A 4D seismic or time lapse survey has been used to investigate the amplitude versus offset (AVO) effects on seismic data in order to identify anomalies in the Gullfaks field for three different reservoir intervals namely the Tarbert, Cook and Statfjord reservoirs. Repeatability analysis has shown that the earlier seismic vintages are the most unreliable for amplitude anomaly analysis as normalised root-mean square (NRMS) values are greater than 50%. This is above the threshold of good and medium repeatability. Fluid substitution models show increases in both P-wave velocity and density for increasing water saturations with a maximum change of 7.33% in the Pwave velocity, and this is in line with predictions from previous work using the Biot -Gassman equations. AVO modelling for the top Tarbert Formation interface produced scenarios of increasing amplitudes with offset for the presence of hydrocarbons, which dim out with 100% brine saturation. This correlates to class III gas sands for different situations of varying Poisson's ratio across an interface, which has been previously modelled. Two anomalies were identified with one being related to increasing pressure due to water injection correlating to poor permeability around injector well 34/10-B-33. The second anomaly is a case of potential unswept hydrocarbons that displayed a consistent bright spot throughout all of the seismic vintages (in-inlines and crosslines). AVO attribute analysis of this event produced a class II anomaly. However, when comparing near and far offset seismic data, dimming effect was observed producing contrasting evidence. The dimming offset is viewed to have been as a result of poor repeatability values at far offsets. The modelling of the fluid contents in the studied formations to conform to existing literatures justifies the efficacy of the method.