Abstract: Under the influence of climate change and human activities，the saltwater intrusion of Modaomen Waterway in the Pearl River Estuary has become increasingly severe，posing a serious threat to the safety of water supply in Macao，Zhuhai and Zhongshan. Based on the data of the dry season from 2021 to 2022，the prediction model of the daily mean chlorinity and the daily exceeding time of saltwater intrusion in the Pinggang pump Station of Modaomen Waterway is established by using multivariate nonlinear regression method. The results show that the daily mean chlorinity of Pinggang Station is negatively correlated with the daily maximum tidal range of Sanzao Station and the river discharge of Wuzhou and Shijiao Station，and the lag time is 3 days. Considering the cumulative effect of tide，runoff and early salinity，the goodness of fit r² of the average daily mean chlorinity at Pinggang Station is 0.85. The relationship between the daily exceedance time and the daily mean chlorinity show the relationship of power function，and the goodness of fit r² is 0.93. The prediction model has high accuracy and can provide scientific reference for regional saltwater intrusion prevention and control.