首页 > , Vol. , Issue () : -
Salt marshes are the most valuable and vulnerable ecosystem in the world, accurate and timely monitoring the distribution of salt marsh vegetation is thus important. With the accumulation of multi-source remote sensing imagery, the time-series method has increasingly become an important means for monitoring coastal areas. However, it is still challenging to construct time series effectively since the number of available observations is relatively low due to frequent cloudy weather in the coastal areas. In this study, we coupled multi-sourced Landsat images and constructed a pixel-level time-series with XGBoost. Based on which, the feasibility and stability for classifying salt marsh vegetation were tested using the three typical sites in Yangtze River Delta. The results showed that, (1) Inter-calibration for multi-sourced images was necessary for not only improving the available of images, but also reducing the spectral differences between sensors. (2) The performance of salt marsh vegetation classification based on the pixel-level time-series was favorable, reflected by 81.50% as the mean overall accuracy and 0.755 as the Kappa coefficient. The classification results were excellent especially for the widely distributed S. salsa and S. alterniflora in Yangtze River Delta. (3) Compared with the single-phrase classifications, the pixel-level time-series based classifications were stable, evidenced by an inter-annual absolute mean error lower than 3.27%. Therefore, our proposed method is expected for dynamic monitoring on salt marsh vegetation, which facilitates to manage coastal resources and implement ecological conservation effectively of China"s coasts.