跳到主要导航 跳到搜索 跳到主要内容

Image-based machine learning for monitoring the dynamics of the largest salt marsh in the Yangtze River Delta

  • East China Normal University
  • Qingdao National Laboratory for Marine Science and Technology
  • the People's Republic of China

科研成果: 期刊稿件文章同行评审

摘要

The extreme decline in fluvial sediment discharge and rapid increase in sea level have increased salt marsh vulnerability in some of the world's mega-delta. However, limited research has addressed both the vertical accretion and horizontal/lateral progradation of salt marshes induced by anthropogenic activities in recent decades. Here, a machine learning-based method for retrieving remote sensing images of the salt marsh along the Eastern Chongming Wetland (ECW), the largest wetland in the Yangtze River Delta, was used to monitor salt marsh dynamics between 2002 and 2019. The results demonstrate that salt marshes have experienced significant expansion, including seaward progradation and accretion with ranges of −18.5–60.6 m/yr and 0.103–0.178 m/yr, respectively. Nevertheless, the bare mudflat areas adjoining the salt marshes have remained almost unchanged, while their progradation and accretion have also shown similar trends with the ranges of −13.3–103.7 m/yr, and 0.066–0.256 m/yr, respectively. Although there was a 70% reduction in fluvial sediment supply in the Yangtze River Delta after the Three Gorges Dam (TGD) began operating in 2003, it is less understood if the constant local suspended sediment concentration (SSC) of the estuary could be responsible for supporting enough sediment to enable salt marsh and mudflat expansions. Meanwhile, the results showed that the seaward expansion of the mudflats provided suitable space for the salt marsh to trap vast amounts of sediment and gradually occupy the adjoining mudflat area. The mudflat progradation further provided a larger space for the growth of salt marsh vegetation and promoted salt marsh expansion. Moreover, the accretion of the ECW indicates the high resilience of these salt marshes to sea-level rise (SLR). The present work highlights the external factors and internal driving forces of the salt marsh evolution process, providing information that can be used by communities and coastal managers to conserve and restore the salt marshes in the future.

源语言英语
文章编号127681
期刊Journal of Hydrology
608
DOI
出版状态已出版 - 5月 2022

学术指纹

探究 'Image-based machine learning for monitoring the dynamics of the largest salt marsh in the Yangtze River Delta' 的科研主题。它们共同构成独一无二的学术指纹。

引用此