Adjacency Effect in Nearshore Aquatic Remote Sensing: Modelling, Correction, and Application

Abstract

The adjacency effect (AE) alters the top-of-atmosphere signals of coastal and inland waters, and it poses a major challenge for remote sensing of nearshore aquatic environments. To address this, we developed a Monte Carlo-based 3D radiative transfer model to study the AE, validated its accuracy against existing codes, and conducted case studies to demonstrate its application in analyzing the impact of AE in custom environments. In addition, we introduced a methodology and code for AE correction and demonstrated significant improvements in satellite-derived water-leaving reflectance retrievals using globally distributed in situ reflectance measurements. The tool, named T-Mart, is open-source and publicly available (https://github.com/yulunwu8/tmart). We applied AE correction and evaluated the performance and limitations of satellite-based water quality retrievals in small rivers traversing agricultural lands in Eastern Ontario, Canada. Satellite-derived reflectance and water quality parameters were validated against in situ measurements collected from May to October 2023. In the South Nation River and the Ottawa River, turbidity can be reliably monitored. Despite the improved retrievals through AE correction, further work is required to accurately monitor coloured dissolved organic matter and chlorophyll-a. While the findings highlight the complexities of satellite-based water quality monitoring applications for small rivers, AE correction represents a crucial step towards more accurate aquatic remote sensing of inland waters, laying the groundwork for more refined methodologies in future studies.