A Comprehensive Review of Sensor-Based and Spectroscopy-Based Systems for Monitoring Water Quality in Freshwater Aquaculture System

Abstract

Ensuring precise water conditions is essential for the economic viability and preservation of aquatic resources in aquaculture, necessitating effective water quality monitoring systems. This research work investigates and reviews water quality monitoring systems for freshwater aquaculture, focusing on electronic sensor-based and spectroscopy-based methods through a comparative analysis. The review categorizes and evaluates machine learning (ML)-based sensor and spectroscopy methods, emphasizing the performance of sensitive spectral bands linked to diverse water quality parameters. Furthermore, the research examines the efficiency and accuracy of water quality parameters in ML-based water quality monitoring systems for freshwater aquaculture. Comparative findings indicate that ML-based sensor methods exhibit superior quality, versatility, and performance, capitalizing on their ability to exploit unique spectral features. The discussion encompasses challenges and issues faced by ML-based water quality monitoring systems in freshwater aquaculture, providing insights into their future perspectives. This comprehensive investigation contributes valuable insights into the intricate relationship between sensing technologies, machine learning, and water quality monitoring in the context of freshwater aquaculture. It serves as a resource for stakeholders, researchers, and policymakers navigating the challenges of improving aquaculture practices while addressing environmental considerations.