Repeating Earthquake Detection and Characterisation in New Zealand: A Catalogue for the Raukumara Peninsula, Northern Hikurangi Subduction Margin

<p>Repeating earthquakes provide a novel way of monitoring how stresses load faults between large earthquakes. In this thesis, we develop a method and composite criterion for identifying repeating earthquakes in New Zealand and present New Zealand’s first long-duration repeating earthquake catalogue. This thesis addresses three primary objectives: (1) develop a method and composite criterion for identifying repeating earthquakes; (2) build a long-duration catalogue of repeating earthquakes for the Raukumara Peninsula; and (3) apply the method and composite criterion in different tectonic settings to investigate whether it can be applied more broadly elsewhere in New Zealand. The systematic identification of repeating earthquakes in New Zealand provides the first step in being able to monitor the state of stresses of New Zealand’s active faults in situ throughout the earthquake cycle. Studies elsewhere, particularly in Japan and California, have developed case-specific criteria for identifying repeating earthquakes. Building on these studies, we develop a method and composite criterion for identifying repeating earthquakes in New Zealand, focusing on seismicity around the Raukumara Peninsula. Our composite criterion states that for events to be identified as repeating earthquakes, two or more events must have a normalised cross-correlation of at least 0.95 at two or more seismic stations, when calculated for 75% of the earthquake coda. Sensitivity to correlation window length, filtering frequency-band and correlation threshold were tested during the development of the composite criterion. These tests indicated that small perturbations to the parameter thresholds did not affect our ability to detect repeating earthquakes using the composite criterion. By applying our composite criterion to seismicity around the Raukumara Peninsula, we identified 62 repeating earthquake families occurring between 2003 and 2018, consisting of 160 individual earthquakes. These families have a magnitude range of MW 1.5–4.5, and have recurrence intervals and family durations of < 1–12 years. High-precision absolute and relative locations were calculated using manual phase picks and cross-correlation re-picking. Focal mechanisms for 56 of the families were also determined, using P-wave first motions, revealing predominantly strike-slip and normal faulting at shallow depths, low-angle reverse faulting along the subduction interface, and normal faulting in the subducting plate. We compared the timing of the repeating earthquakes to slow-slip events previously identified using geodetic measurements around the Raukumara Peninsula and observed that repeating earthquakes occurred during 26 of the 31 identified periods of slow-slip. We also compared the seismic moment– recurrence interval relationship of the Raukumara Peninsula repeating earthquakes to that of earthquakes near Parkfield, California, identified by Nadeau and Johnson (1998), and observed a similar functional relationship. Slip-rates of the Raukumara Peninsula repeating earthquake families were also calculated using a slip-rate–moment relationship and were found to vary from < 10mm/yr to 80mm/yr. We applied the method and composite criterion developed for the Raukumara Peninsula to two other locations to ensure it could be applied successfully in other New Zealand regions with different seismotectonic characteristics. Using our workflow, we successfully identified four families in Marlborough, and three families around Fiordland. These families differ from those identified around the Raukumara Peninsula in that they had relatively short recurrence intervals and family durations, of 2 minutes– 15 months. The ability of the composite criterion to identify these families confirms its suitability for further studies of repeating earthquakes throughout New Zealand.</p>

Repeating Earthquake Detection and Characterisation in New Zealand: A Catalogue for the Raukumara Peninsula, Northern Hikurangi Subduction Margin

<p>Repeating earthquakes provide a novel way of monitoring how stresses load faults between large earthquakes. In this thesis, we develop a method and composite criterion for identifying repeating earthquakes in New Zealand and present New Zealand’s first long-duration repeating earthquake catalogue. This thesis addresses three primary objectives: (1) develop a method and composite criterion for identifying repeating earthquakes; (2) build a long-duration catalogue of repeating earthquakes for the Raukumara Peninsula; and (3) apply the method and composite criterion in different tectonic settings to investigate whether it can be applied more broadly elsewhere in New Zealand. The systematic identification of repeating earthquakes in New Zealand provides the first step in being able to monitor the state of stresses of New Zealand’s active faults in situ throughout the earthquake cycle. Studies elsewhere, particularly in Japan and California, have developed case-specific criteria for identifying repeating earthquakes. Building on these studies, we develop a method and composite criterion for identifying repeating earthquakes in New Zealand, focusing on seismicity around the Raukumara Peninsula. Our composite criterion states that for events to be identified as repeating earthquakes, two or more events must have a normalised cross-correlation of at least 0.95 at two or more seismic stations, when calculated for 75% of the earthquake coda. Sensitivity to correlation window length, filtering frequency-band and correlation threshold were tested during the development of the composite criterion. These tests indicated that small perturbations to the parameter thresholds did not affect our ability to detect repeating earthquakes using the composite criterion. By applying our composite criterion to seismicity around the Raukumara Peninsula, we identified 62 repeating earthquake families occurring between 2003 and 2018, consisting of 160 individual earthquakes. These families have a magnitude range of MW 1.5–4.5, and have recurrence intervals and family durations of < 1–12 years. High-precision absolute and relative locations were calculated using manual phase picks and cross-correlation re-picking. Focal mechanisms for 56 of the families were also determined, using P-wave first motions, revealing predominantly strike-slip and normal faulting at shallow depths, low-angle reverse faulting along the subduction interface, and normal faulting in the subducting plate. We compared the timing of the repeating earthquakes to slow-slip events previously identified using geodetic measurements around the Raukumara Peninsula and observed that repeating earthquakes occurred during 26 of the 31 identified periods of slow-slip. We also compared the seismic moment– recurrence interval relationship of the Raukumara Peninsula repeating earthquakes to that of earthquakes near Parkfield, California, identified by Nadeau and Johnson (1998), and observed a similar functional relationship. Slip-rates of the Raukumara Peninsula repeating earthquake families were also calculated using a slip-rate–moment relationship and were found to vary from < 10mm/yr to 80mm/yr. We applied the method and composite criterion developed for the Raukumara Peninsula to two other locations to ensure it could be applied successfully in other New Zealand regions with different seismotectonic characteristics. Using our workflow, we successfully identified four families in Marlborough, and three families around Fiordland. These families differ from those identified around the Raukumara Peninsula in that they had relatively short recurrence intervals and family durations, of 2 minutes– 15 months. The ability of the composite criterion to identify these families confirms its suitability for further studies of repeating earthquakes throughout New Zealand.</p>

Benefits of Defining Geological Sensitive Zones in the Mitigation of Disasters Along Earthquake Fault Zones in Taiwan – The Case of Milun Fault

In Taiwan, the main purpose of earthquake fault zone legislation is to prevent earthquake-related disasters around the surface traces of active faults, particularly in urban areas. Here, the Geologically Sensitive Area (GSA) of the Milun Fault (Milun Earthquake Fault Zone) is used as an example to reveal the importance of such legislation. Field data collected along the Milun Fault before and after the 2018 Hualien Earthquake were used to reveal the reappearance of damages within the GSA. The 2018 Hualien Earthquake represents one of the shortest recurrence intervals (67 years) among all major faults in Taiwan. Most of the surface ruptures and damaged buildings in Hualien City were within the Milun Fault GSA and concentrated on the hanging wall of the fault. Moreover, 61% (91/148) of the damaged buildings and 83% (692/835) of the surface ruptures occurred within 100 m of the fault line. The results of this study demonstrate the importance of defining GSAs of active faults for mitigating earthquake hazards.

Geologic and geodetic constraints on the seismic hazard of Malawi’s active faults: The Malawi Seismogenic Source Database (MSSD)

Active fault data are commonly used in seismic hazard assessments, but there are challenges in deriving the slip rate, geometry, and frequency of earthquakes along active faults. Herein, we present the open-access geospatial Malawi Seismogenic Source Database (MSSD), which describes the seismogenic properties of faults that have formed during East African rifting in Malawi. We first use empirical observations to geometrically classify active faults into section, fault, and multi-fault seismogenic sources. For sources in the North Basin of Lake Malawi, slip rates can be derived from the vertical offset of a seismic reflector that is estimated to be 75 ka based on dated core. Elsewhere, slip rates are constrained from advancing a ‘systems-based’ approach that partitions geodetically-derived rift extension rates in Malawi between seismogenic sources using a priori constraints on regional strain distribution in magma-poor continental rifts. Slip rates are then combined with source geometry and empirical scaling relationships to estimate earthquake magnitudes and recurrence intervals, and their uncertainty is described from the variability of outcomes from a logic tree used in these calculations. We find that for sources in the Lake Malawi’s North Basin, where slip rates can be derived from both the geodetic data and the offset seismic reflector, the slip rate estimates are within error of each other, although those from the offset reflector are higher. Sources in the MSSD are 5–200 km long, which implies that large magnitude (MW 7–8) earthquakes may occur in Malawi. Low slip rates (0.05–2 mm/yr), however, mean that the frequency of such events will be low (recurrence intervals ~103–104 years). The MSSD represents an important resource for investigating Malawi’s increasing seismic risks and provides a framework for incorporating active fault data into seismic hazard assessment in other tectonically active regions.

Magnitude and source area estimations of severe prehistoric earthquakes in the western Eastern Alps

In slowly deforming intraplate tectonic regions such as the Alps only limited knowledge exists on the occurrence of severe earthquakes, their maximum possible magnitude and their potential source areas. This is mainly due to long earthquake recurrence rates exceeding the time span of instrumental earthquake records and historical documentation. Lacustrine paleoseismology aims at retrieving long-term continuous records of seismic shaking. A paleoseismic record from a single lake provides information on events for which seismic shaking exceeded the intensity threshold at the lake site. In addition, when positive and negative evidence for seismic shaking from multiple sites can be gathered for a certain time period, minimum magnitudes and source locations can be estimated for paleo-earthquakes by a reverse application of an empirical intensity prediction equation in a geospatial analysis. Here, we present potential magnitudes and source locations of four paleo-earthquakes in the western Eastern Alps based on the integration of available and updated lake paleoseismic data. The paleoseismic records at Plansee and Achensee covering the last ~10 kyrs were extended towards the age of lake initiation after deglaciation to obtain the longest possible paleoseismic catalogue at each lake site. Our results show that 25 severe earthquakes are recorded in the four lakes Plansee, Piburgersee, Achensee and potentially Starnbergersee over the last ~16 kyrs, from which four earthquakes are interpreted to left imprints in two or more lakes. Earthquake recurrence intervals range from ca. 1,000 to 2,000 years with a weakly periodic to aperiodic recurrence behavior for the individual records. We interpret that relatively shorter recurrence intervals in the more orogen-internal archives Piburgersee and Achensee are related to enhanced tectonic loading, whereas a longer recurrence rate in the more orogen-external archive Plansee might reflect a decreased stress transfer across the current-day enhanced seismicity zone. Plausible epicenters of paleo-earthquake scenarios coincide with the current enhanced seismicity regions. Prehistoric earthquakes with a minimum moment magnitude (MW) 5.8–6.1 might have occurred around the Inn valley, the Brenner region and the Fernpass-Loisach region, and might have reached up to MW 6.3 at Achensee. The paleo-earthquake catalogue might hint at a shift of severe earthquake activity near the Inn valley from east to west to east during Postglacial times. Shakemaps highlight that such severe earthquake scenarios not solely impact the enhanced seismicity region of Tyrol, but widely affect adjacent regions like southern Bavaria in Germany.

Forecasting excessive rainfall with random forests and a deterministic convection-allowing model

Approximately seven years of daily initializations from the convection-allowing National Severe Storms Laboratory Weather Research and Forecasting model are used as inputs to train random forest (RF) machine learning models to probabilistically predict instances of excessive rainfall. Unlike other hazards, excessive rainfall does not have an accepted definition, so multiple definitions of excessive rainfall and flash flooding – including flash flood reports and 24-hr average recurrence intervals (ARIs) – are used to explore RF configuration forecast sensitivities. RF forecasts are analogous to operational Weather Prediction Center (WPC) day-1 Excessive Rainfall Outlooks (EROs) and their resolution, reliability, and skill are strongly influenced by rainfall definitions and how inputs are assembled for training. Models trained with 1-y ARI exceedances defined by the Stage-IV (ST4) precipitation analysis perform poorly in the northern Great Plains and southwest U.S., in part due to a high bias in the number of training events in these regions. Increasing the ARI threshold to 2 years or removing ST4 data from training, optimizing forecast skill geographically, and spatially averaging meteorological inputs for training generally results in improved CONUS-wide RF forecast skill. Both EROs and RF forecasts have seasonal skill – poor forecasts in the late fall and winter and skillful forecasts in the summer and early fall. However, the EROs are consistently and significantly better than their RF counterparts, regardless of RF configuration, particularly in the summer months. The results suggest careful consideration should be made when developing ML-based probabilistic precipitation forecasts with convection-allowing model inputs, and further development is necessary to consider these forecast products for operational implementation.

Spatial changes in inclusion band spacing as an indicator of temporal changes in slow slip and tremor recurrence intervals

Slow slip and tremor (SST) downdip of the seismogenic zones may trigger megathrust earthquakes by frequently transferring stress to seismogenic zones. Geodetic observations have suggested that the recurrence intervals of slow slip decrease toward the next megathrust earthquake. However, temporal variations in the recurrence intervals of SST during megathrust earthquake cycles remain poorly understood because of the limited duration of geodetic and seismological monitoring of slow earthquakes. The quartz-filled, crack-seal shear veins in the subduction mélange deformed near the downdip limit of the seismogenic zone in warm-slab environments record cyclic changes in the inclusion band spacing in the range from 4 ± 1 to 65 ± 18 μm. The two-phase primary fluid inclusions in quartz between inclusion bands exhibit varying vapor/liquid ratios regardless of inclusion band spacing, suggesting a common occurrence of fast quartz sealing due to a rapid decrease in quartz solubility associated with a large fluid pressure reduction. A kinetic model of quartz precipitation, considering a large fluid pressure change and inclusion band spacing, indicates that the sealing time during a single crack-seal event cyclically decreased and increased in the range from 0.16 ± 0.04 to 2.7 ± 0.8 years, with one cycle lasting at least 27 ± 2 to 93 ± 5 years. The ranges of sealing time and duration of a cycle may be comparable to the recurrence intervals of SST and megathrust earthquakes, respectively. We suggest that the spatial change in inclusion band spacing is a potential geological indicator of temporal changes in SST recurrence intervals, particularly when large fluid pressure reduction occurs by brittle fracturing.

Repeating earthquakes and ground deformation reveal the structure and triggering mechanisms of the Pernicana fault, Mt. Etna

Structure and dynamics of fault systems can be investigated using repeating earthquakes as repeatable seismic sources, alongside ground deformation measurements. Here we utilise a dataset of repeating earthquakes which occurred between 2000 and 2019 along the transtensive Pernicana fault system on the northeast flank of Mount Etna, Italy, to investigate the fault structure, as well as the triggering mechanisms of the seismicity. By grouping the repeating earthquakes into families and integrating the seismic data with GPS measurements of ground deformation, we identify four distinct portions of the fault. Each portion shows a different behaviour in terms of seismicity, repeating earthquakes and ground deformation, which we attribute to structural differences including a segmentation of the fault plane at depth. The recurrence intervals of repeating earthquake families display a low degree of regularity which suggests an episodic triggering mechanism, such as magma intrusion, rather than displacement under a constant stress.