The Yangtze River estuary is the outlet of the largest river in China,the changes and predicions of sediment content directly affect the ecological environment,waterway maintenance and flood control safety in the estuary area.This study develops a machine learning-based model for predicting sediment content dynamics in the Yangtze River estuary region.Considering that sediment content is affected by a variety of hydrological environmental factors,this study collects hydrological data for a year in the Yangtze River estuary area,including flow velocity,tide level,sediment content,etc.and uses time series analysis methods to extract key features and patterns,and a long short-term memory network(LSTM) is selected,trained and tested.The analysis results show that the LSTM-based model shows high accuracy in predicting sediment content in the Yangtze River estuary region.The mean absolute error of the model is 0.146 5,and the coefficient of determination is 0.931 4.