and effluents (Tijani et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), most have been mostly due to industrial and agricultural effluents (Olaojo et al. 2016; Emenike et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the northern and central parts of Nigeria, very few studies have reported water pollution due to industrial effluents Most reports on heavy metal concentration in water have been attributed to mining activities. In the Anka gold mining area of northwestern Nigeria, there have been reported cases of Pb poisoning found in the hairs and nails of children within the vicinity of the mine site; attributed to the ongoing mining activities within the region (Adebwumi 2020). Although a study by Lar et al. (2015) reported the concentration of heavy metals (such as Zn, V, Pb, Cu, Co, Be Cr, As, Cd, Sb, and Se) in drinking water wells due to volcanic eruption in the Panyam volcanic province, majority of the reports on metal concentration in the water have been attributed to mining and poor disposal of its effluents. Pb, Ni, and Hg poisoning have been reported in analyzed surface and groundwater samples from the Bagega gold mine province in Zamfara state, Northwestern Nigeria. The high elevated concentration of carcinogenic elements was attributed to artisanal gold mining (Nuhu et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn southeastern Nigeria, the sources of water pollution are highly ubiquitous-ranging from a wide range of anthropogenic sources (e.g., mining, poor waste disposal activities, poor hygiene, industries, population increase, land use, etc.) and geogenic sources (precipitation, chemical weathering, weathering, and rock-water interaction) (Omeka et al. 2023; Aghamelu et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Although most of the reports on contamination have been from both anthropogenic and geogenic activities (Nnorom et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Edet et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), the majority have been from mining and poor waste disposal activities. Reports from mining have been majorly from the solid mineral-rich zones of the lower Benue trough; with high concentrations of ore minerals occurring in association with Pb-Zn. (Omeka \u0026amp; Igwe \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Adamu et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This has been mainly reported from Ebonyi, Enugu, and Cross River states. A review carried out by Umeoguaju et al. (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) for two decades (2000–2020) revealed that the concentration of heavy metals in most water sources in southeastern Nigeria is attributed majorly to anthropogenic influences from mining and oil exploration. This is in agreement with a study by Opuene \u0026amp; Agbozu \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) on the heavy metal assessment in fish from Taylor Creek, southern Nigeria.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Application of neural networks in Water Quality Monitoring: implications for prospects\u003c/h2\u003e \u003cp\u003eThis section addresses the efficacy of artificial neural networks (ANNs) in water quality assessment, the most frequently used ANN models, the input or predictor variables, and the model accuracy based on the coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e). To achieve this, a literature search was conducted on the application of neural networks in water quality modeling and prediction. The search was conducted on papers from the Scopus and Web of Science databases between 2003–2024, using the keywords “water quality and artificial intelligence” with emphasis on both surface and groundwater. The results from the search were compiled and presented in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, and plots were generated based on the results in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDataset of selected articles from the Web of Science and Scopus databases used for SLR\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLocation\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eANN model\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eInput parameters\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHighest R\u003csup\u003e2\u003c/sup\u003e (during the testing stage)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. Surface and groundwater, Ojoto suburb, southeastern Nigeria\u003c/p\u003e \u003cp\u003e2. Groundwater samples from Osisioma, southeastern Nigeria\u003c/p\u003e \u003cp\u003e3. Groundwater samples from Egbema, southeastern Nigeria\u003c/p\u003e \u003cp\u003e4. surface and groundwater from Okurumutet-Iyamitet mine province, southeastern Nigeria\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTH, T, pH, TDS EC, Cl, Ca, SO\u003csub\u003e4\u003c/sub\u003e, Pb, HCO\u003csub\u003e3\u003c/sub\u003e, Zn Fe.\u003c/p\u003e \u003cp\u003eCu, Pb, Fe As, Cr, Benzene, Ethylbenzene m-Xylene, Toluene, and o-Xylene\u003c/p\u003e \u003cp\u003eFe, Zn, Ni, Cd, Cu and Pb\u003c/p\u003e \u003cp\u003eHCO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e, SO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2−\u003c/sup\u003e, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e, Cl\u003csup\u003e−\u003c/sup\u003e, Mg\u003csup\u003e2+\u003c/sup\u003e, Ca\u003csup\u003e2+\u003c/sup\u003e, K\u003csup\u003e+\u003c/sup\u003e, and Na\u003csup\u003e+\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.878\u003c/p\u003e \u003cp\u003e0.896\u003c/p\u003e \u003cp\u003e0.966\u003c/p\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Egbueri 2023)\u003c/p\u003e \u003cp\u003e(Akakuru et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Akakuru et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003eb)\u003c/p\u003e \u003cp\u003e(Omeka et al. 2023)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. Bouregreg watershed, Morocco.\u003c/p\u003e \u003cp\u003e2. Groundwater from Agartala municipality, India.\u003c/p\u003e \u003cp\u003e3. groundwater from El Kharga Oasis, Western\u003c/p\u003e \u003cp\u003eDesert of Egypt.\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003cp\u003eANFIS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003epH and EC\u003c/p\u003e \u003cp\u003eNa\u003csup\u003e+\u003c/sup\u003e, Mg\u003csup\u003e2+\u003c/sup\u003e, Ca\u003csup\u003e2+\u003c/sup\u003e, EC, HCO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e, and B\u003c/p\u003e \u003cp\u003eEC, pH, T\u003cem\u003e◦\u003c/em\u003e, TDS, Na\u003csup\u003e+\u003c/sup\u003e, Mg\u003csup\u003e2+\u003c/sup\u003e, K\u003csup\u003e+\u003c/sup\u003e, Ca\u003csup\u003e2+\u003c/sup\u003e, Cl\u003csup\u003e\u003cem\u003e−\u003c/em\u003e\u003c/sup\u003e, CO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e2\u003cem\u003e−\u003c/em\u003e\u003c/sup\u003e, SO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2\u003cem\u003e−\u003c/em\u003e\u003c/sup\u003e, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u003cem\u003e−\u003c/em\u003e\u003c/sup\u003eand HCO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u003cem\u003e−\u003c/em\u003e\u003c/sup\u003e,\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.87\u003c/p\u003e \u003cp\u003e0.990\u003c/p\u003e \u003cp\u003e0.997\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Bilali et al. 2022)\u003c/p\u003e \u003cp\u003e(Mallik et al \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Ibrahim 2022)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1) Groundwater wells from Illizi County, Algeria\u003c/p\u003e \u003cp\u003e(2) groundwater from El Merk is an oil field in the SOUTHEAST of Algerian\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBP-NN\u003c/p\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003epH, TH, EC, TDS, Ca, Mg, Na, K, HCO\u003csub\u003e3,\u003c/sub\u003e Cl, SO\u003csub\u003e4\u003c/sub\u003e, and NO\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e \u003cp\u003eTH, NO\u003csub\u003e3\u003c/sub\u003e, and NO\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.8957\u003c/p\u003e \u003cp\u003e0.9967\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Kouadri et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Kouadri \u0026amp; Samir \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. The Elbe River, Germany\u003c/p\u003e \u003cp\u003e2. Groundwater from the Gaza Strip Palestine\u003c/p\u003e \u003cp\u003e3. Yipin River, China\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWavelet and BPNN (W-BPNN)\u003c/p\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFlow, pH, Fe, and DO\u003c/p\u003e \u003cp\u003eAbstraction average rate (AVR), relative humidity (RH), depth from the surface to well screen (DSWS), aquifer thickness (AT), recharge rate (RR), Initial chloride concentration (ICC), and groundwater level (GWL)\u003c/p\u003e \u003cp\u003epH, TP, Temperature, EC, PI, NH\u003csub\u003e3\u003c/sub\u003e-N, COD and TN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.780\u003c/p\u003e \u003cp\u003e0.9770\u003c/p\u003e \u003cp\u003e0.717\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Li et al. 2020)\u003c/p\u003e \u003cp\u003e(Kassem 2020)\u003c/p\u003e \u003cp\u003e(Zhu and Heddam \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2019\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWater samples from the Shivganga River\u003c/p\u003e \u003cp\u003ebasin, India\u003c/p\u003e \u003cp\u003elakes, Tezpur University, India\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBP-NN\u003c/p\u003e \u003cp\u003eMLP-ANN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEC, pH, TH, TDS, Mg, Ca, K, Na, HCO\u003csub\u003e3\u003c/sub\u003e, PO\u003csub\u003e4\u003c/sub\u003e NO\u003csub\u003e3,\u003c/sub\u003e Cl, and SO\u003csub\u003e4\u003c/sub\u003e\u003c/p\u003e \u003cp\u003eBOD and TSS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.932\u003c/p\u003e \u003cp\u003e0.783\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Kadam et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Ahamad et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2018\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. Gorganrood Basin,\u003c/p\u003e \u003cp\u003eIran\u003c/p\u003e \u003cp\u003e2. Gorganrood Basin,\u003c/p\u003e \u003cp\u003eIran\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eANFIS\u003c/p\u003e \u003cp\u003eANFIS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSAR\u003c/p\u003e \u003cp\u003eEC\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Azad et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Azad et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2017\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. Hooghly River, India\u003c/p\u003e \u003cp\u003e2. Hooghly River, India\u003c/p\u003e \u003cp\u003e3. Saint John River,\u003c/p\u003e \u003cp\u003eCanada\u003c/p\u003e \u003cp\u003e4. Langat River and Klang River, Malaysia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNN-CS\u003c/p\u003e \u003cp\u003eNN-GA\u003c/p\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChlorides, turbidity, pH, TA, and residual chlorine TH\u003c/p\u003e \u003cp\u003eChlorides, pH, residual chlorine\u003c/p\u003e \u003cp\u003eTA TH, and turbidity\u003c/p\u003e \u003cp\u003eTSS\u003c/p\u003e \u003cp\u003eBOD, DO, COD, Ph, NH3-N, and TSS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.976\u003c/p\u003e \u003cp\u003e0.7267\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Chatterjee et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Chatterjee et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(El Din and Zhang \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Hameed et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2016\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.Aji-Chay River, Iran\u003c/p\u003e \u003cp\u003e2. Aji-Chay River, Iran\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eW-ANN\u003c/p\u003e \u003cp\u003eW-ANFIS\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSalinity (EC)\u003c/p\u003e \u003cp\u003eSalinity (EC)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.9960\u003c/p\u003e \u003cp\u003e0.9958\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eBarzegar et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e\u003c/p\u003e \u003cp\u003e(Barzegar et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2015\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1. Hilo Bay, Hawaii,\u003c/p\u003e \u003cp\u003eUSA\u003c/p\u003e \u003cp\u003e2. Hilo Bay, Hawaii,\u003c/p\u003e \u003cp\u003eUSA\u003c/p\u003e \u003cp\u003e3. Dahan River, Taiwan\u003c/p\u003e \u003cp\u003e4. Nadong River, South\u003c/p\u003e \u003cp\u003eKorea\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP\u003c/p\u003e \u003cp\u003eW-ANN\u003c/p\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003cp\u003eW-ANN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSalinity, DO, and Temperature\u003c/p\u003e \u003cp\u003eTemperature, DO, and Salinity\u003c/p\u003e \u003cp\u003eNH\u003csub\u003e3\u003c/sub\u003e–N\u003c/p\u003e \u003cp\u003eWater level\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003cp\u003e0.967\u003c/p\u003e \u003cp\u003e0.979\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Alizadeh and Kavianpour\u003c/p\u003e \u003cp\u003e2015)\u003c/p\u003e \u003cp\u003e(Alizadeh and Kavianpour\u003c/p\u003e \u003cp\u003e2015)\u003c/p\u003e \u003cp\u003e(Chang et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Seo et al. \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2014\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKaroon River, Iran\u003c/p\u003e \u003cp\u003eKaroon River, Iran\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDO\u003c/p\u003e \u003cp\u003eCOD\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Emamgholizadeh et al.\u003c/p\u003e \u003cp\u003e2014)\u003c/p\u003e \u003cp\u003e(Emamgholizadeh et al.\u003c/p\u003e \u003cp\u003e2014)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2013\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJohor River, Malaysia\u003c/p\u003e \u003cp\u003eJishan Lake, China\u003c/p\u003e \u003cp\u003eJishan Lake, China\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003cp\u003eW-ANN\u003c/p\u003e \u003cp\u003eENN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEC, TDS, and turbidity\u003c/p\u003e \u003cp\u003eDO, Temperature, and pH\u003c/p\u003e \u003cp\u003eDO, Temperature, and pH\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.799\u003c/p\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Najah et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Xu and Liu \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e(Xu and Liu \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2012\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKinta River, Malaysia\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 parameters\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.765\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Gazzaz et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2011\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNile River, Egypt\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMLP-NN\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 parameters\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(Khalil et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2011\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eAs observed earlier from the bibliographic coupling map in Fig.\u0026nbsp;2, artificial neural networks appear to be the most frequently used machine learning model in water quality monitoring and assessment globally, followed only by support vector machines and decision trees. The wide usage of ANN in water quality studies has been attributed to its efficiency and versatility in predictions even in systems with poor computational strength (Egbueri et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Omeka et al. 2023). According to Ghavidel and Montazeri (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), the unique ability of the ANN model to accurately match a broad range of nonlinear variables makes it a widely accepted ML model in most water quality studies. The ANN architecture is designed to mimic the human neural system; with the unique ability to quickly learn and send signals about a range of linear and nonlinear datasets through its interwoven parts known as “neurons” (Ozel et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This makes it stand out among other water quality modeling techniques. Water quality data are usually nonstationary, random, nonlinear, and unpredictable (Beven, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). This means that the relationship between water quality and its controlling factors will tend to vary over time (as a result of varying anthropogenic and seasonal fluctuations), hence, the application of predictive models that work on historical data will become ineffective over time (Khan et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), hence more focus has been put on ANN due to its unique ability to process complex datasets.\u003c/p\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e4.3.1 Model utilization\u003c/h2\u003e \u003cp\u003eIt can be observed from Fig.\u0026nbsp;8 that the multilayer perceptron artificial neural network (MLP-NN) and back-propagated neural network (BP-NN) are the most frequently used neural network algorithms in water quality monitoring and prediction within the period of 2003–2024. This was closely followed by the adaptive neuro-fuzzy inference system (ANFIS), and other neural networks such as the Wavelet and BPNN (WNN) hybrid model, Neural Network trained by Cuckoo Search (NNCS), Neural Network trained by Genetic Algorithm (NN-GA), Elman Neural Network (ENN) and Wavelet-Adaptive Neural Fuzzy Interference System (WANFIS) in decreasing order. The high frequency of the MLP-NN and BP-NN in water quality monitoring and prediction is in line with a study by Rajaee et al. (\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The wide acceptance of these algorithms has also been attributed to their ability to attain high modeling accuracy with fewer input parameters compared to other white-box models (Yetilmezsoy et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Ighalo et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Omeka 2023).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e4.3.2 Model validation accuracy\u003c/h2\u003e \u003cp\u003eIn this assessment, only studies that used the coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e) as a validation metric were considered. The exclusion was made for the root mean square error (RMSE). The RMSE was excluded because of the difficulty of directly comparing the units to different parameters. This implies that its accuracy is dependent on the size (value) of the parameter being considered. This would mean that a parameter with a large numerical value will show a high RMSE value; resulting in inaccuracy (Ighalo et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Unlike the RMSE, the R\u003csup\u003e2\u003c/sup\u003e is a measure of the extent of variation of a particular dataset, therefore giving better accuracy of parameters in a particular dataset (Adeniyi et al. 2019).\u003c/p\u003e \u003cp\u003eAn observation of Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows that the adaptive neuro-fuzzy inference system (ANFIS), Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS), and the Wavelet and BPNN (W-NN) hybrid models are the most accurate neural networks for both surface and groundwater monitoring and prediction. These findings are in agreement with the observation of Ighalo et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) on the frequently used neural networks for surface water prediction between 2011–2020. This study has revealed that although the MLP-NN and BP-NN seem to be the most popularly used algorithms in water quality prediction, however, they have low modeling accuracy compared to other hybrid models. The high accuracy of the W-ANFIS and W-NN hybrid models has been attributed to the in-cooperation of the wavelet decomposition of covariates (predictor variables) into detail and approximate components (Barzegar et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). On the other hand, the high accuracy of the ANFIS is because its architecture integrates the learning ability of both the Fuzzy Inference System and ANN (Emamgholizadeh et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e4.3.3 Predicted water quality parameters\u003c/h2\u003e \u003cp\u003eIn this section, the most frequently analyzed water quality parameters for both surface and groundwater have been discussed. An observation of Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows that temperature, dissolved oxygen (DO), pH, Electrical conductivity (EC; as a function of salinity), and total dissolved solids (TDS) are the most frequently investigated parameters in water quality prediction. A possible reason for their high prevalence is due to significance in the determination of the overall water quality in both surface and groundwater sources. Ighalo et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) have attributed the prevalence of DO, TDS, and pH to the high availability and low cost of measuring equipment. Moreover, these parameters are readily measured in situ before laboratory analysis because of their high susceptibility to surface environmental changes (Igwe and Omeka \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eWith the aid of meta-analysis from Scopus and Web of Science databases and bibliometric coupling for data extraction and visualization, the present study has explored the recent trends to more hypothetical and advanced model simulation in the application of machine learning (ML) models in water quality monitoring and prediction in Nigeria for the last two decades (2003–2024). Emphasis has been made on the type of ML frequently used, the accuracy of the model in water quality prediction, and the frequently monitored water quality parameters. It was revealed that the use of hybridized ML models in water quality prediction has not been well explored globally; a majority of the prediction has been based on the use of artificial neural networks (ANN). The adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models were observed to be the most accurate ANN algorithms, in prediction. Based on analysis of the literature, temperature, dissolved oxygen (DO), pH, conductivity (EC) and total dissolved solids (TDS) appear to be the most frequently predicted parameters. China and the United States had the highest number of publications concerning the application of ML in water quality prediction; while Nigeria appears among the countries grossly lacking behind.\u003c/p\u003e\u003cp\u003eBased on a literature analysis of the sources of contaminants in aquatic systems in Nigeria, heavy metals (Pb, Cd, Cu, Cr, and Zinc) were reported in more publications; linked majorly to mining and industrial effluents. This result implies that the increasing rate of anthropogenic activities coupled with climatic variation will continue to render water quality data complex; making the use of conventional assessment methods and standalone ML models inefficient. Therefore, future water quality monitoring and prediction studies in developing countries, especially in Nigeria- where data on environmental monitoring is lacking- should take advantage of the rapidly evolving field of machine learning; with more emphasis placed on the hybridized machine learning algorithms.\u003c/p\u003e\u003cp\u003eAlthough this study only considered the Scopus and Web of Science databases in its analysis, it is recommended that future studies should explore other databases such as Science Direct, PubMed, Cochrane Library, Lens, and Dimensions. Other bibliometric analytic techniques such as the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) should be explored as well.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eSAR \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Sodium Absorption Ratio\u003c/p\u003e\n\u003cp\u003eRMSE \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Root Mean Square Error\u003c/p\u003e\n\u003cp\u003eMLP-ANN \u0026nbsp; \u0026nbsp;Multi-layer perceptron neural network\u003c/p\u003e\n\u003cp\u003eTP \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Total Phosphorus\u003c/p\u003e\n\u003cp\u003eNN-CS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Neural Network trained by Cuckoo Search Algorithm\u003c/p\u003e\n\u003cp\u003eTS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Total Solids\u003c/p\u003e\n\u003cp\u003eTDS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Total Dissolved Solids\u003c/p\u003e\n\u003cp\u003eANFIS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Adaptive Neuro-Fuzzy Inference System\u003c/p\u003e\n\u003cp\u003eBOD \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Biochemical Oxygen Demand\u003c/p\u003e\n\u003cp\u003eTN \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Total Nitrogen\u003c/p\u003e\n\u003cp\u003eTA \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Total alkalinity\u003c/p\u003e\n\u003cp\u003eTH \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Total Hardness\u003c/p\u003e\n\u003cp\u003eCOD \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Chemical Oxygen Demand\u003c/p\u003e\n\u003cp\u003eDO \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Dissolved Oxygen\u003c/p\u003e\n\u003cp\u003eTSS \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Total Suspended Solids\u003c/p\u003e\n\u003cp\u003eBPNN \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Back Propagation Neural Network\u003c/p\u003e\n\u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Coefficient of Determination\u003c/p\u003e\n\u003cp\u003ePI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Permanganate Index\u003c/p\u003e\n\u003cp\u003eSVM \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Support Vector Machine\u003c/p\u003e\n\u003cp\u003eW-BPNN \u0026nbsp; \u0026nbsp; Wavelet and back propagation neural network\u003c/p\u003e\n\u003cp\u003eW-ANFIS \u0026nbsp; \u0026nbsp;Wavelet and Adaptive Neuro-Fuzzy Inference System\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contribution\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eMichael E. Omeka Michael E. Omeka conceived and designed the project, conducted bibliometric modeling, and wrote the manuscript. Manuscript review and editing were also carried out by the author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eNot applicable.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eNot applicable.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eNot applicable.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eThe author declares no competing interests regarding this research\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eNo funding was received for this research\u003c/em\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAdamu CI, Nganje TN, Edet A (2015) Heavy metal contamination and health risk assessment associated with abandoned barite mines in Cross River State, southeastern Nigeria, vol 3. Environmental Nanotechnology, monitoring \u0026amp; management, pp 10\u0026ndash;21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdekola FA, Eletta OAA (2007) A study of heavy metal pollution of Asa River, Ilorin. Nigeria; trace metal monitoring and geochemistry. Environ Monit Assess 125:157\u0026ndash;163\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdeniyi AG, Ighalo JO, Marques G (2020) Utilisation of machine learning algorithms for the prediction of syngas composition from biomass bio-oil steam reforming. Int J Sustain Energy. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/14786451.2020.1803862\u003c/span\u003e\u003cspan address=\"10.1080/14786451.2020.1803862\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdewumi AJ (2020) Contamination, sources and risk assessments of metals in media from Anka artisanal gold mining area. Northwest Nigeria Sci Total Environ 718:137235\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgbasi JC, Egbueri JC, Ayejoto DA, Unigwe CO, Omeka ME, Nwazelibe VE, Ighalo JO, Pande CB, Fakoya AA (2023) The impact of seasonal changes on the trends of physicochemical, heavy metal and microbial loads in water resources of southeastern Nigeria: A critical review. In: Egbueri JC, Ighalo JO, Pande CB (eds) Climate change impacts on Nigeria. Springer, Cham. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-21007-5_25\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-21007-5_25\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAghamelu OP, Omeka ME, Unigwe CO (2022) Modeling the vulnerability of groundwater to pollution in a fractured shale aquifer in SE Nigeria using information entropy theory, geospatial, and statistical modeling approaches. Model Earth Syst Environ, 1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhamad KU, Raj P, Barbhuiya NH, Deep A (2019) Surface water\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAjala OA, Oke MR, Ajibade TF, Ajibade FO, Adelodun B, Ighalo JO, Silva LF (2022) Concentrations, bioaccumulation, and health risk assessments of heavy metals in fishes from Nigeria\u0026rsquo;s freshwater: a general overview. Environ Sci Pollut Res 29(55):82660\u0026ndash;82680\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkakuru OC, Adakwa CB, Ikoro DO, Eyankware MO, Opara AI, Njoku AO, Usman A (2023) Application of artificial neural network and multi-linear regression techniques in groundwater quality and health risk assessment around Egbema, Southeastern Nigeria. Environ Earth Sci 82(3):77\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkakuru OC, Njoku UB, Obinna-Akakuru AU, Akudinobi BE, Obasi PN, Aigbadon GO, Onyeanwuna UB (2023) Non-carcinogenic health risk assessment and predicting of pollution indexing of groundwater around Osisioma, Nigeria, using artificial neural networks and multi-linear modeling principles. Stochastic Environmental Research and Risk Assessment, pp 1\u0026ndash;31\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl Aani S, Bonny T, Hasan SW, Hilal N (2019) Can machine language and artificial intelligence revolutionize process automation for water treatment and desalination? Desalination 458:84\u0026ndash;96\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlizadeh MJ, Kavianpour MR (2015) Development of wavelet-ANN models to predict water quality parameters in Hilo Bay Pacific Ocean. Mar Pollut Bull 98:171\u0026ndash;178\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAyejoto DA, Egbueri JC, Agbasi JC, Omeka ME, Unigwe CO, Nwazelibe VE, Ighalo JO, Pande CB (2023) Influence of Seasonal Changes on the Quality of Water Resources in Southwestern Nigeria: A Review. In: Egbueri JC, Ighalo JO, Pande CB (eds) Climate change impacts on Nigeria. Springer, Cham. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-21007-5_22\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-21007-5_22\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAzad A, Karami H, Farzin S, Saeedian A, Kashi H, Sayyahi F (2018) Prediction of water quality parameters using ANFIS optimized by intelligence algorithms (case study: Gorganrood River). KSCE J Civ Eng 22:2206\u0026ndash;2213\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarzegar R, Adamowski J, Moghaddam AA (2016) Application of wavelet-artificial intelligence hybrid models for water quality prediction: a case study in Aji-Chay River, Iran. Stoch Environ Res Risk Assess 30:1797\u0026ndash;1819\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeven K (2016) Advice to a young hydrologist. Hydrol Process 30(20):3578\u0026ndash;3582. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/hyp.10879\u003c/span\u003e\u003cspan address=\"10.1002/hyp.10879\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang F-J, Tsai Y-H, Chen P-A, Coynel A, Vachaud G (2015) Modeling water quality in an urban river using hydrological factors\u0026ndash; Data-driven approaches. J Environ Manage 151:87\u0026ndash;96\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChatterjee S, Sarkar S, Dey N, Ashour AS, Sen S, Hassanien AE (2017) Application of cuckoo search in water quality prediction using artificial neural network. Int J Comput Intell Stud 6:229\u0026ndash;244\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeng T, Duan HF, Keramat A (2022) Spatiotemporal characterization and forecasting of coastal water quality in the semi-enclosed Tolo Harbour based on machine learning and EKC analysis. Eng Appl Comput Fluid Mech 16(1):694\u0026ndash;712. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19942060.2022.2035257\u003c/span\u003e\u003cspan address=\"10.1080/19942060.2022.2035257\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdet AE, Merkel BJ, Offiong OE (2003) Trace element hydrochemical assessment of the Calabar Coastal Plain Aquifer, southeastern Nigeria using statistical methods. Environ Geol 44:137\u0026ndash;149\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgbueri JC, Agbasi JC (2022) Data-driven soft computing modeling of groundwater quality parameters in southeast Nigeria: comparing the performances of different algorithms. Environ Sci Pollut Res 29(25):38346\u0026ndash;38373\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgbueri JC, Igwe O, Omeka ME, Agbasi JC (2023) Development of MLR and variedly optimized ANN models for forecasting the detachability and liquefaction potential index of erodible soils. Geosyst Geoenvironment 2(1):100104\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgbueri JC (2022) Incorporation of information entropy theory, artificial neural network, and soft computing models in the development of integrated industrial water quality index. Environ Monit Assess 194(10):693\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl Bilali A, Taleb A (2020) Prediction of irrigation water quality parameters using machine learning models in a semi-arid environment. J Saudi Soc Agricultural Sci. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jssas.2020.08.001\u003c/span\u003e\u003cspan address=\"10.1016/j.jssas.2020.08.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl Din ES, Zhang Y (2017) Improving the accuracy of extracting surface water quality levels (Swqls) using remote sensing and artificial neural network: a case study in the saint john river, Canada. XLII\u0026ndash;4/W4:245\u0026ndash;249International archives of the photogrammetry. Remote sensing \u0026amp; spatial information sciences\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmamgholizadeh S, Kashi H, Marofpoor I, Zalaghi E (2014) Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models. Int J Environ Sci Technol 11:645\u0026ndash;656\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmenike PC, Tenebe IT, Neris JB, Omole DO, Afolayan O, Okeke CU, Emenike IK (2020) An integrated assessment of land-use change impact, seasonal variation of pollution indices, and human health risk of selected toxic elements in sediments of River Atuwara. Nigeria Environ Pollution 265:114795\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEncyclop\u0026aelig;dia, Britannica (2023) Inc\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaniyu SA, Mabunmi AA, Olurin OT, Adeyemi AA, Jegede OA, Okeh A (2021b) Assessment of microbial and heavy metal contamination in shallow hand-dug wells bordering Ona River, Southwest Nigeria. Environ Monit Assess 193:1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaniyu SA, Oyadeyi AT, Adeyemi AA (2021a) Assessment of heavy metals contamination and associated risks in shallow groundwater sources from three different residential areas within the Ibadan metropolis, southwest Nigeria. Appl Water Sci 11(5):81\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGazzaz NM, Yusof MK, Aris AZ, Juahir H, Ramli MF (2012) Artificial neural network modeling of the water quality index for Kinta River (Malaysia) using water quality variables as predictors. Mar Pollut Bull 64:2409\u0026ndash;2420\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeetha S, Gouthami S (2016) Internet of things enabled real-time water quality monitoring system Smart. Water 2:1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40713-017-0005-y\u003c/span\u003e\u003cspan address=\"10.1186/s40713-017-0005-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhavidel SZZ, Montazeri M (2014) Application of different data-driven methods for the prediction of total dissolved solids in the Zarinehroud basin. Stoch Environ Res Risk Ass 28(8):2101\u0026ndash;2118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00477-014-0899-y\u003c/span\u003e\u003cspan address=\"10.1007/s00477-014-0899-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHameed M, Sharqi SS, Yaseen ZM, Afan HA, Hussain A, Elshafe A (2017) Application of artificial intelligence (AI) techniques in water quality index prediction: a case study in the tropical region, Malaysia. Neural Comput Appl 28:893\u0026ndash;905\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHasan R, Raghav A, Mahmood S, Hasan MA (2011) Artificial Intelligence in River Quality Assessment. In: International Conference on Information Management, Innovation Management and Industrial Engineering, 2011. IEEE, pp 491\u0026ndash;495\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIbrahim H, Yaseen ZM, Scholz M, Ali M, Gad M, Elsayed S, Khadr M, Hussein H, Ibrahim HH, Eid MH et al (2023) Evaluation and Prediction of Groundwater Quality for Irrigation Using an Integrated Water Quality Indices, Machine Learning Models and GIS Approaches: A Representative Case Study. Water 15:694. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/w15040694\u003c/span\u003e\u003cspan address=\"10.3390/w15040694\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIfediegwu SI (2022) Assessment of groundwater potential zones using GIS and AHP techniques: a case study of the Lafa district, Nasarawa State, Nigeria. Appl Water Sci 12:10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13201-021-01556-5\u003c/span\u003e\u003cspan address=\"10.1007/s13201-021-01556-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIghalo JO, Adeniyi AG, Marques G (2020) Artificial intelligence for surface water quality monitoring and assessment A systematic literature analysis. Model Earth Syst Environ. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40808-020-01041-z\u003c/span\u003e\u003cspan address=\"10.1007/s40808-020-01041-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIgwe O, Omeka ME (2021) Hydrogeochemical and pollution assessment of water resources within a mining area, SE Nigeria, using an integrated approach. Int J Energy Water Resour 1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKadam AK, Wagh VM, Muley AA, Umrikar BN, Sankhua RN (2019) Prediction of water quality index using artificial neural network and multiple linear regression modeling approaches in Shivganga River basin. Model Earth Syst Environ India. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40808-019-00581-3\u003c/span\u003e\u003cspan address=\"10.1007/s40808-019-00581-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhalil B, Ouarda T, St-Hilaire A (2011) Estimation of water quality characteristics at ungauged sites using artificial neural networks and canonical correlation analysis. J Hydrol 405:277\u0026ndash;287\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan N, Shahid S, Ismail TB, Behlil F (2021) Prediction of heat waves over Pakistan using support vector machine algorithm in the context of climate change. Stoch Environ Res Risk Assess 1\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00477-020-01963-1\u003c/span\u003e\u003cspan address=\"10.1007/s00477-020-01963-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKouadri S, Elbeltagi A, Islam ARM, Kateb S (2021) Performance of machine learning methods in predicting water quality index based on the irregular data set: application on Illizi region (Algerian southeast). Appl Water Sci 11(12):1\u0026ndash;20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKouadri S, Samir K (2021) Hydro-chemical study with geospatial analysis of groundwater Quality Illizi Region, South-East of Algeria. Iran J Chem Chem Eng (IJCCE) 40(4):1315\u0026ndash;1333. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.30492/ijcce.2020.39800\u003c/span\u003e\u003cspan address=\"10.30492/ijcce.2020.39800\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKouadri S, Samir K (2021) Hydro-chemical study with geospatial analysis of groundwater Quality Illizi Region, South-East of Algeria. Iran J Chem Chem Eng (IJCCE) 40(4):1315\u0026ndash;1333. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.30492/ijcce.2020.39800\u003c/span\u003e\u003cspan address=\"10.30492/ijcce.2020.39800\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLar UA, Gusikit RB (2015) Environmental and health impact of potentially harmful elements distribution in the Panyam (Sura) volcanic province, Jos Plateau, Central Nigeria. Environ Earth Sci 74:1699\u0026ndash;1710\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMallik S, Chakraborty A, Mishra U, Paul N (2022) Prediction of irrigation water suitability using geospatial computing approach: a case study of Agartala city, India. Environmental Science and Pollution Research, pp 1\u0026ndash;16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins P, Forstinus NO, Ikechukwu NE (2016) Water and Waterborne Diseases: A Review Water and Waterborne Diseases: A Review. on. JANUARY\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohammadpour R, Shaharuddin S, Zakaria N, Ghani A, Vakili M, Chan N (2016) Prediction of water quality index in free surface constructed wetlands. Environ Earth Sci 75:1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12665-015-4905-6\u003c/span\u003e\u003cspan address=\"10.1007/s12665-015-4905-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorounke SG, Ayorinde JB, Benedict AO, Adedayo FF, Adewale FO, Oluwadamilare I, Sokunle SS, Benjamin A (2017) Epidemiology and Incidence of Common Cancers in Nigeria. J Cancer Biol Res 5(3):1105\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNajah A, El-Shafe A, Karim OA, El-Shafe AH (2013) Application of artificial neural networks for water quality prediction. Neural Comput Appl 22:187\u0026ndash;201\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNayak S (2020) Assessment of water quality of Brahmani river using correlation and regression analysis\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNigerian National System of Cancer Registries (2013) Federal Ministry of Health of Nigeria, Federal Secretariat Complex Shehu Shagari Way, Garki, Abuja P.M.B. 083, GarkiAbuja ISBN: 978-978-955-030-2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNnaji CC, Chibueze CV, Mama CN, Alum OL, Afangideh CB (2023) Impact of anthropogenic and environmental conditions on surface run-off quality: a case study of Nsukka, Eastern Nigeria. International Journal of Environmental Science and Technology, pp 1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNnorom IC, Ewuzie U, Eze SO (2019) Multivariate statistical approach and water quality assessment of natural springs and other drinking water sources in Southeastern Nigeria. Heliyon, 5(1), e01123\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNuhu AA, Sallau MS, Majiya MH (2014) Heavy metal pollution: the environmental impact of artisanal gold mining on Bagega village of Zamfara state, Nigeria. Res J Pharm Biol Chem Sci 5(6):306\u0026ndash;313\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eObasi PN, Akudinobi BB (2020) Potential health risk and levels of heavy metals in water resources of lead\u0026ndash;zinc mining communities of Abakaliki, southeast Nigeria. Appl Water Sci 10:184. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13201-020-01233-z\u003c/span\u003e\u003cspan address=\"10.1007/s13201-020-01233-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOgundiran OO, Afolabi TA (2008) Assessment of the physicochemical parameters and heavy metals toxicity of leachates from the municipal solid waste open dumpsite. Int J Environ Sci Technol 5:243\u0026ndash;250\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOkamkpa JR, Omeka ME, Igwe O, Iyiokwu MU (2022) An integrated geochemical and spatiotemporal assessment of groundwater resources within an industrial suburb, Southeastern Nigeria. Int J Energy Water Resour 1\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s42108-022-00183-3\u003c/span\u003e\u003cspan address=\"10.1007/s42108-022-00183-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOkolo CC, Oyedotun TDT, Akamigbo FOR (2018) Open cast mining: a threat to water quality in the rural community of Enyigba in south-eastern Nigeria. Appl water Sci 8:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlaojo AA, Oladunjoye MA, Sanuade OA (2018) Geoelectrical assessment of polluted zone by sewage effluent in University of Ibadan campus southwestern Nigeria. Environ Monit Assess 190:1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOmeka ME, Egbueri JC (2022) Hydrogeochemical assessment and health-related risks due to toxic element ingestion and dermal contact within the Nnewi-Awka urban areas, Nigeria. Environ Geochem Health 1\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10653-022-01332-7\u003c/span\u003e\u003cspan address=\"10.1007/s10653-022-01332-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOmeka ME, Ezugwu AL, Agbasi JC, Egbueri JC, Abugu HO, Aralu CC, Ucheana IA (2024) A review of the status, challenges, trends, and prospects of groundwater quality assessment in Nigeria: an evidence-based meta-analysis approach. Environmental Science and Pollution Research, pp 1\u0026ndash;24\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOmeka ME, Egbueri JC, Unigwe CO (2022a) Investigating the hydrogeochemistry, corrosivity, and scaling tendencies of groundwater in an agrarian area (Nigeria) using graphical, indexical, and statistical modeling. Arab J Geosci 15(13):1\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12517-022-10514-7\u003c/span\u003e\u003cspan address=\"10.1007/s12517-022-10514-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOmeka ME (2022) Evaluation and prediction of irrigation water quality of an agricultural district, SE Nigeria: an integrated heuristic GIS-based and machine learning approach. Environ Sci Pollut Res. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11356-022-25119-6\u003c/span\u003e\u003cspan address=\"10.1007/s11356-022-25119-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOpuene K, Agbozu IE (2008) Relationships between heavy metals in shrimp (Macro brachium felicinum) and metal levels in the water column and sediments of Taylor Creek\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOzel HU, Gemici BT, Gemici E, Ozel HB, Cetin M, Sevik H (2020) Application of artificial neural networks to predict the heavy metal contamination in the Bartin River. Environ Sci Pollut Res. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11356-020-10156-w\u003c/span\u003e\u003cspan address=\"10.1007/s11356-020-10156-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePour SH, Wahab AKA, Shahid S (2020) Physicalempirical models for prediction of seasonal rainfall extremes of Peninsular Malaysia. Atmos Res 233:104720. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.atmosres.2019\u003c/span\u003e\u003cspan address=\"10.1016/j.atmosres.2019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalamdhad A, Singh J, Dhamodharan K (eds) Advances in waste management. Springer, Singapore. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-981-13-0215-2_15\u003c/span\u003e\u003cspan address=\"10.1007/978-981-13-0215-2_15\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRajaee T, Khani S, Ravansalar M (2020) Artificial intelligence-based single and hybrid models for prediction of water quality in rivers: a review. ChemometrIntell Lab Syst 200:103978\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRicolfi L, Barbieri M, Muteto PV, Nigro A, Sappa G, Vitale S (2020) Potential toxic elements in groundwater and their health risk assessment in drinking water of Limpopo National Park, Gaza Province, Southern Mozambique. Environmental Geochemistry and Health, pp 1\u0026ndash;13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeo Y, Kim S, Kisi O, Singh VP (2015) Daily water level forecasting using wavelet decomposition and artificial intelligence techniques. J Hydrol 520:224\u0026ndash;243\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun K, Rajabtabar M, Samadi S, Rezaie-Balf M, Ghaemi A, Band S, Mosavi A (2021) An integrated machine learning, noise suppression, and population-based algorithm to improve total dissolved solids prediction. Eng Appl Comput Fluid Mech 15(1):251\u0026ndash;271. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19942060.2020.1861987\u003c/span\u003e\u003cspan address=\"10.1080/19942060.2020.1861987\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTijani MN, Jinno K, Hiroshiro Y (2004) Environmental impact of heavy metals distribution in water and sediments of Ogunpa River, Ibadan area, southwestern Nigeria. J Min Geol 40(1):73\u0026ndash;83\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUmeoguaju FU, Akaninwor JO, Essien EB, Amadi BA (2022) Heavy metal profile of surface and groundwater samples from the Niger Delta region of Nigeria: a systematic review and meta-analysis. Environ Monit Assess 194:1\u0026ndash;37\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Ech NJ, Waltman L (2020) VOSviewer Manual. Universiteit Leiden, The Centre for Science and Technology Studies, Meaningful metrics. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.vosviewer.com/\u003c/span\u003e\u003cspan address=\"https://www.vosviewer.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed February 2023\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organisation/United Nations Children Fund (WHO/UNICEF) 2010. Progress on drinking water and sanitation\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao Y, Liu K, Hao Q, Li J, Zhang Y, Cui W, Pei Q (2021) Hydrogeochemical features and genesis of confined groundwater and health perspectives for sustainable development in urban Hengshui. North China Plain, J Chem\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu L, Liu S (2013) Study of short-term water quality prediction model based on wavelet neural network. Math Comput Model 58:807\u0026ndash;813\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYetilmezsoy K, Ozkaya B, Cakmakci M (2011) Artificial intelligence-based prediction models for environmental engineering. Neural Netw World 21:193\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu S, Heddam S (2020) Prediction of dissolved oxygen in urban rivers at the Three Gorges Reservoir, China: extreme learning machines (ELM) versus artificial neural network (ANN). Water Qual Res J 55:106\u0026ndash;118\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, Pollution, Artificial intelligence, Drinking water quality, Nigeria","lastPublishedDoi":"10.21203/rs.3.rs-4429965/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4429965/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn recent decades, machine learning (ML) artificial intelligence has found wide application in water quality monitoring and prediction due to the increasing complexity of water quality data. This complexity has been attributed to the global upsurge in anthropogenic activities and climatic variations. It is therefore critical to identify the most accurate and suitable ML model for water quality prediction. In this study, a systematic literature review (SLR) was carried out to explore the trend and progress in the application of ML models in water quality monitoring and prediction in Nigeria from 2003\u0026ndash;2024. A comprehensive review of the effectiveness of advanced ML models as well as the gaps in their application in the area of water quality assessment and monitoring was also carried out using the PRISMA-P meta-analysis technique. Forty publications were used to perform bibliographic analysis and visualization using the VOS viewer software. The study found that globally, the use of hybrid ML models in water quality prediction has not been well explored; a majority of the prediction has been based on the use of artificial neural networks (ANN). Among the ANN algorithms, the adaptive neuro-fuzzy inference system (ANFIS), and Wavelet-Adaptive Neural Fuzzy Interference System (W-ANFIS) hybrid models are the most accurate in prediction; with temperature, dissolved oxygen (DO), pH, conductivity (EC), and total dissolved solids (TDS) among the most frequently predicted parameters. Nigeria is grossly lagging in the application of ML in water quality prediction. This limitation is largely attributed to inadequate data on environmental monitoring. It is critical therefore for future water quality monitoring and prediction studies in Nigeria to take advantage of the rapidly evolving field of machine learning; with more emphasis placed on the hybridized machine learning algorithms\u003c/p\u003e","manuscriptTitle":"Exploring the recent trends, progresses, and challenges in the application of Artificial intelligence in water quality assessment and monitoring in Nigeria: A systematic review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-10 11:18:51","doi":"10.21203/rs.3.rs-4429965/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"cff7fc62-c737-4acc-93d7-d597da4ce9f6","owner":[],"postedDate":"June 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-06-28T09:48:11+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-10 11:18:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4429965","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4429965","identity":"rs-4429965","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}