Seismically vulnerable buildings constitute a major problem for the safety of human beings. In many parts of the world, reinforced concrete (RC) frame buildings designed and constructed with substandard detailing, no consideration of capacity design principles, and improper or no inclusion of the seismic actions, have been identified. Amongst those vulnerable building, one particular typology representative of the construction practice of the years previous to the 1970’s, that most likely represents the worst case scenario, has been widely investigated in the past. The deficiencies of that building typology are related to non-ductile detailing in beam column joints such as the use of plain round bars, the lack of stirrups inside the joint around the longitudinal reinforcement of the column, the use of 180° end hooks in the beams, the use of lap splices in potential ‘plastic hinge’ regions, and substandard quality of the materials. That type of detailing and the lack of a capacity design philosophy create a very fragile fuse in the structure where brittle inelastic behaviour is expected to occur, which is the panel zone region of exterior beam column joints. The non-ductile typology described above was extensively investigated at the University of Canterbury in the context of the project ‘Retrofit Solutions for New Zealand Multi-Storey Buildings’ (2004-2011), founded by the ‘Foundation for Research, Science and Technology’ Tūāpapa Rangahau Pūtaiao. The experimental campaign prior to the research carried out by the author consisted of quasi-static tests of beam column joint subassemblies subjected to lateral loading regime, with constant and varying axial load in the column. Most of those specimens were representative of a plane 2D frame (knee joint), while others represented a portion of a space 3D frame (corner joints), and only few of them had a floor slab, transverse beams, and lap splices. Using those experiments, several feasible, cost-effective, and non-invasive retrofit techniques were developed, improved, and refined. Nevertheless, the slow motion nature of those experiments did not take into account the dynamical component inherent to earthquake related problems. Amongst the set of techniques investigated, the use of FRP layers for strengthening beam column joints is of particular interest due to its versatility and the momentum that its use has gained in the current state of the practice. That particular retrofit technique was previously used to develop a strengthening scheme suitable for plane 2D and space 3D corner beam column joints, but lacking of floor slabs. In addition, a similar scheme was not developed for exterior joints of internal frames, referred here as ‘cruciform’. In this research a 2/5 scale RC frame model building comprising of two frames in parallel (external and internal) joined together by means of floor slabs and transverse beams, with non-ductile characteristics identical to those of the specimens investigated previously by others, and also including lap splices, was developed. In order to investigate the dynamic response of that building, a series of shake table tests with different ground motions were performed. After the first series of tests, the specimen was modified by connecting the spliced reinforcement in the columns in order to capture a different failure mode. Ground motions recorded during seismic events that occurred during the initial period of the experimental campaign (2010) were used in the subsequent experiments. The hierarchy of strengths and sequence of events in the panel zone region were evaluated in an extended version of the bending moment-axial load (M-N) performance domain developed by others. That extension was required due to the asymmetry in the beam cross section introduced by the floor slab. In addition, the effect of the torsion resistance provided by the spandrel (transverse beam) was included. In order to upgrade the brittle and unstable performance of the as-built/repaired specimen, a practical and suitable ad-hoc FRP retrofit intervention was developed, following a partial retrofit strategy that aimed to strengthen exterior beam column joints only (corner and cruciform). The ability of the new FRP scheme to revert the sequence of events in the panel zone region was evaluated using the extended version of the M-N performance domain as well as the guidelines for strengthening plane joints developed by others. Weakening of the floor slab in a novel configuration was also incorporated with the purpose of reducing the flexural capacity of the beam under negative bending moment (slab in tension), enabling the damage relocation from the joint into the beam. The efficacy of the developed retrofit intervention in upgrading the seismic performance of the as-built specimen was investigated using shake table tests with the input motions used in the experiments of the as-built/repaired specimen. Numerical work aimed to predict the response of the model building during the most relevant shake table tests was carried out. By using a simple numerical model with concentrated plasticity elements constructed in Ruaumoko2D, the results of blind and post-experimental predictions of the response of the specimen were addressed. Differences in the predicted response of the building using the nominal and the actual recorded motions of the shake table were investigated. The dependence of the accuracy of the numerical predictions on the assumed values of the parameters that control the hysteresis rules of key structural members was reviewed. During the execution of the experimental campaign part of this thesis, two major earthquakes affected the central part of Chile (27 of February 2010 Maule earthquake) and the Canterbury region in New Zealand (22 February 2011 Canterbury earthquake), respectively. As the author had the opportunity to experience those events and investigate their consequences in structures, the observations related to non-ductile detailing and drawbacks in the state of the practice related to reinforced concrete walls was also addressed in this research, resulting in preliminary recommendations for the refinement of current seismic code provisions and assessment guidelines. The investigations of the ground motions recorded during those and other earthquakes were used to review the procedures related to the input motions used for nonlinear dynamic analysis of buildings as required by most of the current code provisions. Inelastic displacement spectra were constructed using ground motions recorded during the earthquakes mentioned above, in order to investigate the adequacy of modification factors used to obtain reduced design spectra from elastic counterparts. Finally a simplified assessment procedure for RC walls that incorporates capacity compatible spectral demands is proposed.
This study investigates the uncertainty of simulated earthquake ground motions for smallmagnitude events (Mw 3.5 – 5) in Canterbury, New Zealand. 148 events were simulated with specified uncertainties in: event magnitude, hypocentre location, focal mechanism, high frequency rupture velocity, Brune stress parameter, the site 30-m time-averaged shear wave velocity (Vs30), anelastic attenuation (Q) and high frequency path duration. In order to capture these uncertainties, 25 realisations for each event were generated using the Graves and Pitarka (2015) hybrid broadband simulation approach. Monte-Carlo realisations were drawn from distributions for each uncertainty, to generate a suite of simulation realisations for each event and site. The fit of the multiple simulation realisations to observations were assessed using linear mixed effects regression to generate the systematic source, path and site effects components across all ground motion intensity measure residuals. Findings show that additional uncertainties are required in each of the three source, path, and site components, however the level of output uncertainty is promising considering the input uncertainties included.
This study provides an initial examination of source parameter uncertainty in a New Zealand ground motion simulation model, by simulating multiple event realisations with perturbed source parameters. Small magnitude events in Canterbury have been selected for this study due to the small number of source input parameters, the wealth of recorded data, and the lack of appreciable off-fault non-linear effects. Which provides greater opportunity to identify systematic source, path and site effects, required to robustly investigate the causes of uncertainty.
This dissertation addresses several fundamental and applied aspects of ground motion selection for seismic response analyses. In particular, the following topics are addressed: the theory and application of ground motion selection for scenario earthquake ruptures; the consideration of causal parameter bounds in ground motion selection; ground motion selection in the near-fault region where directivity effect is significant; and methodologies for epistemic uncertainty consideration and propagation in the context of ground motion selection and seismic performance assessment. The paragraphs below outline each contribution in more detail. A scenario-based ground motion selection method is presented which considers the joint distribution of multiple intensity measure (IM) types based on the generalised conditional intensity measure (GCIM) methodology (Bradley, 2010b, 2012c). The ground motion selection algorithm is based on generating realisations of the considered IM distributions for a specific rupture scenario and then finding the prospective ground motions which best fit the realisations using an optimal amplitude scaling factor. In addition, using different rupture scenarios and site conditions, two important aspects of the GCIM methodology are scrutinised: (i) different weight vectors for the various IMs considered; and (ii) quantifying the importance of replicate selections for ensembles with different numbers of desired ground motions. As an application of the developed scenario-based ground motion selection method, ground motion ensembles are selected to represent several major earthquake scenarios in New Zealand that pose a significant seismic hazard, namely, Alpine, Hope and Porters Pass ruptures for Christchurch city; and Wellington, Ohariu, and Wairarapa ruptures for Wellington city. A rigorous basis is developed, and sensitivity analyses performed, for the consideration of bounds on causal parameters (e.g., magnitude, source-to-site distance, and site condition) for ground motion selection. The effect of causal parameter bound selection on both the number of available prospective ground motions from an initial empirical as-recorded database, and the statistical properties of IMs of selected ground motions are examined. It is also demonstrated that using causal parameter bounds is not a reliable approach to implicitly account for ground motion duration and cumulative effects when selection is based on only spectral acceleration (SA) ordinates. Specific causal parameter bounding criteria are recommended for general use as a ‘default’ bounding criterion with possible adjustments from the analyst based on problem-specific preferences. An approach is presented to consider the forward directivity effects in seismic hazard analysis, which does not separate the hazard calculations for pulse-like and non-pulse-like ground motions. Also, the ability of ground motion selection methods to appropriately select records containing forward directivity pulse motions in the near-fault region is examined. Particular attention is given to ground motion selection which is explicitly based on ground motion IMs, including SA, duration, and cumulative measures; rather than a focus on implicit parameters (i.e., distance, and pulse or non-pulse classifications) that are conventionally used to heuristically distinguish between the near-fault and far-field records. No ad hoc criteria, in terms of the number of directivity ground motions and their pulse periods, are enforced for selecting pulse-like records. Example applications are presented with different rupture characteristics, source-to-site geometry, and site conditions. It is advocated that the selection of ground motions in the near-fault region based on IM properties alone is preferred to that in which the proportion of pulse-like motions and their pulse periods are specified a priori as strict criteria for ground motion selection. Three methods are presented to propagate the effect of seismic hazard and ground motion selection epistemic uncertainties to seismic performance metrics. These methods differ in their level of rigor considered to propagate the epistemic uncertainty in the conditional distribution of IMs utilised in ground motion selection, selected ground motion ensembles, and the number of nonlinear response history analyses performed to obtain the distribution of engineering demand parameters. These methods are compared for an example site where it is observed that, for seismic demand levels below the collapse limit, epistemic uncertainty in ground motion selection is a smaller uncertainty contributor relative to the uncertainty in the seismic hazard itself. In contrast, uncertainty in ground motion selection process increases the uncertainty in the seismic demand hazard for near-collapse demand levels.
Since the early 1980s seismic hazard assessment in New Zealand has been based on Probabilistic Seismic Hazard Analysis (PSHA). The most recent version of the New Zealand National Seismic Hazard Model, a PSHA model, was published by Stirling et al, in 2012. This model follows standard PSHA principals and combines a nation-wide model of active faults with a gridded point-source model based on the earthquake catalogue since 1840. These models are coupled with the ground-motion prediction equation of McVerry et al (2006). Additionally, we have developed a time-dependent clustering-based PSHA model for the Canterbury region (Gerstenberger et al, 2014) in response to the Canterbury earthquake sequence. We are now in the process of revising that national model. In this process we are investigating several of the fundamental assumptions in traditional PSHA and in how we modelled hazard in the past. For this project, we have three main focuses: 1) how do we design an optimal combination of multiple sources of information to produce the best forecast of earthquake rates in the next 50 years: can we improve upon a simple hybrid of fault sources and background sources, and can we better handle the uncertainties in the data and models (e.g., fault segmentation, frequency-magnitude distributions, time-dependence & clustering, low strain-rate areas, and subduction zone modelling)? 2) developing revised and new ground-motion predictions models including better capturing of epistemic uncertainty – a key focus in this work is developing a new strong ground motion catalogue for model development; and 3) how can we best quantify if changes we have made in our modelling are truly improvements? Throughout this process we are working toward incorporating numerical modelling results from physics based synthetic seismicity and ground-motion models.
The performance of conventionally designed reinforced concrete (RC) structures during the 2011 Christchurch earthquake has demonstrated that there is greater uncertainty in the seismic performance of RC components than previously understood. RC frame and wall structures in the Christchurch central business district were observed to form undesirable cracks patterns in the plastic hinge region while yield penetration either side of cracks, and into development zones, were less than theoretical predictions. The implications of this unexpected behaviour: (i) significantly less available ductility; (ii) less hysteretic energy dissipation; and (iii) the localization of peak reinforcement strains, results in considerable doubt for the residual capacity of RC structures. The significance of these consequences has prompted a review of potential sources of uncertainty in seismic experimentation with the intention to improve the current confidence level for newly designed conventional RC structures. This paper attempts to revisit the principles of RC mechanics, in particular, to consider the influence of loading history, concrete tensile strength, and reinforcement ratio on the performance of ‘real’ RC structures compared to experimental test specimens.
This paper investigates the effects of variability in source rupture parameters on site-specific physics-based simulated ground motions, ascertained through the systematic analysis of ground motion intensity measures. As a preliminary study, we consider simulations of the 22 February 2011 Christchurch earthquake using the Graves and Pitarka (2015) methodology. The effects of source variability are considered via a sensitivity study in which parameters (hypocentre location, earthquake magnitude, average rupture velocity, fault geometry and the Brune stress parameter) are individually varied by one standard deviation. The sensitivity of simulated ground motion intensity measures are subsequently compared against observational data. The preliminary results from this study indicate that uncertainty in the stress parameter and the rupture velocity have the most significant effect on the high frequency amplitudes. Conversely, magnitude uncertainty was found to be most influential on the spectral acceleration amplitudes at low frequencies. Further work is required to extend this preliminary study to exhaustively consider more events and to include parameter covariance. The ultimate results of this research will assist in the validation of the overall simulation method’s accuracy in capturing various rupture parameters, which is essential for the use of simulated ground motion models in probabilistic seismic hazard analysis.
In the last two decades, New Zealand (NZ) has experienced significant earthquakes, including the 2010 M 7.2 Darfield, 2011 M 6.2 Christchurch, and 2016 M 7.8 Kaikōura events. Amongst these large events, tens of thousands of smaller earthquakes have occurred. While previous event and ground-motion databases have analyzed these events, many events below M 4 have gone undetected. The goal of this study is to expand on previous databases, particularly for small magnitude (M<4) and low-amplitude ground motions. This new database enables a greater understanding of regional variations within NZ and contributes to the validity of internationally developed ground-motion models. The database includes event locations and magnitude estimates with uncertainty considerations, and tectonic type assessed in a hierarchical manner. Ground motions are extracted from the GeoNet FDSN server and assessed for quality using a neural network classification approach. A deep neural network approach is also utilized for picking P and S phases for determination of event hypocentres. Relative hypocentres are further improved by double-difference relocation and will contribute toward developing shallow (< 50 km) seismic tomography models. Analysis of the resulting database is compared with previous studies for discussion of implications toward national hazard prediction models.
One of the most controversial issues highlighted by the 2010-2011 Christchurch earthquake series and more recently the 2016 Kaikoura earthquake, has been the evident difficulty and lack of knowledge and guidelines for: a) evaluation of the residual capacity damaged buildings to sustain future aftershocks; b) selection and implementation of a series of reliable repairing techniques to bring back the structure to a condition substantially the same as prior to the earthquake; and c) predicting the cost (or cost-effectiveness) of such repair intervention, when compared to fully replacement costs while accounting for potential aftershocks in the near future. As a result of such complexity and uncertainty (i.e., risk), in combination with the possibility (unique in New Zealand when compared to most of the seismic-prone countries) to rely on financial support from the insurance companies, many modern buildings, in a number exceeding typical expectations from past experiences at an international level, have ended up being demolished. This has resulted in additional time and indirect losses prior to the full reconstruction, as well as in an increase in uncertainty on the actual relocation of the investment. This research project provides the main end-users and stakeholders (practitioner engineers, owners, local and government authorities, insurers, and regulatory agencies) with comprehensive evidence-based information to assess the residual capacity of damage reinforced concrete buildings, and to evaluate the feasibility of repairing techniques, in order to support their delicate decision-making process of repair vs. demolition or replacement. Literature review on effectiveness of epoxy injection repairs, as well as experimental tests on full-scale beam-column joints shows that repaired specimens have a reduced initial stiffness compared with the undamaged specimen, with no apparent strength reduction, sometimes exhibiting higher displacement ductility capacities. Although the bond between the steel and concrete is only partially restored, it still allows the repaired specimen to dissipate at least the same amount of hysteretic energy. Experimental tests on buildings subjected to earthquake loading demonstrate that even for severe damage levels, the ability of the epoxy injection to restore the initial stiffness of the structure is significant. Literature review on damage assessment and repair guidelines suggests that there is consensus within the international community that concrete elements with cracks less than 0.2 mm wide only require cosmetic repairs; epoxy injection repairs of cracks less and 2.0 mm wide and concrete patching of spalled cover concrete (i.e., minor to moderate damage) is an appropiate repair strategy; and for severe damaged components (e.g., cracks greater than 2.0 mm wide, crushing of the concrete core, buckling of the longitudinal reinforcement) local replacement of steel and/or concrete in addition to epoxy crack injection is more appropriate. In terms of expected cracking patterns, non-linear finite element investigations on well-designed reinforced concrete beam-to-column joints, have shown that lower number of cracks but with wider openings are expected to occur for larger compressive concrete strength, f’c, and lower reinforcement content, ρs. It was also observed that the tensile concrete strength, ft, strongly affects the expected cracking pattern in the beam-column joints, the latter being more uniformly distributed for lower ft values. Strain rate effects do not seem to play an important role on the cracking pattern. However, small variations in the cracking pattern were observed for low reinforcement content as it approaches to the minimum required as per NZS 3101:2006. Simple equations are proposed in this research project to relate the maximum and residual crack widths with the steel strain at peak displacement, with or without axial load. A literature review on fracture of reinforcing steel due to low-cycle fatigue, including recent research using steel manufactured per New Zealand standards is also presented. Experimental results describing the influence of the cyclic effect on the ultimate strain capacity of the steel are also discussed, and preliminary equations to account for that effect are proposed. A literature review on the current practice to assess the seismic residual capacity of structures is also presented. The various factors affecting the residual fatigue life at a component level (i.e., plastic hinge) of well-designed reinforced concrete frames are discussed, and equations to quantify each of them are proposed, as well as a methodology to incorporate them into a full displacement-based procedure for pre-earthquake and post-earthquake seismic assessment.
Using case studies from the 2010-2011 Canterbury, New Zealand earthquake sequence, this study assesses the accuracies of paleoliquefaction back-analysis methods and explores the challenges, techniques, and uncertainties associated with their application. While liquefaction-based back-analyses have been widely used to estimate the magnitudes of paleoearthquakes, their uncertain efficacies continue to significantly affect the computed seismic hazard in regions where they are relied upon. Accordingly, their performance is evaluated herein using liquefaction data from modern earthquakes with known magnitudes. It is shown that when the earthquake source location and mechanism are known, back-analysis methods are capable of accurately deriving seismic parameters from liquefaction evidence. However, because the source location and mechanism are often unknown in paleoseismic studies, and because accurate interpretation is shown to be more difficult in such cases, new analysis techniques are proposed herein. An objective parameter is proposed to geospatially assess the likelihood of any provisional source location, enabling an analyst to more accurately estimate the magnitude of a liquefaction-inducing paleoearthquake. This study demonstrates the application of back-analysis methods, provides insight into their potential accuracies, and provides a framework for performing paleoliquefaction analyses worldwide.
Previous earthquakes demonstrated destructive effects of soil-structure interaction on structural response. For example, in the 1970 Gediz earthquake in Turkey, part of a factory was demolished in a town 135 km from the epicentre, while no other buildings in the town were damaged. Subsequent investigations revealed that the fundamental period of vibration of the factory was approximately equal to that of the underlying soil. This alignment provided a resonance effect and led to collapse of the structure. Another dramatic example took place in Adapazari, during the 1999 Kocaeli earthquake where several foundations failed due to either bearing capacity exceedance or foundation uplifting, consequently, damaging the structure. Finally, the Christchurch 2012 earthquakes have shown that significant nonlinear action in the soil and soil-foundation interface can be expected due to high levels of seismic excitation and spectral acceleration. This nonlinearity, in turn, significantly influenced the response of the structure interacting with the soil-foundation underneath. Extensive research over more than 35 years has focused on the subject of seismic soil-structure interaction. However, since the response of soil-structure systems to seismic forces is extremely complex, burdened by uncertainties in system parameters and variability in ground motions, the role of soil-structure interaction on the structural response is still controversial. Conventional design procedures suggest that soil-structure interaction effects on the structural response can be conservatively ignored. However, more recent studies show that soil-structure interaction can be either beneficial or detrimental, depending on the soil-structure-earthquake scenarios considered. In view of the above mentioned issues, this research aims to utilise a comprehensive and systematic probabilistic methodology, as the most rational way, to quantify the effects of soil-structure interaction on the structural response considering both aleatory and epistemic uncertainties. The goal is achieved by examining the response of established rheological single-degree-of-freedom systems located on shallow-foundation and excited by ground motions with different spectral characteristics. In this regard, four main phases are followed. First, the effects of seismic soil-structure interaction on the response of structures with linear behaviour are investigated using a robust stochastic approach. Herein, the soil-foundation interface is modelled by an equivalent linear cone model. This phase is mainly considered to examine the influence of soil-structure interaction on the approach that has been adopted in the building codes for developing design spectrum and defining the seismic forces acting on the structure. Second, the effects of structural nonlinearity on the role of soil-structure interaction in modifying seismic structural response are studied. The same stochastic approach as phase 1 is followed, while three different types of structural force-deflection behaviour are examined. Third, a systematic fashion is carried out to look for any possible correlation between soil, structural, and system parameters and the degree of soil-structure interaction effects on the structural response. An attempt is made to identify the key parameters whose variation significantly affects the structural response. In addition, it is tried to define the critical range of variation of parameters of consequent. Finally, the impact of soil-foundation interface nonlinearity on the soil-structure interaction analysis is examined. In this regard, a newly developed macro-element covering both material and geometrical soil-foundation interface nonlinearity is implemented in a finite-element program Raumoko 3D. This model is then used in an extensive probabilistic simulation to compare the effects of linear and nonlinear soil-structure interaction on the structural response. This research is concluded by reviewing the current design guidelines incorporating soil-structure interaction effects in their design procedures. A discussion is then followed on the inadequacies of current procedures based on the outcomes of this study.
The overarching goal of this dissertation is to improve predictive capabilities of geotechnical seismic site response analyses by incorporating additional salient physical phenomena that influence site effects. Specifically, multidimensional wave-propagation effects that are neglected in conventional 1D site response analyses are incorporated by: (1) combining results of 3D regional-scale simulations with 1D nonlinear wave-propagation site response analysis, and (2) modelling soil heterogeneity in 2D site response analyses using spatially-correlated random fields to perturb soil properties. A method to combine results from 3D hybrid physics-based ground motion simulations with site-specific nonlinear site response analyses was developed. The 3D simulations capture 3D ground motion phenomena on a regional scale, while the 1D nonlinear site response, which is informed by detailed site-specific soil characterization data, can capture site effects more rigorously. Simulations of 11 moderate-to-large earthquakes from the 2010-2011 Canterbury Earthquake Sequence (CES) at 20 strong motion stations (SMS) were used to validate simulations with observed ground motions. The predictions were compared to those from an empirically-based ground motion model (GMM), and from 3D simulations with simplified VS30- based site effects modelling. By comparing all predictions to observations at seismic recording stations, it was found that the 3D physics-based simulations can predict ground motions with comparable bias and uncertainty as the GMM, albeit, with significantly lower bias at long periods. Additionally, the explicit modelling of nonlinear site-response improves predictions significantly compared to the simplified VS30-based approach for soft-soil or atypical sites that exhibit exceptionally strong site effects. A method to account for the spatial variability of soils and wave scattering in 2D site response analyses was developed and validated against a database of vertical array sites in California. The inputs required to run the 2D analyses are nominally the same as those required for 1D analyses (except for spatial correlation parameters), enabling easier adoption in practice. The first step was to create the platform and workflow, and to perform a sensitivity study involving 5,400 2D model realizations to investigate the influence of random field input parameters on wave scattering and site response. Boundary conditions were carefully assessed to understand their effect on the modelled response and select appropriate assumptions for use on a 2D model with lateral heterogeneities. Multiple ground-motion intensity measures (IMs) were analyzed to quantify the influence from random field input parameters and boundary conditions. It was found that this method is capable of scattering seismic waves and creating spatially-varying ground motions at the ground surface. The redistribution of ground-motion energy across wider frequency bands, and the scattering attenuation of high-frequency waves in 2D analyses, resemble features observed in empirical transfer functions (ETFs) computed in other studies. The developed 2D method was subsequently extended to more complicated multi-layer soil profiles and applied to a database of 21 vertical array sites in California to test its appropriate- ness for future predictions. Again, different boundary condition and input motion assumptions were explored to extend the method to the in-situ conditions of a vertical array (with a sensor embedded in the soil). ETFs were compared to theoretical transfer functions (TTFs) from conventional 1D analyses and 2D analyses with heterogeneity. Residuals of transfer-function- based IMs, and IMs of surface ground motions, were also used as validation metrics. The spatial variability of transfer-function-based IMs was estimated from 2D models and compared to the event-to-event variability from ETFs. This method was found capable of significantly improving predictions of median ETF amplification factors, especially for sites that display higher event-to-event variability. For sites that are well represented by 1D methods, the 2D approach can underpredict amplification factors at higher modes, suggesting that the level of heterogeneity may be over-represented by the 2D random field models used in this study.
Recent surface-rupturing earthquakes in New Zealand have highlighted significant exposure and vulnerability of the road network to fault displacement. Understanding fault displacement hazard and its impact on roads is crucial for mitigating risks and enhancing resilience. There is a need for regional-scale assessments of fault displacement to identify vulnerable areas within the road network for the purposes of planning and prioritising site-specific investigations. This thesis employs updated analysis of data from three historical surface-rupturing earthquakes (Edgecumbe 1987, Darfield 2010, and Kaikoūra 2016) to develop an empirical model that addresses the gap in regional fault displacement hazard analysis. The findings contribute to understanding of • How to use seismic hazard model inputs for regional fault displacement hazard analysis • How faulting type and sediment cover affects the magnitude and spatial distribution of fault displacement • How the distribution of displacement and regional fault displacement hazard is impacted by secondary faulting • The inherent uncertainties and limitations associated with employing an empirical approach at a regional scale • Which sections of New Zealand’s roading network are most susceptible to fault displacement hazard and warrant site-specific investigations • Which regions should prioritise updating emergency management plans to account for post-event disruptions to roading. I used displacement data from the aforementioned historical ruptures to generate displacement versus distance-to-fault curves for various displacement components, fault types, and geological characteristics. Using those relationships and established relationships for along-strike displacement, displacement contours were generated surrounding active faults within the NZ Community Fault Model. Next, I calculated a new measure of 1D strain along roads as well as relative hazard, which integrated 1D strain and normalised slip rate data. Summing these values at the regional level identified areas of heightened relative hazard across New Zealand, and permits an assessment of the susceptibility of road networks using geomorphon land classes as proxies for vulnerability. The results reveal that fault-parallel displacements tend to localise near the fault plane, while vertical and fault-perpendicular displacements sustain over extended distances. Notably, no significant disparities were observed in off-fault displacement between the hanging wall and footwall sides of the fault, or among different surface geology types, potentially attributed to dataset biases. The presence of secondary faulting in the dataset contributes to increased levels of tectonic displacement farther from the fault, highlighting its significance in hazard assessments. Furthermore, fault displacement contours delineate broader zones around dip-slip faults compared to strike-slip faults, with correlations identified between fault length and displacement width. Road ‘strain’ values are higher around dip-slip faults, with notable examples observed in the Westland and Buller Districts. As expected, relative hazard analysis revealed elevated values along faults with high slip rates, notably along the Alpine Fault. A regional-scale analysis of hazard and exposure reveals heightened relative hazard in specific regions, including Wellington, Southern Hawke’s Bay, Central Bay of Plenty, Central West Coast, inland Canterbury, and the Wairau Valley of Marlborough. Notably, the Central West Coast exhibits the highest summed relative hazard value, attributed to the fast-slipping Alpine Fault. The South Island generally experiences greater relative hazard due to larger and faster-slipping faults compared to the North Island, despite having fewer roads. Central regions of New Zealand face heightened risk compared to Southern or Northern regions. Critical road links intersecting high-slipping faults, such as State Highways 6, 73, 1, and 2, necessitate prioritisation for site-specific assessments, emergency management planning and targeted mitigation strategies. Roads intersecting with the Alpine Fault are prone to large parallel displacements, requiring post-quake repair efforts. Mitigation strategies include future road avoidance of nearby faults, modification of road fill and surface material, and acknowledgement of inherent risk, leading to prioritised repair efforts of critical roads post-quake. Implementing these strategies enhances emergency response efforts by improving accessibility to isolated regions following a major surface-rupturing event, facilitating faster supply delivery and evacuation assistance. This thesis contributes to the advancement of understanding fault displacement hazard by introducing a novel regional, empirical approach. The methods and findings highlight the importance of further developing such analyses and extending them to other critical infrastructure types exposed to fault displacement hazard in New Zealand. Enhancing our comprehension of the risks associated with fault displacement hazard offers valuable insights into various mitigation strategies for roading infrastructure and informs emergency response planning, thereby enhancing both national and global infrastructure resilience against geological hazards.
