Abstract A study of the dynamic adsorption of an anionic surfactant, Neodol® 25-3S, on Berea sandstone is presented. Cylindrical cores with axisymmetric radial flow are used in order to simulate the flow pattern near an oilfield injection wellbore, and also to allow greater sensitivity in testing various transport and adsorption mechanism. Mathematical models based on the advective-dispersive transport equation of mass coupled with an adsorption model were applied. Numerical solutions of the coupled transport and adsorption equations were obtained using the Galerkin finite element method. Equilibrium and kinetic Langmuir, two-site, and bilayer adsorption models were evaluated. Linear (Henry's law) and non-linear (Langmuir) equilibrium adsorption models were found to be unsuitable for modelling the surfactant effluent profiles. Improved fits to the experimental data were obtained using a kinetic Langmuir model, but the total amount of adsorption was over-estimated by a factor of three. It was found that the two-site or bilayer model was required to produce consistent model parameters for the experimental conditions employed. The effect of salinity on the adsorption of surfactant was examined using the two-site model. A small decrease in the rate constant for the first type of adsorption kinetics and an increase for the second type were observed. At high salinity, the axisymmetric radial transport equation was not able to predict the tracer and surfactant profiles during the brine flood. At such conditions, the mobility ratio was not favourable and, as the result, the displacement front was not stable. The flow pattern became two-dimensional and could not be described by the one-dimensional model used. An increase in the rate of adsorption was observed with increasing injection flow rate. It is proposed that the adsorption kinetics depend strongly on the concentration of surfactant monomers in the bulk solution. The results from this study can be used to estimate surfactant loss due to adsorption during many surfactant-based EOR processes. The surfactant loss in the wellbore region is a very important parameter for determining surfactant propagation in the reservoir.