【摘要】 文章采用面向对象图像分割与随机森林分类相结合的方法，基于Landsat和Sentinel-2卫星影像提取了2011年、2016年、2021年江苏滨海湿地生态系统的空间分布信息，分析了湿地植被群落的时空动态变化及湿地植被变化成因。结果显示：基于Sentinel-2卫星影像的分类精度较高；江苏滨海湿地植被群落主要集中分布在新洋港与斗龙港之间，少部分沿海堤呈条带状分布，植被类型主要为互花米草、碱蓬和芦苇；2011—2021年江苏滨海湿地植被呈先减少后增加的状态，湿地开垦、城市化进程等人类活动是驱使面积缩减的原因，同时互花米草的扩张及湿地保护修复等政策的实施，使湿地面积不断增加。更多还原

【Abstract】 In the paper, using a combination of object-oriented image segmentation and random forest classification, the spatial distribution information of Jiangsu coastal wetland ecosystems in 2011, 2016 and 2021 was extracted based on Landsat and Sentinel-2 satellite images, and the spatiotemporal dynamic changes of the wetland vegetation communities and the causes of wetland vegetation changes were analyzed. The results show that the classification accuracy based on Sentinel-2 satellite images is high. The vegetation community of Jiangsu coastal wetland is mainly distributed between Xinyang Port and Doulong Port, with a few coastal embankments distributed in strips. The vegetation types are mainly Spartina alterniflora, Suaeda salsa, and Phragmites australis. From 2011 to 2021, the vegetation of Jiangsu coastal wetland shows the state of decreasing and then increasing. The human activities, such as the reclamation of wetland and the process of urbanization, are the reasons of the area shrinkage. At the same time, the expansion of Spartina alterniflora and the implementation of policies of wetland protection and restoration make the wetland area increase continuously.更多还原