Salinity is an important parameter influencing the water quality of estuaries. Prediction of salt intrusion length in estuaries is a challenge for managers as well as scientists in this field. Several numerical and empirical models have been developed for the prediction of salt intrusion length in recent decades. The aim of this study is to evaluate the performance of empirical models in the Arvand river estuary in south-west Iran. A laterally averaged, two-dimensional hydrodynamics–water quality model called CE-QUAL-W2 was used for numerical simulations. Salt intrusion lengths under different discharge and tidal conditions were first predicted by the validated numerical model, then the semi-empirical models were applied to the same hydrological/tidal conditions. From the comparison of the results, it was found that the model of Savenije, that properly considers the variation of bathymetry along the estuary, outperformed the others and could be used as a rapid assessment tool in the Arvand river estuary.