This report investigates the potential of applying salt water modelling using the Laser Induced Fluorescence (LIF) flow visualisation technique to study balcony spill plume phenomena. A 1120 scale perspex model was used to conduct a series of salt water experiments. The testing parameters include two balcony settings (125mm and 250mm in model scale) and two spilling densities (0.5% and 1.0% of salt by weight). Through the study, results showed that good flow visualisation could be achieved using the LIF technique. The main advantage of using the salt water modelling technique was the ease in modifying the model's geometry or testing conditions. These modifications would not result in huge changes in the data acquisition systems as in either the full or small scale fire tests. The smoke layer within the compartment was simulated by the injection of a saline layer. By doing this, a quantitatively correct counter flow at the doorway was achieved, as in the real fire situation. This method also provided greater control over the injected flow and the flow was found to be repeatable. From the results obtained, it was found that there was a small degree of entrainment at the rotational region of the balcony spill plume. It was also observed that the presence of the soffit at the opening would result in significant entrainment into the underbalcony flow layer. Smoke logging on the upper balcony was found to be more severe with a shorter balcony than a longer balcony; this was due to the local deepening effect. The salt water results collected in this study could not be converted into the equivalent fire results at this stage. There are still many questions regarding the issue of scaling salt water results to full scale fire cases. Future research on the scaling laws needs to be done before the full potential of the salt water modelling technique could be utilised.