Spatiotemporal Monitoring of Oxygen-Consuming Organic Pollutants in the Pearl River Delta Estuaries Based on Sentinel-2 MSI Observations and Physics-Based Approach

Estuarine ecosystems, critical interfaces between anthropogenic activities and natural processes, exhibit complex organic pollution patterns that demand sophisticated monitoring approaches. The permanganate index (chemical oxygen demand using potassium permanganate (KMnO<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub>) as oxidant, COD<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Mn</sub>), a principal parameter in China's environmental monitoring framework for aquatic systems, provides crucial insights into organic contamination dynamics. The existing monitoring paradigms predominantly focus on inland water bodies and large-scale coastal regions, leaving estuarine environments characterized by complex hydrodynamic interactions and sediment dynamics insufficiently addressed. Moreover, conventional data-driven empirical models face significant limitations in these transitional ecosystems due to sampling data scarcity and compromised representativeness under variable geomorphologic conditions. This study developed a physics-based inversion model, using Sentinel-2 multispectral data, to overcome these challenges. A comprehensive spatiotemporal analysis of COD<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Mn</sub> distribution patterns and their driving factors was then conducted. The results demonstrated that the model had robust performance (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.7838, mean absolute percentage difference (MAPD) = 13.9%, root-mean-square deviation (RMSD) = 0.3791 mg/L, mean normalized difference (MND) = -5.99%), and outperformed the comparison model. Distinct distribution patterns were first recognized in the boundaries between different estuarine zones of four major Pearl River estuaries: as the pollutants move from the near-mouth section to the river-mouth section, the COD<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Mn</sub> concentration decreases sharply, while pollutants accumulate and linger at the boundary between the river-mouth section and the mouth-outside seashore. Four driving factors (i.e., increased terrestrial organic matter inputs, seasonal variations in rainfall, inadequate urban wastewater infrastructure management, and estuarine engineering modifications) regulate the transport pathways and residence time of pollutants. This study provides new insights for the dynamic monitoring of organic pollution, supporting adaptive management strategies in transitional ecosystems, particularly for pollution source identification, remediation effectiveness assessment, and ecological restoration planning in anthropogenically stressed estuaries.