Trends in Remote Sensing of Water Quality Parameters in Inland Water Bodies: A Systematic Review

Sinesipho Ngamile ^1,2 Sabelo Madonsela ^1,3 Mahlatse Kganyago ^1*

• ¹ University of Johannesburg, Johannesburg, Gauteng, South Africa
• ² South African National Space Agency, Pretoria, South Africa
• ³ Council for Scientific and Industrial Research (CSIR), Pretoria, South Africa

Monitoring water quality is crucial for sustainable water management and meeting the United Nations Sustainable Development Goals. Urbanisation, agricultural practices, industrial activities, and population growth increase the presence of biological, chemical and physical properties in water bodies. Traditional water quality monitoring methods (laboratory and in-situ measurements) are limited spatially, temporarily and are costly. Satellite remote sensing has been shown to provide a systematic, cost-effective, and near-real-time alternative. This paper analysed 127 peer-reviewed articles published between 2002 and 2024 from Web of Science and Scopus databases. The final included articles in the review were achieved through the PRISMA flowchart. The review revealed that low-resolution sensors with long-term records, such as MODIS, were commonly applied to study large lakes. In contrast, sensors such as Landsat-8 and Sentinel-2 were applied for both lakes and dams. These sensors contain necessary spectral regions for monitoring water quality, where it was shown that the 500-600nm region is critical for chlorophyll assessment, while the 640-670nm region is used for turbidity. The Secchi disk depth and the total suspended solids were assessed using regions 860-1040nm and 1570-1650nm. Water quality research also focused on countries such as China, India, Brazil, and South Africa, with an emphasis on optically active parameters. There is, however, limited research on non-optically active parameters, such as nitrogen, phosphorus, and temperature, especially in small inland water bodies. Therefore, there is a need for more research in these areas, using direct and indirect methods of water quality parameter estimation with the integration of machine learning algorithms.