Abstract: Spartina alterniflora is a common invasive species in coastal wetlands of China. Rapid and accurate identification of S. alterniflora has been one of the key focuses in coastal wetland ecological research. Traditional identification methods primarily rely on manual field surveys, which are labor-intensive, time-consuming, and costly. Moreover, existing remote sensing-based identification studies largely depend on a single normalized vegetation index, which often fails to adequately address spectral confusion between S. alterniflora and co-occurring vegetation species. To address these limitations, taking Xiangshan Port in Ningbo City, Zhejiang Province as the study area, this research utilized the Red Edge Index (REI) generated from the red-edge bands of Sentinel-2 remote sensing imagery, combining with the phenological characteristics of S. alterniflora during its peak growth season (from June to July 2022) and senescence period (from December 2022 to January 2023), constructed an identification model integrating temporal spectral and phenological information. Through the Google Earth Engine platform, obtain and preprocess image data to identify S. alterniflora in the study area, and systematically optimize the number of decision trees to improve model performance. The results showed that compared to the commonly used Normalized Difference Vegetation Index (NDVI), REI provided better identification performance for S. alterniflora. Furthermore, compared to remote sensing identification methods without phenological features, incorporating the phenological characteristics of S. alterniflora reduced the misidentification rate of surrounding plants. When the number of random forest decision trees was set to 7, the overall identification accuracy combining S. alterniflora vegetation phenological features reached 99.18%, with a Kappa coefficient of 0.981 7. The results is significantly better than the classification results using only NDVI or single-phase data. Spatial analysis further indicates that this method can effectively suppress spectral confusion caused by seasonal phase changes and reduce misjudgment between the edge area of the tidal ditch and the remaining reed plant area. This study verified the potential of the red-edge band of sentine1-2 in the identification of vegetation phenology, providing an efficient and feasible technical approach for high-precision remote sensing monitoring of invasive plants in coastal wetlands.