Gisborne seismic and tsunami hazard: constraints from marine terraces at Puatai Beach
Authors: N J Litchfield, U A Cochran, K R Berryman, K C Clark, B G McFadgen, R Steele, GNS Science
Paper number: 3797 (EQC 14/666)
Abstract
The East Coast, Gisborne District, is at risk from large earthquakes on active faults lying offshore to the east. These earthquakes are also a likely trigger for tsunamis. Earthquakes on offshore faults can raise the coastline, creating uplifted beaches, and so we can use these landforms to provide a record of past earthquakes and tsunamis. In this study we obtain much greater insight into raised beach records than previous studies by excavating a 90 m long trench across three raised beaches at Puatai Beach. The beach is approximately 30 km north of Gisborne. The trench provided information on the timing, magnitude and frequency of past earthquakes on the Gable End Fault. This fault lies about 5 km offshore and runs from just north of Gisborne to Tolaga Bay. The trench also provided information on the height and frequency of past tsunamis that have struck the East Coast.
Investigation of the trench confirmed that there are three raised beaches at Puatai Beach. The beaches were raised in three earthquakes, and were then covered by slopewash, most likely from slips and washed down onto the beaches during storms. The earthquakes occurred approximately 1800, 1200, and 400 years ago and each raised the Puatai Beach coast by around 3.5 metres. These results confirm previous calculations that the Gable End Fault ruptures roughly every 750 years, producing earthquakes of around magnitude 7.2. However, the average amount (3–4 m) and rate of uplift is higher than the long-term average, and may suggest that the involvement of other faults such as the Waihau Fault. The time since the last uplift event (400 years) is shorter than the average repeat times (750 years), suggesting the next earthquake is likely not imminent.
Within the slopewash sediments on each raised beach are thin sandy layers, many containing shells. We interpret these to be storm or tsunami deposits, with most probably of tsunami origin. Two of the tsunamis occurred approximately 1000 and 300 years ago. These ages are slightly younger than the earthquakes which raised two of the beaches and within slopewash, so the tsunamis were probably not triggered by the Gable End Fault. There is evidence for tsunamis at other sites along the North Island East Coast, as well as raised beaches and submarine landslides which occurred at approximately 1000 and 300 years ago. Further work could pinpoint the sources of these tsunamis. The highest tsunami sand layers were deposited 9–12 m above mean sea level. These heights that are within, but at the upper end of, the expected tsunami heights for this coastline as modelled in the National Tsunami Hazard Model.
In addition to the data on earthquakes and tsunami, the trenches also included archaeological remains. These were mainly fireplaces, probably from prehistoric transient Māori occupation, not established settlements. It is possible that Māori witnessed the last earthquake around 400 years ago and tsunami around 300 years ago.
Technical Abstract
The Gisborne District has a high seismic and tsunami hazard owing to its position in the northern Hikurangi Margin. In this study we place constraints on the seismic and tsunami hazard through a study of marine terraces at Puatai Beach, on the East Coast ~30 km north of Gisborne, and ~5 km west of the landward-dipping, Gable End Fault. Previous reconnaissance work at Puatai Beach documented evidence for three anomalously high and young marine terraces, which led us to speculate that some of the ages were inadvertently collected from tsunami deposits draped over the marine terraces. This study gives a more detailed analysis of the terraces and potential tsunami deposits, through excavation of a
~90-m-long continuous trench across the three terraces. The analysis addresses the aims of: 1) better constraining: the recurrence interval, single event displacement, and timing of the most recent earthquake for the Gable End Fault; and 2) if tsunami deposits are present, developing a paleotsunami record.
The Puatai Beach trench confirmed the presence of three marine terraces, characterised by stepped shore platforms and fossil sea cliffs cut into bedrock, and overlain by shelly sand and gravel beach deposits and silty colluvium. The surveyed inner edges of the shore platforms (shoreline angles) are 9.5 ± 0.5, 6.6 ± 0.5, and 3.0 ± 0.5 m amsl, from which an average uplift-per-event of 3.4 ± 1 m can be calculated. The mean estimate is slightly higher than the mean single event displacement (2.9 m) calculated for the Gable End Fault in the National Seismic Hazard Model. Radiocarbon ages from the youngest shells in the beach deposits on each terrace show the terrace ages are 1920–1650 (upper), 1270–1030 (middle), and 520–320 (lower) cal. yr BP respectively, which we interpret to be the timing of large (M7+) paleoearthquakes on the Gable End Fault. From these, are calculated inter-event times of 890–380 and 950–510 years, which are similar to the calculated recurrence interval on the Gable End Fault in the National Seismic Hazard Model (760 years). These results confirm that the terraces are relatively high and young, and indicate an uplift rate of 5.6 ± 1 mm/yr. This is lower than previously reported using terrace surface heights 7–8 mm/yr), but is higher than the vertical slip rate for the Gable End Fault (3.8 ± 0.5 mm/yr) and may suggest an additional fault, such as the nearby Waihau Fault, is involved in the terrace uplift mechanism. The youngest terrace age (520–320 cal. yr BP) provides the first estimate for the most recent event on the Gable End Fault and could have been witnessed by early Maori. We also note that the elapsed time is less than the average recurrence interval.
The age for the most recent uplift event falls within the period of Māori occupation of the coast, and the event would almost certainly have been witnessed by early Māori living in the area. The trenches also included archaeological remains. These were mainly fireplaces, probably from prehistoric transient Māori occupation, not established settlements. No other early occupation remains were seen at Puatai Beach. Whether this is because Māori recognised the dangers of living too close to the beach within range of tsunamis, and hence lived higher up the hills, is a subject for future research into the Māori stories and traditions of the area and into the archaeology.
The silty colluvium layers overlying the beach deposits contain thin semi-continuous sand layers. Based on chronological, geomorphological, sedimentological and biological considerations we suggest that at least some of these sand layers are tsunami deposits. Three sand layers are dated at 1190–930, 400–100, and 450–150 cal. yr BP, and the overlap of the latter two suggests they could be the same event. The inferred tsunami ages are slightly younger than the uplift of the marine terraces, and this as well as their position within colluvium, suggests they were not necessarily triggered by rupture of the Gable End Fault. The younger ages potentially overlap tsunami deposits from nearby sites, and there are correlative paleotsunami deposits, uplift events, and turbidites from elsewhere along the Hikurangi Margin, suggesting potential sources that could be investigated with further research. Estimates of tsunami run-up can be obtained from the surveyed maximum heights, and making allowance for the terrace uplift, they are 9.3 ± 0.5 m, 12.6 ± 0.5 m and 4.2–1.2 ± 0.5 m amsl, for the two dated and one un-dated paleotsunamis respectively. These heights are at the upper end of height estimates for tsunamis in Waihau Bay in the National Tsunami Hazard Model, and this research confirms that tsunamis of this height are possible along the Gisborne coast.
Order a research paper
Many of these research papers have PDF downloads available on the site.
If you'd like to access a paper that doesn't have a download, get in touch to ask for a copy.