Using synchronous observational water temperature and salinity data collected simultaneously by 21 ships in summer and a three-dimensional robust diagnostic model, we calculated the density-driven current in Jiaozhou Bay (JZB), a semienclosed bay in the Yellow Sea. Special attention was paid to the influences of intratidal variations in temperature and salinity on the density-driven current. The density-driven current in JZB has a maximum speed of similar to 0.1 m s(-1) and is stronger than the tide-induced residual current in some places. The density-driven current is characterized by the intrusion of high-density (low-density) water in deep (shallow) areas. The results of the diagnostic model depend heavily on the observational data. For example, the density-driven current calculated from nonsynchronous data obtained by one ship at the same 21 stations is not consistent with that calculated from synchronous data because the nonsynchronous data correspond to different tidal phases at different stations. The intratidal variations of the density field result in a false spatial variation of density in the nonsynchronous data, which induces a false density-driven current that is of the same order as that calculated from the synchronous data. In contrast, the tidally averaged water temperature and salinity, which were used to remove intratidal variations from the synchronous data, diagnosed a density-driven current consistent with that from synchronous data. We, therefore, conclude that it is not necessary to explicitly resolve the intratidal variations in density in the calculation of density-driven current, but it is necessary to remove intratidal variations in the density field before the calculation.