Power, W.L.; Fry, B.; Gusman, A.R.; Burbidge, D.R.; Brewer, M.; Wang, X. 2023 Preliminary DART buoy network design. Lower Hutt, NZ: GNS Science. GNS Science report 2023/26. 49 p.; doi: 10.21420/0G31-AT33
Abstract:
New Zealand’s location on the Pacific ‘Ring of Fire’ makes us and our Pacific Island neighbours vulnerable to tsunami, one of the most dangerous natural hazards. The risk posed by tsunami to coastal populations can be mitigated by effective tsunami warning systems applied in conjunction with public education. DART buoys are an essential component of tsunami detection and impact forecasting systems. Recent international forums have confirmed DART buoys are the only currently available technology capable of providing timely and accurate calibration of tsunami forecast models. This conclusion has been strengthened by the availability of fourth generation DART buoys, which can be located much closer to potential tsunami sources, allowing for faster detection and forecasting. New Zealand has made extensive and effective use of earlier-generation DART buoys operated by the United States of America and Chile during responses todistant (trans-Pacific) tsunami threats. Recent experience during exercises and events has demonstrated the difficulties of providing well-calibrated tsunami threat forecasts for local and regional tsunami sources without the availability of DART data. The compelling case for the deployment of a DART buoy network for New Zealand comes from investigation of the tsunami threats facing our communities from earthquakes along the Hikurangi and Kermadec Trenches. Earthquakes in the Kermadec region are poorly felt in north-western New Zealand, including along the coast from the Bay of Plenty to Northland, which includes cities such as Tauranga, Auckland and Whanga;rei. Plausible scenarios indicate the possibility of waves of over 5 m impacting New Zealand coastlines following ground motions too weak to trigger widespread natural warning evacuations. Similarly, there have been two historically recorded earthquakes in 1947 that were weakly felt on land but caused dangerous tsunami. Offshore tsunamigenic earthquakes are also difficult to seismically characterise with the terrestrial GeoNet network. We conducted a scientific evaluation of the tsunami sources that threaten New Zealand and our southwest Pacific neighbours. Using the evaluation, we designed a network of DART buoys for the region. Tsunami source regions to be monitored were decided through a series of expert workshops, and DART buoy locations were selected based on a trade-off between quick detection and the need for accurate forecasts and reliable buoy operation. The highest priority tsunami source region was the Kermadec Trench, a region where the Pacific tectonic plate is being subducted below the Australian plate, which extends from the East Cape of the North Island for more than 1000 km in the direction of Tonga. This region was assessed to pose a high risk of life-threatening tsunami to the northern half of the North Island, yet attenuation of seismic waves by the offshore extension of the Taupo; Volcanic Zone, and the distance and orientation of the trench relative to the North Island coastline, means that, in many scenarios, the earthquake shaking is unlikely to be sufficiently strongly felt to prompt widespread self-evacuation of the most at-risk coasts. The Kermadec Trench is also very difficult to monitor by other means (e.g. rapid seismic analysis), and the travel time to the most-affected coasts indicate that the availability of DART buoy data is essential for effective early warning. Other high-priority tsunami source regions were the Hikurangi Margin (which sits along the east coast of the North Island), the Tonga Trench (east of the Tongan Island chain), the southern New Hebrides Trench (east of New Caledonia) and the Puysegur Trench (south of Fiordland). The Puysegur Trench would have been higher in the priority list except for two Australian DART buoys that are already located in that area. The DART buoy network design is based on enabling the detection of a tsunami within 20 minutes for the major tsunami sources close to New Zealand and within 30 minutes for other regions of the southwest Pacific not already monitored by Australian buoys. DART buoy locations were selected based on a trade-off between quick detection and the need for accurate forecasts and reliable buoy operation. The full form of the proposed network consists of 12 buoys, with an optional 13th buoy near Fiordland that could be included to reduce the time to detect and forecast tsunami generated on the Puysegur Trench by placing it closer to the trench than the earlier-generation Australian DART buoys. In the event that the full network cannot be funded, we also proposed two sub-network designs. One is an eight-buoy network (nine if the optional Fiordland buoy is included) focused on the specific requirements for accurate tsunami threat-level forecasting for New Zealand. This sub-network still provides some benefits to other Pacific Island states, although the detection times are noticeably increased for the source regions further from New Zealand. The other sub-network design involves seven buoys and is focused on providing timely information for Pacific Island states (auth)
