Abstract: The Yangtze River estuary has been strongly influenced by human activities, and the consistency of its runoff-sediment sequence has been damaged. Based on the monthly runoff and sediment data from 1965 to 2019 at Datong Station, a sliding window algorithm based on Copula is applied to investigate the variation characteristics of runoff and sediment in the Yangtze estuary and the consistency of its abundance-depletion, and the uncertainty of the marginal distribution selection for the joint runoff-sediment distribution model is also analyzed. The results show that: (1) The runoff-sediment combination at Datong Station shows a decreasing trend from 1965 to 2019 due to human activities, and changes abruptly in 1979 and 2000. The optimal function models for the runoff-sediment series in 1965—1979 and 1980—2000 are Clayton, while 2001—2019 is best modeled by Frank; (2) the synchronous frequencies of the runoff-sediment series in the three periods are 84.52%, 84.40% and 83.20% respectively, which are much greater than the asynchronous frequencies of abundance-depletion, revealing the strong consistency of the incoming runoff-sediment in the upper reaches of the Yangtze River; (3) the 95% confidence interval of the PE3-PE3 marginal distribution combination is smaller than that of the PE3-GPD combination, indicating that the PE3-PE3 combination would reduce the uncertainty of function selection. By exploring the joint variation characteristics of incoming runoff and sediment at Datong station, this study provides a reference for water resources management and river regulation in the Yangtze estuary.