Assessment of Optical and Near-Infrared Proximal Remote Sensing for Suspended Sediment Concentration Estimation under Artificial and Ambient Illumination

Proximal remote sensing represents an effective approach for water quality monitoring, enabling the estimation of turbidity and suspended sediment concentration (SSC) through spectral indices, such as red–green band ratios. Low-cost RGB cameras are widely adopted for this purpose, however, their measurements are strongly affected by variations in illumination, shadows, surface glint, and ambient environmental conditions, which can compromise data consistency and reliability. Extending the spectral coverage into the near-infrared (NIR) domain has the potential to enhance sensitivity to suspended sediments and reduce the influence of variable lighting conditions. Although hyperspectral sensors remain costly and impractical for routine monitoring, the analysis of hyperspectral data provides valuable insights into the most informative wavelengths and supports the targeted integration of RGB imagery with selected NIR bands for future field applications.

In this laboratory study, proximal hyperspectral sensing was employed to investigate SSC under both artificial and ambient illumination conditions, using two sediment types with contrasting optical properties (yellowish soil and white China clay). The experiments assess the influence of illumination conditions and sediment characteristics on spectral signatures, and compare the performance of reflectance information derived from the RGB and NIR spectral ranges. The results offer initial insights into sediment–reflectance interactions and contribute to the development of more robust and cost-effective proximal remote sensing strategies for water quality monitoring in real-world environments.

 

Keywords: Proximal remote sensing; hyperspectral data; suspended sediment concentration; laboratory experiments