Biomass prediction and invasion assessment of Spartina alterniflora driven by remote sensing big data

Coastal wetlands, as one of the three major ecosystems on Earth, play a critical role in maintaining biodiversity and ecological balance. In recent years, the invasive species Spartina alterniflora has rapidly spread along China's coastal areas, posing a considerable threat to coastal wetland ecosystems. Based on marine remote sensing big data, this study integrated multi-source remote sensing imagery and multi-site observational characteristics to develop a standardized dataset for estimating S. alterniflora biomass, and created corresponding automated identification and remote sensing inversion methods for biomass estimation. Systematic monitoring in the Hangzhou Bay area revealed that between 2019 and 2022, S. alterniflora biomass density continued to increase, with a maximum annual expansion area reaching 1135.7 ha, a northward shift in distribution centroid, and edge-expansion patterns dominating the spread. However, from 2023 to 2024, with enhanced artificial control measures, its distribution area sharply decreased from a peak of 3387.1 ha to 608.1 ha, accompanied by a significant reduction in biomass. This research demonstrates the effectiveness of combining remote sensing technology with machine learning algorithms for biomass monitoring, provides a scientific basis for developing effective management strategies against invasive species, and holds important practical significance for the conservation and restoration of coastal wetland ecosystems.