Power, W.L.; Henderson, A.; Lukovic, B. 2024 Agent-based tsunami evacuation modelling of Papamoa. Lower Hutt, NZ: GNS Science. GNS Science report 2024/18. 33 p.; doi: 10.21420/7JE4-NW55
Abstract
Papamoa is a suburb of Tauranga in the Bay of Plenty which has undergone remarkable population growth in recent years. As a low-lying coastal town, it is vulnerable to tsunamis, and in particular to tsunamis caused by earthquakes on the Kermadec subduction zone which lies to the north-east of the Bay of Plenty. Tsunamis from the southern part of the Kermadec subduction zone are expected to reach the Papamoa coastline around 50 minutes after the earthquake. It is consequently important that the population are able to promptly respond to Long or Strong shaking in the event of such an earthquake by self-evacuation from the tsunami evacuation zone. In this study we have used Agent Based Modelling (ABM) as a tool to evaluate the potential for the population of Papamoa to be able to leave the evacuation zone on foot within the time available, and to try and identify obstacles that could prevent a successful evacuation and identify potential improvements that could assist. Agent Based Modelling is a technique in which the movement of ‘agents’, who are simulated individual people, is calculated by numerical modelling. In this case we model agents as pedestrians attempting to leave their homes or workplaces and reach the safe area outside of the evacuation zone. One benefit of the agent-based approach is that it provides a way to identify ‘bottlenecks’ and other areas of congestion that may occur during an evacuation. The modelling approximates a very complicated process, so cannot be completely accurate. But it is hoped that it can still provide useful insights. Among the key approximations here are that only pedestrians are modelled, whereas in reality it can be expected that some people will attempt to use vehicles, and that the streets and paths are usable as usual, i.e. not affected by debris or liquefaction. One clear insight from the study is the importance of people making the decision to evacuate and leaving their homes or workplaces promptly. The evacuation is much less effective if a proportion of people do not respond to the earthquake by evacuating promptly, as may be anticipated to occur based on a recent survey of evacuation intentions of people living on the east coast of the North Island. In eastern/central Papamoa we find many agents use routes that cross farmland to leave the evacuation zone in the vicinity of the Bruce Road services. Passage across these fields may be relatively difficult because of the rough ground, and the fields are not available at certain times of year. We recommend that the potential for additional routes to reach safety in this area avoiding the farmland be investigated. Vertical Evacuation Structures (VES) can provide an alternative refuge from tsunamis for those who will struggle to reach safety in time by leaving the evacuation zone entirely. These are structures in which people are elevated above the anticipated tsunami water level. The Gordon Spratt evacuation mound is the only purpose-built VES in New Zealand. We have modelled the number of people using the evacuation mound and the similarly isolated area around the Te Manawa school. However there is variation in public opinion around the use of mounds, which is very hard to estimate. Future work to survey public attitudes would be valuable here. There are several drainage channels in Papamoa, some of which run approximately parallel to the coast, and which are not safely crossable by pedestrians except where there are bridges. In the study we considered one possible additional bridge location, which was helpful in terms of increased safety during the evacuation, and we recommend careful consideration of other crossing point options. (auths)