Abstract: As an invasive salt marsh plant, Spartina alterniflora poses a significant threat to the ecological health of coastal areas in our country. Due to different objects have same spectrum and the same objects have different spectrums, identifying Spartina alterniflora with remote sensing imagery is often plagued by low accuracy. Based on time series satellite remote sensing data, this study constructed a remote sensing index of Spartina alterniflora based on phenological characteristics to accurately identify the spatial distribution of Spartina alterniflora. Initially, the time series satellite remote sensing data were screened for cloud cover and median synthesis, and the Normalized Difference Vegetation Index (NDVI) was calculated to construct NDVI time series data. Subsequently, the NDVI time series were smoothed using the Savitzky-Golay (S-G) filter, and coefficients of phenological characteristic were computed by the two-term Fourier function. These coefficients helped analyze the characteristic curves and distribution patterns of various objects within the phenological characteristic space, enabling the construction of a tailored remote sensing index for Spartina alterniflora. Finally, the threshold method was used to segment the remote sensing index of Spartina alterniflora, and post-processing was performed in combination with the Digital Elevation Model (DEM) and mathematical morphology methods to obtain the remote sensing precise extraction results of Spartina alterniflora. The experimental results on the south bank of Hangzhou Bay showed that the proposed method achieved accurate identification of Spartina alterniflora, with the overall accuracy (OA), producer accuracy (PA), user accuracy (UA), and Kappa coefficient of 90.50%, 91.50%, 89.71% and 0.81 respectively. This study can offer a feasible solution for monitoring and extracting Spartina alterniflora, thereby providing important support for the high-quality, sustainable development of coastal resources.