The propagation of flood inundation in urban catchments is highly dependent on the surface topography, land cover and the representation of structure in elevation data used in modelling. Studies have shown that subtle variations in terrain have a considerable influence on flood propagation patterns in urban watersheds. High resolution LIDAR DEM and DSM datasets are applied in a hydraulic model to determine floodplain inundation in this study. Datasets of Land cover, soil type and precipitation are used for model parameterisation. A comparison of results from both elevation data types was done to assess the impact of the presence of structures on the flood regime. Results from both models show variation in flooded locations amounting to about 17% difference in flooded areas of inhabited environments. Simulation using the DSM resulted in notable flood diversions in the urban areas with a general increase in flood depth and coverage due to its structure representation. Flood characteristics of depth and velocity are used as exposure indicators in mapping the vulnerability of properties to flood hazards. The two measures show significant variation in the risk levels associated to property units, yet the same properties are affected using both. A portable android application was prototyped for disseminating time-series flood model information and for reporting details of flood events as they occur to serve for model calibration and enhancement, thereby completing the flood modelling lifecycle.