ALBERT

Description: At the southern Hikurangi margin, New Zealand, we use salt marsh stratigraphy, sedimentology, micropaleontology, and radiocarbon dating to document evidence of two earthquakes producing coseismic subsidence and (in one case) a tsunami over the past 1000 yrs. The earthquake at 520–470 yrs before present (B.P.) produced 0.25±0.1 m of subsidence at Big Lagoon. The earthquake at 880–800 yrs B.P. produced 0.45±0.1 m of subsidence at Big Lagoon and was accompanied by a tsunami that inundated ≥360 m inland with a probable height of ≥3.3 m. Distinguishing the effects of upper plate faulting from plate interface earthquakes is a significant challenge at this margin. We use correlation with regional upper plate paleoearthquake chronologies and elastic dislocation modeling to determine that the most likely cause of the subsidence and tsunami events is subduction interface rupture, although the older event may have been a synchronous subduction interface and upper plate fault rupture. The southern Hikurangi margin has had no significant ( M 〉6.5) documented subduction interface earthquakes in historic times, and previous assumptions that this margin segment is prone to rupture in large to great earthquakes were based on seismic and geodetic evidence of strong contemporary plate coupling. This is the first geologic evidence to confirm that the southern Hikurangi margin ruptures in large earthquakes. The relatively short-time interval between the two subduction earthquakes (~350 yrs) is shorter than in current seismic-hazard models. Online Material: Historical accounts, description of vertical deformation, core names, foraminifera census and abundance, diatom census, modern analog samples, map of cores collected, stratigraphic correlation diagram for all cores, and detailed core logs.

Description: The Alpine fault in south Westland, New Zealand, releases strains of Pacific–Australian relative plate motion in large earthquakes with an average interevent spacing of ~330 years. A new record of earthquake recurrence has been developed at Hokuri Creek, with evidence for 22 events. The youngest Hokuri Creek earthquake overlaps in time and is believed to be the same as the oldest of another site about 100 km to the northwest near Haast. The combined record spans the last 7900 years and includes 24 events. We study the recurrence rate and conditional probability of ground ruptures from this record using a new likelihood-based approach for estimation of recurrence model parameters. Paleoseismic parameter estimation includes both dating and natural recurrence uncertainties. Lognormal and Brownian passage time (BPT) models are considered. The likelihood surface has distribution location and width parameters as axes, the mean and standard deviation of the log recurrence for the lognormal, and the mean and coefficient of variation for the BPT. The maximum-likelihood (ML) point gives the parameters most likely to have given rise to the data. The ML point, 50-year conditional probabilities of a ground-rupturing earthquake are 26.8% and 26.1% for the lognormal and BPT models, respectively. Contours of equal likelihood track the parameter pairs that are equally probable to have given rise to the observed data. Conditional probabilities on the lognormal 95% boundary around the ML point range from 18.2% to 35.8%. An empirical distribution model completely based on past recurrence times gives a similar conditional probability of 27.1% (9.6%–50.2%). In contrast, the time-independent conditional probability estimate of 13.6% (8.8%–19.1%) is about half that of the time-dependent models. A nonparametric test of earthquake recurrence at Hokuri Creek indicates that time-dependent recurrence models best represent the southern Alpine fault of the South Island, New Zealand.

Description: Most paleoenvironmental assessments of fossil salt-marsh foraminiferal faunas are based on modern analogue samples from the surface 1–2 cm. Usually no account is taken of the faunal modifications that result from the 60–90% of the fauna that live at greater infaunal depths or from the patchy loss of agglutinated tests that mostly occurs in the oxic and taphonomically active zone (TAZ, upper 10–15 cm). Here we provide examples of the highly variable foraminiferal test distribution in New Zealand salt marsh cores and surface transects. Using these, we suggest two simple adjustments that could be made to quantitative estimates of paleoelevation derived from fossil salt marsh faunas based on modern surface analogues: 1) Determine a minimum dead specimen density (8 per cm 3 in this study) below which all fossil faunas are considered to have suffered significant taphonomic loss (in the TAZ) and their paleoelevation estimates are deemed suspect and may be applied tentatively to core depths ~15+ cm above. 2) Move the core depths of all acceptable paleoelevation estimates upwards by the mean infaunal depth of the dominant species (2.5–8 cm in this study). We applied these adjustments to two late Holocene cores located 5 m apart in salt marsh in tidal Big Lagoon, Marlborough, New Zealand. Fifty percent (18) of the fossil faunas have unacceptably low specimen densities (〈8 per cm 3 ). A 15 cm thick fine pebbly sand bed contains a mix of calcareous tests derived from subtidally offshore and from intertidally within the adjacent sheltered lagoon, and agglutinated tests of salt marsh taxa inferred to be displaced and/or infaunally in-situ. These faunas provide strong evidence for sand deposition on top of the high salt marsh by the outgoing surge of a tsunami. The upper 7 and 13 cm of the sand lacks calcareous tests, which are inferred to have been dissolved by acidic pore waters following colonisation by salt marsh vegetation. Our taphonomically- and infaunally-adjusted paleoelevation estimates, generated by modern analogue technique using a training set of 1017 modern New Zealand faunas, provide near-identical elevational histories for both cores. Both core sequences (70 and 85 cm thick) were deposited within a 50 cm-wide elevational envelope in middle- and high-tidal salt marsh. Both record an inferred 0.5 m co-seismic subsidence event coincident with emplacement of the tsunami sand (~840 cal yrs BP) and a second smaller, ~0.3 m, presumably co-seismic, subsidence event coincident with a sharp peat-mud contact in one core (~500 cal yrs BP).

Description: Stratigraphic investigations of three coastal waterbodies in southeastern Tasmania reveal major paleoenvironmental phases related to sea level change and anomalous deposits consistent with tsunami inundation. Twenty-two short sediment cores were examined for their sedimentology and fossil diatom, foraminifera and macrofossil assemblages; nine radiocarbon ages were obtained. Despite diverse Holocene histories at each site, four common phases of Holocene paleoenvironmental evolution can be distinguished. In Phase I (pre-8000 yr BP) terrestrial environments existed. During Phase II (8000–6500 yr BP) ponded freshwater environments formed behind transgressive coastal barriers. In Phase III (6500–2000 yr BP) the sites were subject to varying degrees of marine influence, resulting in environments ranging from current-swept tidal inlets to sheltered brackish-marine lagoons. In Phase IV (2000 yr BP to present) there was a decrease in marine influence, one site changed to a freshwater wetland environment while the other two changed to ephemeral salt pans. This study suggests that postglacial sea level rise culminated after c . 7300 cal. yr BP in southeastern Tasmania and that there was probably a late-Holocene fall in sea level. These paleoenvironmental histories provide a framework within which to identify anomalous deposits and assess them for likely causes. Five anomalous deposits are identified, three of which are considered likely to have been deposited by tsunami occurring at c . 4000 cal. yr BP, c . 2000 cal. yr BP and 〈2000 cal. yr BP, although deposition by large storms cannot be ruled out.

Description: Sudden changes in microfossils and lithologies in Holocene sediments of a former tidal inlet on the Hikurangi subduction margin provide evidence of 10 large earthquakes. Studies were focused in three former embayments where intertidal shelly sediment interfingers with freshwater and salt-marsh peat. Paleoelevation histories were reconstructed using the modern analogue technique with foraminiferal assemblages. Land elevation record analysis indicates 8–9 m of mid- to late Holocene tectonic subsidence occurred prior to 1.5 m of uplift during the A.D. 1931 Hawkes Bay earthquake. Chronologies of displacement events were constrained using 50 radiocarbon dates and three widespread air-fall tephras. We infer the following earthquakes: earthquake 1: 7.3–7.0 ka (–1.1 ± 0.3 m), earthquake 2: 5.6–5.1 ka (+0.4 ± 0.4 m), earthquake 3: 5.2–4.9 ka (–0.5 ± 0.5 m), earthquake 4: 4.4–3.8 ka (–0.6 ± 0.5 m), earthquake 5: 2.8–2.4 ka (–0.9 ± 0.5 m), earthquake 6: 1.73–1.70 ka (–1.0 ± 0.3 m), earthquake 7: 1.5–1.3 ka (–0.7 ± 0.5 m), earthquake 8: 1.04–0.89 ka (–1.2 ± 0.4 m), earthquake 9: 0.60–0.44 ka (–0.8 ± 0.6 m), and earthquake 10: A.D. 1931 (+1.5 ± 0.3 m). A further 1.6–2.6 m of subsidence could have occurred by gradual aseismic slip or in smaller earthquakes. The age ranges of four of the recognized earthquakes (earthquakes 1, 6, 8, and 9) overlap with other documented displacement events onshore along 250–600 km of the Hikurangi subduction margin, and with turbidites offshore 100–300 km to the north. These four are considered strong candidates for large subduction-interface earthquakes. The other five inferred earthquakes are less strongly correlated with along-margin displacement events and offshore turbidites. These could have been caused by upper-plate fault ruptures (like historic earthquake 10), but subduction-interface sources cannot be ruled out. This evidence for repeated coseismic vertical deformation suggests large coseismic slip on a part of the subduction interface beneath Hawkes Bay that is currently dominated by aseismic creep processes, such as transient slow-slip events. This clearly indicates multiple slip processes are possible in a single location on a subduction interface.

Description: A sedimentary sequence that was highly sensitive to fault rupture–driven changes in water level and sediment supply has been used to extract a continuous record of 22 large earthquakes on the Alpine fault, the fastest-slipping fault in New Zealand. At Hokuri Creek, in South Westland, an 18 m thickness of Holocene sediments accumulated against the Alpine fault scarp from ca. A.D. 800 to 6000 B.C. We used geomorphological mapping, sedimentology, and paleoenvironmental reconstruction to investigate the relationship between these sediments and Alpine fault rupture. We found that repeated fault rupture is the most convincing mechanism for explaining all the features of the alternating peat and silt sedimentary sequence. Climate has contributed to sedimentation but is unlikely to be the driver of these cyclical changes in sediment type and paleoenvironment. Other nontectonic causes for the sedimentary alternations do not produce the incremental increase in basin accommodation space necessary to maintain the shallow-water environment for 6800 yr. Our detailed documentation of this near-fault sedimentary basin sequence highlights the advantages of extracting paleoearthquake records from such sites—the continuity of sedimentation, abundance of dateable material, and pristine preservation of older events.

Description: A database of historical (preinstrumental) and modern (instrumentally recorded) tsunamis that have impacted or been observed in New Zealand has been compiled and published online. New Zealand’s tectonic setting, astride an obliquely convergent tectonic boundary on the Pacific Rim, means that it is vulnerable to local, regional, and circum-Pacific tsunamis. Despite New Zealand’s comparatively short written historical record of about 200 years, there is a wealth of information about the impact of past tsunamis. The New Zealand Tsunami Database (NZTD) currently has 800+ entries that describe 〉50 high-validity tsunamis. Sources of historical information include witness reports recorded in diaries, notes, newspapers, books, and photographs. Information on recent events comes from tide gauges and other instrumental recordings such as Deep-ocean Assessment and Reporting of Tsunamis (DART) buoys, and media of greater variety, for example, video and online surveys. The NZTD is an ongoing project with information added as further historical records come to light. Modern tsunamis are also added to the database once the relevant data for an event have been collated and edited. This article briefly overviews the procedures and tools used in the recording and analysis of New Zealand’s historical tsunamis, with emphasis on database content.