Water quality control and management in water resources are important for providing clean and safe water to the public. Due to their large area, collection, analysis, and management of a large amount of water quality data are essential. Water quality data are collected mainly by manual field sampling, and recently real-time sensor monitoring has been increasingly applied for efficient data collection. However, real-time sensor monitoring still relies on only a few parameters, such as water level, velocity, temperature, conductivity, dissolved oxygen (DO), and pH. Although advanced sensing technologies, such as hyperspectral images (HSI), have been used for the areal monitoring of algal bloom, other water quality sensors for organic compounds, phosphorus (P), and nitrogen (N) still need to be further developed and improved for field applications. The utilization of information and communications technology (ICT) with sensor technology shows great potential for the monitoring, transmission, and management of field water-quality data and thus for developing effective water quality management. This paper presents a review of the recent advances in ICT and field applicable sensor technology for monitoring water quality, mainly focusing on water resources, such as rivers and lakes, and discusses the challenges and future directions.

Discharge from sewage treatment plants (STPs) is a significant pathway of entry for microplastics (MPs) to the environment. Therefore, STPs should be considered as an important barrier to the distribution and circulation of MPs in the aquatic environment. In this study, the fate and material-specific properties of MPs were investigated in an STP-equipped and granule-activated carbon (GAC) tower with a thermal regeneration system. This system functioned with a tertiary treatment unit. The GAC with thermal regeneration removed 92.8% of MPs and was useful for removing MPs with a specific gravity less than that of water and with a size of 20–50 µm, which had negligible removal in the conventional STP process. In addition, a lab-scale electric-coagulation experiment was conducted to examine its potential utility as a pretreatment process for further enhancing the removal efficiency of MPs by GAC. After 30 min of electro-coagulation using aluminum electrodes, 90% of MPs were converted into separable flocs by centrifugation. These flocs may be effectively removed by GAC or other tertiary treatment steps. This study demonstrates that GAC with thermal regeneration is a tertiary process that can efficiently prohibit the release of MPs from STPs and circulation of MPs in the natural environment.

Algal blooms have various effects on drinking water supply systems; thus, proper monitoring is essential. Traditional visual identification using a microscope is a time-consuming method and requires extensive labor. Recently, advanced machine learning algorithms have been increasingly applied for the development of object detection models. The You-Only-Look-Once (YOLO) model is a novel machine learning algorithm used for object detection; it has been continuously improved in newer versions, and a tiny version of each standard model presented. The tiny versions applied a less complicated architecture using a smaller number of convolutional layers to enable faster object detection than the standard version. This study compared the applicability of the YOLO models for algal image detection from a practical aspect in terms of classification accuracy and inference time. Therefore, automated algal cell detection models were developed using YOLO v3 and YOLO v4, in which a tiny version of each model was also applied. The cell images of 30 algal genera were used for training and testing the models. The model performances were compared using the mean average precision (mAP). The mAP values of the four models were 40.9, 88.8, 84.4, and 89.8 for YOLO v3, YOLO v3-tiny, YOLO v4, and YOLO v4-tiny, respectively, demonstrating that YOLO v4 is more precise than YOLO v3. The tiny version models presented noticeably higher model accuracy than the standard models, allowing up to ten times faster object detection time. These results demonstrate the practical advantage of tiny version models for the application of object detection with a limited number of object classes.