Monitoring Wetland Changes and Analyzing the Spartina alterniflora Invasion in the Yellow River Delta Over the Past 30 Years Based on Google Earth Engine

The spectral similarity of regular and irregular ground objects and the problem of image noise and abnormal pixels in a wide range of ground object extraction have been difficult for accurate wetland classification. This letter proposes a two-order hierarchical classification based on SNIC segmentation (HSNIC), and three new indices are constructed to extract salt marsh vegetation more accurately. Finally, the classification results of Yellow River Delta wetlands from 1990 to 2023 were obtained, and the invasion mechanism of Spartina alterniflora was further analyzed. The results show that: 1) the classification accuracy of the Yellow River Delta by HSNIC reached 88.48%. Compared with RF and object-oriented RF classification, the accuracy is improved; 2) introducing the three new indices significantly enhanced the classification accuracy; 3) the invasion of S. alterniflora influenced the survival environment of local salt marsh vegetation. These research findings provide a basis for decision-making regarding wetland restoration and conservation in the Yellow River Delta.