Abstract
Due to the increasingly serious water environment pollution, the difficulty of Water Quality Monitoring (abbreviated as WQM for convenience) is also constantly increasing, which puts forward more requirements for the capabilities of various aspects of WQM systems. However, the current WQM method has drawbacks such as slow speed, long monitoring time, complex operation, poor stability, and the inability to obtain accurate information on water pollution in the first time, as well as the generation of toxic and harmful secondary pollutants after some measurement parameters are tested. To address these issues and ensure water quality safety, this paper investigated the algorithm for monitoring water quality parameters using artificial intelligence data mining optical systems. This article applied an artificial intelligence data mining system to detect water quality and designed various system through this method to improve system performance. To verify the actual effectiveness of artificial intelligence data mining systems, this article selected 10 water plants as experimental research subjects and compared the differences between traditional WQM methods and WQM methods based on artificial intelligence data mining systems in terms of WQM time, accuracy, sensitivity, and protective performance. The experimental results showed that the optical system based on artificial intelligence data mining took an average of 2.7 days in WQM and the average accuracy was 85.95%. The average sensitivity value was 84.19% and the average protective score was 8.46 points. This indicated that artificial intelligence data mining optical technology had vital significance and value for WQM.
Keywords: Water quality monitoring, Optical systems for data mining, Water pollution, Visible spectrum technology
Subject terms: Environmental sciences, Environmental chemistry, Environmental impact
Introduction
Water is essential to humans, people’s daily production and life cannot do without water resources. It is the pillar of human life and also a necessary resource for human production and labor. However, with the progress of technology and the rapid development of the economy, the pollution of water resources is becoming more serious. People need to monitor water quality in real time, protect and treat water pollution as soon as possible to ensure people’s water safety. Facing the enormous pressure of water pollution, it is very important to scientifically and effectively measure relevant water quality indicators. With the continuous progress of technology, the monitoring technology for water quality parameters is also constantly upgrading and improving. Artificial intelligence data mining optical systems are an important method for monitoring water quality. Through optoelectronic detection technology (such as external spectroscopy, laser light propagation, etc.), water quality phase data and information can be directly monitored without time-consuming or laborious work. It can monitor water quality information in real-time without generating additional pollutants to the environment and has strong practicality in water environments.
The main causes of water pollution are industrial production water, farm water and people’s living water. Monitoring water quality parameters can determine the distribution of pollutants in water bodies, and predict the trend of water pollution changes, so as to determine the impact of water pollution on human health and evaluate the actual effectiveness of pollution prevention and control measures. It is precisely because monitoring water quality is so important that many experts and scholars have conducted extensive research on monitoring water quality parameters. Wang S M believed that remote sensing technology could provide support for monitoring lake water quality, overcoming the characteristics of low manual sampling cost and poor timeliness. Therefore, it was widely applied in lake WQM. Remote sensing could be used to monitor many indicators, including suspended solids, chlorophyll, soluble organic matter, dissolved oxygen, transparency and so on. Although remote sensing provided a new method for monitoring lake water quality, there were still some problems in practical applications1. Vikesland P J’s research found that nanomaterial sensors were being designed for efficient, multifunctional and highly flexible sensing applications. He explored how nanotechnology supported sensors had enormous, yet to be realized and promising prospects for providing widespread and potential low-cost monitoring applications for chemicals, microorganisms, and other analytes in water2. Prapti D R studied the key indicators for WQM in aquaculture and found that temperature and dissolved oxygen were priority water quality parameters in Internet of Things aquaculture. In aquaculture, monitoring water quality was a very important thing. The temperature and dissolved oxygen in water were key factors for biological survival, so real-time monitoring of these indicators was very important3. Ighalo J O found through research that adaptive neural fuzzy inference systems and artificial neural networks were the most widely used artificial intelligence models in WQM and evaluation in the past decade. The neural fuzzy inference system and artificial neural network had the highest prediction accuracy for surface water quality. He evaluated recent literature and proposed knowledge gaps and future prospects in this research field4.
Artificial intelligence data mining optical systems provide a large amount of data and technical support in monitoring water quality, which can provide scientific basis for relevant departments to formulate plans for water resource management. Many people have conducted research on the application of data mining optical systems in monitoring water quality parameters. Guo H selected bands and adjusts parameters based on new data from different fields. The water quality mapping obtained from Sentinel-2 images provided a complete spatial coverage of the entire water surface water quality characterization and helped identify illegal discharges into urban water bodies. This study provided a practical and efficient new strategy for WQM for the management of small water bodies5. Lakshmikantha V believed that water pollution was one of the main threats in recent years as applied water was being polluted. Polluted water could bring various diseases to humans and animals, thereby affecting the lifecycle of ecosystems. If water pollution was detected early, appropriate measures could be taken to avoid serious situations. In order to ensure the supply of purified water, real-time detection of water quality was necessary6. Alam A U believed that rapid, accurate and affordable monitoring of water quality parameters was essential for sustained water security, especially in areas with limited resources. Most traditional sensing systems either could only monitor one parameter at a time or lack user-friendly on-site monitoring capabilities. A fully integrated electrochemical sensor array was a good solution to this obstacle7. Wang X believed that significant progress was made in using remote sensing technology to monitor and evaluate water quality, especially in inland lakes, in the past few decades. However, there was still room for further improvement in terms of spatiotemporal coverage, indicator list, human land fusion, inversion accuracy and model generalization ability8.
In order to explore the effectiveness of artificial intelligence data mining optical systems for monitoring water quality parameters, traditional water quality parameter monitoring methods would be introduced first and then the application of artificial intelligence data mining optical systems in monitoring water quality parameters would be introduced. Finally, experiments were conducted to examine the advantages and disadvantages of the artificial intelligence based data mining optical system in monitoring water quality parameters compared to traditional water quality parameter monitoring methods. Finally, the effectiveness of the artificial intelligence based data mining system in detecting water quality parameters was evaluated through experimental data analysis.
Monitoring and evaluation of water quality parameters
Problems with traditional WQM methods
After years of development, the existing methods for detecting water quality mainly include physical monitoring, electrochemical monitoring, and biological detection. However, the existing detection methods have shortcomings and are no longer able to meet the current needs of environmental WQM. More advanced monitoring technologies are needed9. Water pollution mainly includes agricultural pollution, industrial pollution and residential water pollution, as shown in Fig. 1. However, traditional WQM still has shortcomings in monitoring several aspects of water quality, mainly reflected in incomplete monitoring projects, insufficient monitoring frequency, inaccurate monitoring data, unreasonable layout of monitoring stations, and insufficient diversification of governance methods.
Fig. 1.
Water pollution in daily life.
The project is not comprehensive: Currently, WQM projects still lean towards traditional chemical indicators such as PH (Potential of Hydrogen), dissolved oxygen, ammonia nitrogen, etc., while lacking comprehensive monitoring of other harmful substances such as microorganisms, organic matter, heavy metals, etc. in water10.
Insufficient frequency: Traditional WQM methods rely on regular sampling and monitoring, but this method cannot reflect the real-time status of water quality and is difficult to detect unexpected events. It is necessary to increase the monitoring frequency through real-time monitoring and remote monitoring methods.
Inaccurate data: Currently, the accuracy of WQM data needs to be improved, which often leads to errors and omissions. Therefore, it is necessary to introduce more precise monitoring equipment and technology to ensure the accuracy and reliability of data.
Unreasonable site layout: There are deficiencies in the current WQM site layout, with insufficient coverage in some areas and repeated coverage in some places. Therefore, it is necessary to optimize and adjust the monitoring points to ensure comprehensive coverage.
Treatment means are not diversified enough: For water quality problems, the current treatment means focus on water purification and purification, lacking more targeted and diversified treatment means, such as ecological restoration, to improve the self purification ability of water quality.
In summary, traditional environmental WQM has multiple problems, and measures need to be strengthened in terms of technological innovation, layout optimization, and diversification of treatment methods to improve the level and effectiveness of WQM11.
Design of WQM system
System evaluation
By analyzing the WQM system, the safe and normal operation of the monitoring system was ensured. Although traditional WQM systems have to some extent achieved the detection and analysis of water quality indicators such as COD (Chemical Oxygen Demand), TOC (Total Organic Carbon), TURB (Turbidity), and nitrate nitrogen12,13, due to the lack of effective monitoring methods, potential safety hazards cannot be detected in a timely manner and cannot be analyzed. Therefore, it is necessary to build a dynamic monitoring system to monitor the monitoring equipment in the system. Due to the complexity of WQM systems, it is very difficult to safely control them. Although WQM systems have developed well, they have a strong dependence on management. Therefore, it is necessary to have a technical method that can monitor in real-time14.
Water quality parameter monitoring system based on data mining optical technology
When optical systems monitor water quality parameters through spectroscopy, a large amount of spectral data is generated. Different water quality parameters (such as pollutant concentration, turbidity, dissolved oxygen, etc.) will show unique spectral characteristics within a specific wavelength range. Through data mining technology, the characteristic values or patterns in the spectral signal are analyzed and extracted to find the spectral characteristics related to specific water quality parameters. After feature extraction, the signal is interpreted in combination with artificial intelligence algorithms to map complex spectral data to specific water quality parameters. Data mining plays a core intelligent role in the construction of the entire optical system. From the processing of raw data to the self-optimization of the model, it greatly improves the automation and accuracy of the water quality monitoring system, helps predict future trends in water quality changes, and provides early warnings. Compared with traditional methods such as chemical, electrochemical, and biosensing for WQM, spectral based WQM technology is widely used in modern WQM systems and is an important research direction for modern WQM. At present, WQM technologies based on spectroscopy mainly include hyperspectral remote sensing, atomic absorption spectroscopy and molecular absorption spectrometry15,16. The hyperspectral remote sensing method is mainly used to establish data analysis model by using the water body atlas information and qualitative monitoring and analysis of the water environment. It has the advantages of high Spectral resolution, narrow band, wide spectral range and the integration of the atlas. Atomic absorption spectroscopy is based on the selective absorption of light radiation of different wavelengths when the outer atomic electrons of the element to be measured transition from the ground state to the excited state, so as to achieve the purpose of qualitative and quantitative detection of the element to be measured. Atomic emission spectrum analysis technology is one of the important technical means to detect the content of metal elements in the water environment at present. Molecular absorption spectroscopy is currently the most widely used WQM technology. It monitors the water environment based on the selective absorption of characteristic electromagnetic radiation by the molecules of the substance to be measured. According to the different bands of substance monitoring spectra, molecular absorption spectroscopy water quality detection techniques can be divided into two categories: infrared spectroscopy water quality detection and ultraviolet visible spectroscopy water quality detection. Infrared spectroscopy is a qualitative and quantitative analysis and detection of water quality using the selective absorption characteristics of the molecules to be tested for infrared rays. It has the advantages of less time consumption, simple operation, and no secondary pollution. It is mainly used as a national standard method for detecting oil content in water bodies. All three methods have their advantages. However, when detecting in complex water environments, there are problems such as severe band overlap and large spectral width, which also limit the widespread application of this method.
In comparison, the ultraviolet-visible spectroscopy water quality detection technology has advantages such as simple operation, fast analysis speed, high measurement accuracy, good repeatability, and no secondary pollution. It is suitable for real-time, online, in situ, and multi-parameter measurement of the water environment, meeting the technical requirements of modern water quality detection and monitoring. It has become an important research direction for the development of modern water quality detection instruments and equipment.
Ultraviolet-visible spectroscopy WQM technology
The ultraviolet-visible spectrum WQM technology is an important component of artificial intelligence data mining optical systems. Currently, the ultraviolet-visible spectrum WQM technology is mainly applied to the detection and analysis of water quality indicators such as COD, TOC, TURB, and nitrate nitrogen17. Different water quality has different spectral characteristics. The ultraviolet-visible spectrum WQM technology can effectively observe the distribution of chlorophyll, suspended solids, etc. in the water body, so as to identify common problems in water pollution such as Eutrophication and algae outbreaks. At the same time, it can carry out large-scale WQM, dynamic tracking pollution and other advantages. According to different detection methods, ultraviolet-visible spectroscopy water quality detection techniques are divided into three categories: single wavelength method18, multi wavelength method19, and full spectrum method20,21.
The single wavelength water quality detection technology is mainly based on the good correlation between the absorbance of the tested substance at a single wavelength and water quality parameters in the water environment. By establishing a mathematical model to analyze the data, the content of water quality indicators can be accurately measured.
Multi wavelength method: Compared to the application of single wavelength method, the multi wavelength water quality detection technology mainly uses reference wavelength as turbidity compensation for measuring and analyzing water quality parameters, thereby eliminating the impact of most suspended solids in water on absorbance.
Full spectrum method: The full spectrum water quality detection technology utilizes the comprehensive information of the tested substance carried in the ultraviolet-visible spectral band to establish the full spectrum and corresponding parameter content. Compared to single wavelength and multi wavelength methods, the full spectrum method effectively reduces interference caused by background noise, stray light, and unstable light sources. It is more suitable for water environments and greatly improves the measurement accuracy of water quality parameters.
Due to the fact that the electrons in the atoms or molecules that make up matter are always in a state of motion, the matter keeps moving, and the electrons in their respective energy levels have a certain amount of energy. When these electrons are excited by physical conditions and absorb the external radiation energy, they would transition from a lower energy level to a higher energy level. The generation mechanism of ultraviolet-visible spectrum is due to the combination of external factors such as light, electricity, and heat, which are used for the movement of electrons in matter, causing them to undergo energy level transitions and absorb or radiate a certain amount of energy. The relationship between the energy difference between the two energy levels before and after the transition and the wavelength of the absorbed radiation spectral line follows the Planck condition, which is as follows:
 |
1 |
In the formula, 
is the energy level of the electron; h is the Planck constant; v is the radiation frequency; 
is the wavelength.
The transition occurs only when the electron satisfies Formula (1), and the molecular absorption spectrum of a substance is generated during the transition. The complexity of molecular absorption spectra is due to the complex energy level changes involved in the internal motion of molecules. The total energy E of a molecule is the sum of its intrinsic energy 
, continuous translational energy 
, vibrational energy 
, rotational energy 
, and quantized electron motion energy 
. The formula is as follows:
 |
2 |
The internal motion of molecules mainly includes the motion of electrons relative to the atomic nucleus, the relative vibration of atoms at equilibrium positions, and the rotation of molecules themselves around the center of mass. Therefore, the energy change 
of a molecule absorbing external radiation is the sum of vibrational energy 
, rotational energy 
, and electron motion energy 
. The formula is as follows:
 |
3 |
The material shows the ability to selectively absorb light of different wavelengths, and the energy selectively absorbed is equivalent to the Radiant energy of energy E. When the electrons inside the molecule undergo energy level transition, the vibration energy level and rotation energy level would also undergo transition, and the absorption or radiation is the spectrum in the ultraviolet and visible regions. Due to the vastly different quantization energy levels of various substance molecules, their selective absorption of light reflects their different internal structures, which is also the theoretical analysis basis for spectroscopic water quality detection technology.
Implementation of artificial intelligence data mining monitoring optical system
This article introduced the design and implementation method of a WQM system based on artificial intelligence data mining technology. The designed WQM system achieves real-time collection of common water quality parameter information. It is based on real-time analysis to monitor various water quality information in real-time, as well as to monitor various abnormal situations and issue real-time equipment alarms, ensuring the mining and integrated analysis of historical water quality data. In addition, it can also integrate and analyze data. The monitoring system mainly includes control and data acquisition units, sensor measurement modules, gas circuit units, water circuit units, dissolved oxygen sensors, chlorophyll sensors, etc. The WQM system based on data mining optical technology mainly consists of six functional modules: data monitoring, data audit, data analysis, data validation, platform management, and module management. The specific interface of the WQM system is shown in Fig. 2:
Fig. 2.
WQM system interface.
Data monitoring refers to the use of optical systems to easily see WQM data at different points, thus reflecting the distribution and changes of river and lake water quality over a wider range of time and space, and achieving timely warning, tracing, and disposal of abnormal conditions such as water pollution events. This provides support for achieving long-term standardized and refined management of water quality.
Data auditing22 refers to the verification and inspection of collected data, as well as the verification of historical data to ensure its authenticity and reliability.
Data analysis refers to the analysis of data obtained through data auditing to understand the water quality situation of a certain region during a certain period of time, and provide a basis for water pollution prevention and control.
Data validation refers to conducting experiments on audited and analyzed data to compare and verify the authenticity and usability of the data.
Effectiveness experiment of WQM
WQM is an important prerequisite for water environment protection. Environmental protection workers can trace the source of pollution23,24 to the bottom based on monitoring data results, conduct comprehensive analysis and research on the actual distribution of water pollution, and ultimately form scientific, objective and authentic environmental evaluation results. Where there is water, WQM exists, especially for living sewage, industrial sewage, surface water, etc. It is not only necessary to achieve 24-hour continuous monitoring, but also to scientifically and effectively manage the water quality that does not meet the standards after testing, and to discharge and use it. Therefore, WQM is extremely important. Optical systems based on artificial intelligence data mining technology are extremely important for monitoring water quality, and are an innovation in the WQM industry, providing new ideas for water environment protection. A comparative experiment was conducted here, with the experimental group monitoring water quality using artificial intelligence data mining technology and optical technology, while the control group monitoring water quality using traditional methods. In order to investigate the effectiveness of artificial intelligence data mining technology and optical technology in WQM, four indicators of monitoring time, accuracy, sensitivity, and protective performance were set in the experiment, and 10 water plants were selected as the experimental subjects. The specific evaluation results of each indicator are shown in the following experiment.
Monitoring time
WQM time refers to the amount of time required for monitoring work. The more monitoring projects there are, the higher the difficulty and the longer the required time. The time of WQM is also an important indicator to measure the efficiency of a certain monitoring method, and the length of monitoring time not only affects the progress and effectiveness of work. Only by shortening the duration of WQM and improving efficiency can WQM work be completed efficiently. In a series of links in WQM, various links may affect the duration of WQM. The method used for WQM also affects the duration of WQM. According to experiments, the duration of WQM is generally within one week. If this time is exceeded, the WQM time would be longer when the monitoring project is fixed. If the WQM time can be controlled within one week, or even shorter, the efficiency of WQM is relatively high. Based on artificial intelligence data mining, optical systems for monitoring water quality can effectively reduce the time required for WQM and improve the efficiency of monitoring. The traditional WQM method not only requires a lot of time and manpower, but also requires a lot of money. Therefore, optical systems based on artificial intelligence data mining have significant advantages in terms of WQM time. The WQM time is shown in Fig. 3.
Fig. 3.
Monitoring time.
From the figure, it could be seen that in the monitoring of water quality based on artificial intelligence data mining optical systems, the third and ninth water plants had the longest WQM time, which was 4 days. The shortest WQM time was at the first, fourth, sixth, seventh, and tenth water plants, which took 2 days. The average monitoring time was 2.7 days. Based on traditional monitoring methods, the fourth and seventh water plants had the longest WQM time, which was 6 days. The shortest WQM was at the 8th water plant, which took 3 days. The average detection time was 4.6 days. The experimental results showed that using artificial intelligence based data mining optical systems to monitor water quality could shorten detection time, improve monitoring efficiency, and bring more profits and market competitive advantages to enterprises.
-
(2)
Accuracy
The accuracy of WQM refers to the degree to which the measured values of WQM match the true values. The accuracy process of monitoring data is influenced by the on-site fixation, storage, transmission, and analysis of samples. Generally, it is reflected by the accuracy of monitoring data. The improvement of accuracy can make the results of WQM more convincing and authentic. In traditional WQM methods, the WQM system is huge. This monitoring method has slow speed, long monitoring time, complex operation, poor stability, and cannot obtain accurate information on water pollution in the first time. Based on artificial intelligence data mining, optical systems can accurately extract and save data from various stages of WQM, ensuring the accuracy of WQM data. In addition to monitoring methods, the main factors that affect the accuracy of WQM include the source and type of water quality. The accuracy of WQM is shown in Fig. 4.
Fig. 4.
Accuracy of WQM.
From the figure, it could be seen that in the monitoring of water quality using artificial intelligence data mining optical systems, the 10th water plant had the highest accuracy in WQM, accounting for 92.6%. The fifth water plant had the lowest accuracy in WQM, accounting for 80.1%. The average accuracy of WQM was 85.95%. Based on traditional monitoring methods, the 10th water plant had the highest accuracy in WQM, with a rate of 90%. The eighth water plant had the lowest accuracy in WQM, accounting for 50.4%. The average accuracy of WQM was 74.08%. Therefore, using artificial intelligence data mining optical systems to monitor water quality could provide more accurate WQM data.
-
(3)
Sensitivity
The water quality data monitored by WQM systems is usually an important standard for measuring water quality safety. With the improvement of water quality standards, the accuracy and precision of past water quality analyzers25 can no longer meet the current monitoring needs, so it is necessary to improve the sensitivity of WQM instruments. Sensitivity26 is a very important indicator in WQM, and the importance of sensitivity is reflected in the early detection of water pollution problems through high sensitivity monitoring methods, thereby improving the accuracy of monitoring. Traditional WQM methods have problems with low sensitivity, slow monitoring speed, and complex operation in terms of WQM sensitivity. The optical system based on artificial intelligence data mining has the characteristics of high sensitivity, high precision, fast monitoring, and easy operation in monitoring water quality, which can accurately detect various water quality indicators, such as PH value, dissolved oxygen, hardness, ammonia nitrogen27,28, etc. The sensitivity of WQM is shown in Fig. 5.
Fig. 5.
Sensitivity of WQM.
From the graph, it could be seen that the first water plant had the highest sensitivity in WQM based on artificial intelligence data mining optical systems, with a sensitivity of 90.36%. The lowest sensitivity of WQM was the fourth water plant, which was 70.65%. The average sensitivity of WQM was 84.19%. Based on traditional monitoring methods, the fourth water plant had the highest sensitivity in WQM, accounting for 86.32%. The sensitivity of WQM was the lowest at the first water plant, which was 56.32%. The average sensitivity of WQM was 66.84%. Therefore, based on artificial intelligence data mining, optical systems had higher sensitivity in monitoring water quality.
-
(4)
Protective
The protective nature of WQM refers to the fact that WQM not only provides data and information for environmental management29, but also serves as an important foundation for treating water pollution, managing and protecting water resources. Therefore, WQM has protective properties. WQM can effectively monitor water pollution30 issues. By detecting data and timely controlling and treating water pollution, it can effectively control water environmental pollution, prevent and maintain pollution sources, and achieve the goal of protecting the environment. It has a certain supervisory function and can be promptly treated to reduce water source pollution when discovering pollution sources. The traditional WQM method mainly relies on monitoring data without developing prevention and control plans for potential hazards. Based on artificial intelligence data mining, optical systems can detect potential hazards in WQM in advance and develop targeted prevention and control plans. As shown in Table 1, the protective scores for WQM are shown.
Table 1.
WQM protection score.
| Water works | Optical based system | Based on traditional methods |
|---|---|---|
| 1 | 8.5 | 6.0 |
| 2 | 9.0 | 7.5 |
| 3 | 9.5 | 8.0 |
| 4 | 7.0 | 6.7 |
| 5 | 7.5 | 5.5 |
| 6 | 8.5 | 7.6 |
| 7 | 8.0 | 9.0 |
| 8 | 8.8 | 6.8 |
| 9 | 8.6 | 7.2 |
| 10 | 9.2 | 7.0 |
| Average | 8.46 | 7.13 |
From Table 1, the protective scores of WQM based on artificial intelligence optical systems and monitoring based on traditional methods could be obtained. Among them, the highest score for the protection of WQM based on artificial intelligence optical systems was 9.2 points, and the lowest score was 7.0 points. The average protective score was 8.46 points. The highest score for the protection of WQM based on traditional methods was 9.0, and the lowest score was 5.5. The average protective score was 7.13 points. From this, it could be concluded that the optical system monitoring water quality based on artificial intelligence data mining had good protection.
Conclusions
With the development of science and technology, using artificial intelligence data mining optical systems to monitor water quality has become a trend in various countries’ WQM. Due to the problems of slow speed, long monitoring time, complex operation, and poor stability in traditional WQM methods, this paper studied the monitoring of water quality using optical systems based on artificial intelligence data mining. It was found that this mode had higher accuracy, sensitivity, and protection, and could improve the efficiency of WQM to meet the needs of customers and the market. Specifically, using artificial intelligence data mining optical systems to monitor water quality had strong advantages, and this mode was of great significance for the governance of the water environment. Although this model had enormous potential and development space in clothing design, there were also some problems and challenges, such as the difficulty of applying data mining optical systems, water quality control measures, and so on. Therefore, in the future, it is necessary to continuously conduct in-depth research and exploration in the development of technology and application implementation, so as to achieve the widespread application and promotion of digital image processing in clothing design.
Author contributions
All authors reviewed the manuscript.
Data availability
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Wang, S. M. & Qin, B. Q. Research progress on remote sensing monitoring of lake water quality parameters. Huanjing Kexue. 44 (3), 1228–1243 (2023). [DOI] [PubMed] [Google Scholar]
- 2.Vikesland, P. J. Nanosensors for water quality monitoring. Nat. Nanotechnol.13 (8), 651–660 (2018). [DOI] [PubMed] [Google Scholar]
- 3.Prapti, D. R., Mohamed Shariff, A. R. & Che Man, H. Internet of things (IoT)-based aquaculture: an overview of IoT application on water quality monitoring. Reviews Aquaculture. 14 (2), 979–992 (2022). [Google Scholar]
- 4.Ighalo, J. O., Adeniyi, A. G. & Marques, G. Artificial intelligence for surface water quality monitoring and assessment: a systematic literature analysis. Model. Earth Syst. Environ.7 (2), 669–681 (2021). [Google Scholar]
- 5.Guo, H., Huang, J. J. & Chen, B. A machine learning-based strategy for estimating non-optically active water quality parameters using Sentinel-2 imagery. Int. J. Remote Sens.42 (5), 1841–1866 (2021). [Google Scholar]
- 6.Lakshmikantha, V., Hiriyannagowda, A. & Manjunath, A. IoT based smart water quality monitoring system. Glob. Transit. Proc.2(2), 181–186 (2021).
- 7.Alam, A. U., Clyne, D. & Jin, H. Fully integrated, simple, and low-cost electrochemical sensor array for in situ water quality monitoring. ACS Sens.5 (2), 412–422 (2020). [DOI] [PubMed] [Google Scholar]
- 8.Wang, X. & Yang, W. Water quality monitoring and evaluation using remote sensing techniques in China: a systematic review. Ecosyst. Health Sustain.5 (1), 47–56 (2019). [Google Scholar]
- 9.Akhter, F., Siddiquei, H. R. & Alahi, M. E. E. An IoT-enabled portable water quality monitoring system with MWCNT/PDMS multifunctional sensor for agricultural applications. IEEE Internet Things J.9 (16), 14307–14316 (2021). [Google Scholar]
- 10.Neale, P. A., Escher, B. I. & de Baat, M. L. Application of effect-based methods to water quality monitoring: answering frequently asked questions by water quality managers, regulators, and policy makers. Environ. Sci.Technol. 57(15), 6023–6032 (2023). [DOI] [PubMed] [Google Scholar]
- 11.Nugroho, Y. A. & Pratama, M. F. Designing a mobile application for pool water quality monitoring based on the internet of things. Tekinfo Jurnal Ilmiah Teknik Industri Dan. Informasi. 9 (1), 47–55 (2021). [Google Scholar]
- 12.Al, R. Relevance of sensor and chemical technology in water quality monitoring. Turkish J. Comput. Math. Educ. (TURCOMAT). 12 (3), 3815–3818 (2021). [Google Scholar]
- 13.Dutta, K., Daverey, A. & Sarkar, A. Surface and groundwater pollution: monitoring and remediation methods. Environ. Qual. Manage.31 (4), 9–19 (2022). [Google Scholar]
- 14.Akhter, F., Alahi, M. & Mukhopadhyay, S. C. Recent advancement of the sensors for monitoring the water quality parameters in smart fisheries farming. Computers. 10 (26), 1–20 (2021). [Google Scholar]
- 15.Mokua, N., Maina, C. W. & Kiragu, H. A raw water quality monitoring system using wireless sensor networks. Int. J. Comput. Appl.174 (21), 35–42 (2021). [Google Scholar]
- 16.Junior, A., Munoz, R. & Quezada, A. Internet of water things: a remote raw water monitoring and control system. IEEE Access.9 (99), 35790–35800 (2021). [Google Scholar]
- 17.Liang, Y., Rozemeijer, J. C. & Broers, H. P. Drivers of nitrogen and phosphorus dynamics in a groundwater-fed urban catchment revealed by high-frequency monitoring. Hydrol. Earth Syst. Sci.25 (1), 69–87 (2021). [Google Scholar]
- 18.Feder, I., Duadi, H. & Fixler, D. Single wavelength measurements of absorption coefficients based on iso-pathlength point. Biomedical Opt. Express. 11 (10), 5760–5771 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yao, Y. et al. Broadband X-ray ptychography using multi-wavelength algorithm. J. Synchrotron Radiat.28 (1), 309–317 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Guo, Y. et al. Advances on water quality detection by uv-vis spectroscopy. Appl. Sci.10 (19), 6874 (2020). [Google Scholar]
- 21.Jan, F., Min-Allah, N. & Düştegör, D. Iot based smart water quality monitoring: recent techniques, trends and challenges for domestic applications. Water. 13 (13), 1729 (2021). [Google Scholar]
- 22.Eilifsen, A. et al. An exploratory study into the use of audit data analytics on audit engagements. Acc. Horizons. 34 (4), 75–103 (2020). [Google Scholar]
- 23.Madhav, S. et al. Water pollutants: sources and impact on the environment and human health. Sens. Water Pollutants Monitoring: Role Mater. 43–62. (2020).
- 24.Yang, Z. et al. Spatiotemporal variation characteristics and source identification of water pollution: insights from urban water system. Ecol. Ind.139, 108892 (2022). [Google Scholar]
- 25.Lin, B. et al. A multi-parameter in-situ water quality analyzer based on a portable document scanner and 3D printed self-sampling cells. Anal. Chim. Acta. 1101, 176–183 (2020). [DOI] [PubMed] [Google Scholar]
- 26.Calvo-López, A., Alonso-Chamarro, J. & Puyol, M. Highly versatile and automated total ammonia nitrogen compact analyzer suitable for different types of water samples. Environ. Science: Water Res. Technol.9 (12), 3366–3376 (2023). [Google Scholar]
- 27.Akhtar, N. et al. Modification of the water quality index (WQI) process for simple calculation using the multi-criteria decision-making (MCDM) method: a review. Water. 13 (7), 905 (2021). [Google Scholar]
- 28.Uddin, M. G. et al. A comprehensive method for improvement of water quality index (WQI) models for coastal water quality assessment. Water Res.219, 118532 (2022). [DOI] [PubMed] [Google Scholar]
- 29.Cyr-Gagnon, J. & Rodriguez, M. J. Optimizing data management for municipal source water protection. Land. use Policy. 100, 103788 (2021). [Google Scholar]
- 30.Martínez, R. et al. On the use of an IoT integrated system for water quality monitoring and management in wastewater treatment plants. Water. 12 (4), 1096 (2020). [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.