Development of a Multi-temporal Image Differencing approach for identifying invasive water hyacinth (Eichhornia crassipes)

Invasive aquatic plants, such as water hyacinth ( Eichhornia crassipes ), pose severe ecological threats in China by obstructing waterways, altering hydrodynamics, and affecting biodiversity. Effective monitoring and management of these species are critical for ecosystem protection and sustainable water resource management. Remote sensing provides a scalable approach for mapping spatial-temporal dynamics of invasive vegetation, enabling timely intervention and control. However, conventional remote sensing approaches, including end-of-season phenometrics and supervised classification, struggle with the dynamic changes of floating vegetation due to spectral confusion and variable background reflectance caused by hydrological processes. To address these limitations, we developed a novel Multi-Temporal Image Differencing (MID) framework for monitoring water hyacinth. Our approach integrates high-water-content vegetation (HWV) extraction with pixel-wise temporal differencing to construct a spectral-temporal profile. This profile effectively discriminates water hyacinth from co-occurring native vegetation through dynamic differencing, which captures its free-floating signature, as while suppressing interference from stable background features. Validated with high-resolution Sentinel-2 time-series data, MID effectively captures floating vegetation dynamics and can be applied across diverse flowing water bodies, supporting scalable monitoring under variable hydrological conditions. Comparison with a Random Forest classifier trained on spectral and field data showed that MID more effectively detected dynamic vegetation changes, while the classifier achieved slightly higher overall spatial accuracy (≈ 92%). Independent validation confirmed the robustness of MID under variable hydrological conditions. Overall, these results demonstrate that MID provides a robust and scalable monitoring framework, offering an effective tool for ecological management and decision-making of invasive aquatic plants in flowing waters.