Real-time monitoring of physicochemical parameters in water using big data and smart IoT sensors

Naresh Sharma,Rohit Sharma

DOI: https://doi.org/10.1007/s10668-022-02142-8

2022-02-11

Abstract:Water pollution is the most important factor affecting the environment. Appropriate monitoring is a big challenge to make sustainable growth by maintaining it for society. In recent times, water monitoring has turned into a smart monitoring system for water pollution (SMS-wp), with the advances on the Internet of things (IoT), machine learning (ML), and the improvement in current sensors. River Ganga is one of the major sources of water for drinking, irrigation, and industries in the northern part of India. Day by day, Ganga River is getting polluted, due to anthropogenic activities, such as the construction of dams, extensive use of fertilizers in agriculture, and untreated discharges from industries. Contamination in the river water is adversely affecting human health and river biota. Therefore, to improve the river ecosystem and to check infections and diseases, water quality assessment is very much important. The main aim of this study is to determine the Water Quality Index (WQI) of the River Ganges at the upper part of the Indo-Gangetic plain, just downstream of the Himalayan foothill using the last 3 years of data (2017–2019). Trend analysis for River Ganga water at considered locations is also a part of this study. Trend analysis is presenting the water quality of river Ganga in the coming years up to 2025. Twelve physicochemical parameters (TDS, chlorides, alkalinity, DO, temperature, COD, BOD, pH, magnesium, hardness, total coliform, and calcium) were analyzed to determine the water quality of River Ganga. As a result, WQI for next 5 years (from 2020 to 2025) is forecasted as an increment of 17.34% at Haridwar, 4.16% at Roorkee, and 21.63% at Dehradun. Results of the study indicated that WQI values just downstream of the Himalayan foothills in the upper reaches of the Gangetic plain are increasing every year. The authors have concentrated on how the advances in sensor innovation, the Internet of things, and machine learning techniques make water pollution monitoring a genuinely brilliant checking framework. Finally, the system of robust strategies for ML, denoising techniques, and advancement of appropriate guidelines for wireless sensor networks (WSNs) have been recommended.