In 2006, Waikato Regional Council (WRC) provided research funding for the Coastal Marine Group, Department of Earth and Ocean Sciences, University of Waikato, to undertake research on the impacts of tsunami inundation on Whitianga town and harbour. The main aims of the study are to: ● Identify potential tsunami sources; ● Establish an understanding of tsunami inundation impacts in Whitianga township and harbour – including the hydrodynamic processes and responses of Mercury Bay, Buffalo Bay, Whitianga Harbour and adjacent land to tsunami wave action; ● Develop tsunami inundation maps, showing depth and velocity (speed) of tsunami waves; and ● Provide sound evidence upon which to base community risk mitigation measures – including recommendations for evacuation planning, public education and awareness, protection of infrastructure, management of impacts to marine vessels and future land use planning. The numerical model used in this study is the 3DD Suite-Computational Marine and Freshwater Laboratory model (Black, 2001). The model has demonstrated the ability to accurately reproduce tsunami hydrodynamics during propagation and run up on both laboratory and real-world scales. There are three primary tsunami sources that could potentially affect Whitianga from the Kermadec Trench, and beyond the New Zealand continental shelf, being: 1. Mt Healy undersea volcano eruptions (15th Century event); 2. Large earthquakes along segments 1 and 2 of the Kermadec Trench subduction zone; and 3. A 1960 Chilean-type earthquake event. Each source is modelled, and the results that show the greatest risk and impacts to Whitianga are used as the basis for the hazard maps and hazard zones. Modelling results indicate that: • The Mt Healy type of eruption produced a minimal impact on Whitianga. The tsunami waves generated from this event did not inundate Whitianga. Despite this, strong currents of up to 2.5 m/s were generated inside Buffalo Bay and at the Whitianga Harbour inlet. •The Kermadec Trench earthquake scenarios with both positive and negative leading waves, as a result of a subduction fault dislocation along segments 1 and 2, have a significant impact on Whitianga. The waves inundate the coastal area up to 2.5 and 3 km inland for the subduction thrust fault and normal fault events respectively, and affect the entire area of Whitianga Harbour. The normal fault event that produces positive leading waves has more impact than the thrust fault event that produces negative leading waves. • The 1960 Chilean-type earthquake event produced tsunami waves that inundated Buffalo Beach Road and houses in Whitianga, as observed by eyewitnesses. Strong currents of up to 5 m/s are generated inside Buffalo Bay and the Whitianga Harbour inlet. The modelling indicates that: • Whitianga would be inundated five times by a Kermadec Trench earthquake scenario, and three times by a 1960 Chilean-type of tsunami • For the Kermadec Trench scenario (normal fault), the first waves penetrate Mercury Bay within 75–98 minutes after the fault rupture • Regardless of the tsunami source, it takes 11–18 minutes for waves to arrive at the Whitianga foreshore once they have entered Mercury Bay. The modelling indicates that the period between waves is 40 – 60 minutes, which is consistent with the 1960 Chilean event. The geometry of Buffalo Bay and Mercury Bay amplify the incoming tsunami waves, and the sea level inside the bay continues to oscillate, even after the sea level outside of Mercury Bay returns to normal. This situation is consistent with eyewitness accounts of the 1960 Chilean tsunami. Modelling also shows that strong currents are produced within Buffalo Bay and Whitianga Harbour, as well as during the overland flows - especially in areas adjacent to the Taputapuatea Stream and in the foreshore area between Albert Street and the wharf. The flow speed ranged from 1.5 m/s to 8 m/s for overland flows, and above 8 m/s within the entrance of Whitianga Harbour and in the middle of Buffalo Bay. For the first time in New Zealand, a combination of non-ground striking and ground striking LIDAR data was used in modelling tsunami inundation, which increased the accuracy of the modelling results considerably. Inundation flow behaviour and the effect of topography, as well as land use, can be analysed more accurately, and a more precise hazard map can be produced accordingly. Mitigation measures suggested to protect the Whitianga waterfront include a combination of enhanced coastal sand dunes and planted forest belts, which could be done along both sides of the Taputapuatea Stream. A stop gate could also be constructed at the entrance of the Taputapuatea Stream to minimise the impact of the tsunami flows upstream. With respect to evacuation, it is concluded that due to the lag time between a local event from the Kermadec Trench and wave arrival at the Whitianga foreshore, there is enough time for residents to be evacuated to shelter sites using major roads. Three locations are identified as evacuation shelter sites. These are the marina parking area and Buffalo Beach scenic reserve (both of which are located on high ground adjacent to the high-risk zone), and further inland at the airfield. Vertical evacuation sites are needed inside the high-risk zone, and recommendations on potential locations are provided. It is important that vertical evacuation measures are integrated into community response plans, and that they be reviewed and revised regularly. Overland flow information derived from modelling using the ground-striking and non-ground striking LIDAR data provides a basis to influence new development that occurs within tsunami hazard risk zones. Overland flow information also indicates the areas of existing development that need protection from future tsunami events. Risk mitigation may be accomplished through redevelopment, retrofit, coastal defence measures, safety planning for ships and boats, land reuse plans, and also via public education and awareness programmes. A major challenge of risk mitigation is to maintain emergency preparedness programmes and procedures when the threat of tsunami is perceived as remote. Periodic exercises are essential to maintain awareness, and regular information should be provided for those occupying tsunami hazard areas. Tsunami are rare events, but their impacts on coastal communities can be devastating. It is quite dangerous to believe that the impacts of a tsunami can be completely prevented by man-made structures (Horikawa and Shuto, 1983). However, possible impacts may be minimised through careful design of solutions based on systematic research. An important consideration for risk mitigation works is that they may affect the quality of daily life, and risk mitigation involves choices and trade-offs between risk management and other uses. Video animations of each scenario are provided at regional, intermediate and local scales. The animations cover tsunami wave behaviour during generation, propagation, run up, and overland flows, and may be used to inform land use planning and public education and awareness programmes.