Recently developed performance-based earthquake engineering framework, such as one provided by PEER (Deierlein et al. 2003), assist in the quantification in terms of performance such as casualty, monetary losses and downtime. This opens up the opportunity to identify cost-effective retrofit/rehabilitation strategies by comparing upfront costs associated with retrofit with the repair costs that can be expected over time. This loss assessment can be strengthened by learning from recent earthquakes, such as the 2010 Canterbury and 2016 Kaikoura earthquakes. In order to investigate which types of retrofit/rehabilitation strategies may be most cost-effective, a case study building was chosen for this research. The Pacific Tower, a 22-storey EBF apartment located within the Christchurch central business district (CBD), was damaged and repaired during the 2010 Canterbury earthquake series. As such, by taking hazard levels accordingly (i.e. to correspond to the Christchurch CBD), modelling and analysing the structure, and considering the vulnerability and repair costs of its different components, it is possible to predict the expected losses of the aforementioned building. Using this information, cost-effective retrofit/rehabilitation strategy can be determined. This research found that more often than not, it would be beneficial to improve the performance of valuable non-structural components, such as partitions. Although it is true that improving such elements will increase the initial costs, over time, the benefits gained from reduced losses should be expected to overcome the initial costs. Aftershocks do increase the predicted losses of a building even in lower intensities due to the fact that non-structural components can get damaged at such low intensities. By comparing losses computed with and without consideration of aftershocks for a range of historical earthquakes, it was found that the ratio between losses due to main shock with aftershocks to the losses due to the main shock only tended to increase with increasing main shock magnitude. This may be due to the fact that larger magnitude earthquakes tend to generate larger magnitude aftershocks and as those aftershocks happen within a region around the main shock, they are more likely to cause intense shaking and additional damage. In addition to this observation, it was observed that the most significant component of loss of the case study building was the non-structural partition walls.
he 2016 Building (Earthquake Prone Building) Amendment Act aims to improve the system for managing earthquake-prone buildings. The proposed changes to the Act were precipitated by the Canterbury earthquakes, and the need to improve the seismic safety of New Zealand’s building stock. However, the Act has significant ramifications for territorial authorities, organisations and individuals in small New Zealand towns, since assessing and repairing heritage buildings poses a major cost to districts with low populations and poor rental returns on commercial buildings.
In September 2010 and February 2011, the Canterbury region experienced devastating earthquakes with an estimated economic cost of over NZ$40 billion (Parker and Steenkamp, 2012; Timar et al., 2014; Potter et al., 2015). The insurance market played an important role in rebuilding the Canterbury region after the earthquakes. Homeowners, insurance and reinsurance markets and New Zealand government agencies faced a difficult task to manage the rebuild process. From an empirical and theoretic research viewpoint, the Christchurch disaster calls for an assessment of how the insurance market deals with such disasters in the future. Previous studies have investigated market responses to losses in global catastrophes by focusing on the insurance supply-side. This study investigates both demand-side and supply-side insurance market responses to the Christchurch earthquakes. Despite the fact that New Zealand is prone to seismic activities, there are scant previous studies in the area of earthquake insurance. This study does offer a unique opportunity to examine and document the New Zealand insurance market response to catastrophe risk, providing results critical for understanding market responses after major loss events in general. A review of previous studies shows higher premiums suppress demand, but how higher premiums and a higher probability of risk affect demand is still largely unknown. According to previous studies, the supply of disaster coverage is curtailed unless the market is subsidised, however, there is still unsettled discussion on why demand decreases with time from the previous disaster even when the supply of coverage is subsidised by the government. Natural disaster risks pose a set of challenges for insurance market players because of substantial ambiguity associated with the probability of such events occurring and high spatial correlation of catastrophe losses. Private insurance market inefficiencies due to high premiums and spatially concentrated risks calls for government intervention in the provision of natural disaster insurance to avert situations of noninsurance and underinsurance. Political economy considerations make it more likely for government support to be called for if many people are uninsured than if few people are uninsured. However, emergency assistance for property owners after catastrophe events can encourage most property owners to not buy insurance against natural disaster and develop adverse selection behaviour, generating larger future risks for homeowners and governments. On the demand-side, this study has developed an intertemporal model to examine how demand for insurance changes post-catastrophe, and how to model it theoretically. In this intertemporal model, insurance can be sought in two sequential periods of time, and at the second period, it is known whether or not a loss event happened in period one. The results show that period one demand for insurance increases relative to the standard single period model when the second period is taken into consideration, period two insurance demand is higher post-loss, higher than both the period one demand and the period two demand without a period one loss. To investigate policyholders experience from the demand-side perspective, a total of 1600 survey questionnaires were administered, and responses from 254 participants received representing a 16 percent response rate. Survey data was gathered from four institutions in Canterbury and is probably not representative of the entire population. The results of the survey show that the change from full replacement value policy to nominated replacement value policy is a key determinant of the direction of change in the level of insurance coverage after the earthquakes. The earthquakes also highlighted the plight of those who were underinsured, prompting policyholders to update their insurance coverage to reflect the estimated cost of re-building their property. The survey has added further evidence to the existing literature, such as those who have had a recent experience with disaster loss report increased risk perception if a similar event happens in future with females reporting a higher risk perception than males. Of the demographic variables, only gender has a relationship with changes in household cover. On the supply-side, this study has built a risk-based pricing model suitable to generate a competitive premium rate for natural disaster insurance cover. Using illustrative data from the Christchurch Red-zone suburbs, the model generates competitive premium rates for catastrophe risk. When the proposed model incorporates the new RMS high-definition New Zealand Earthquake Model, for example, insurers can find the model useful to identify losses at a granular level so as to calculate the competitive premium. This study observes that the key to the success of the New Zealand dual insurance system despite the high prevalence of catastrophe losses are; firstly the EQC’s flat-rate pricing structure keeps private insurance premiums affordable and very high nationwide homeowner take-up rates of natural disaster insurance. Secondly, private insurers and the EQC have an elaborate reinsurance arrangement in place. By efficiently transferring risk to the reinsurer, the cost of writing primary insurance is considerably reduced ultimately expanding primary insurance capacity and supply of insurance coverage.
Semi-empirical models based on in-situ geotechnical tests have become the standard of practice for predicting soil liquefaction. Since the inception of the “simplified” cyclic-stress model in 1971, variants based on various in-situ tests have been developed, including the Cone Penetration Test (CPT). More recently, prediction models based soley on remotely-sensed data were developed. Similar to systems that provide automated content on earthquake impacts, these “geospatial” models aim to predict liquefaction for rapid response and loss estimation using readily-available data. This data includes (i) common ground-motion intensity measures (e.g., PGA), which can either be provided in near-real-time following an earthquake, or predicted for a future event; and (ii) geospatial parameters derived from digital elevation models, which are used to infer characteristics of the subsurface relevent to liquefaction. However, the predictive capabilities of geospatial and geotechnical models have not been directly compared, which could elucidate techniques for improving the geospatial models, and which would provide a baseline for measuring improvements. Accordingly, this study assesses the realtive efficacy of liquefaction models based on geospatial vs. CPT data using 9,908 case-studies from the 2010-2016 Canterbury earthquakes. While the top-performing models are CPT-based, the geospatial models perform relatively well given their simplicity and low cost. Although further research is needed (e.g., to improve upon the performance of current models), the findings of this study suggest that geospatial models have the potential to provide valuable first-order predictions of liquefaction occurence and consequence. Towards this end, performance assessments of geospatial vs. geotechnical models are ongoing for more than 20 additional global earthquakes.
Between 2010 and 2011, Canterbury experienced a series of four large earthquake events with associated aftershocks which caused widespread damage to residential and commercial infrastructure. Fine grained and uncompacted alluvial soils, typical to the Canterbury outwash plains, were exposed to high peak ground acceleration (PGA) during these events. This rapid increase in PGA induced cyclic strain softening and liquefaction in the saturated, near surface alluvial soils. Extensive research into understanding the response of soils in Canterbury to dynamic loading has since occurred. The Earthquake Commission (EQC), the Ministry of Business and Employment (MBIE), and the Christchurch City Council (CCC) have quantified the potential hazards associated with future seismic events. Theses bodies have tested numerous ground improvement design methods, and subsequently are at the forefront of the Canterbury recovery and rebuild process. Deep Soil Mixing (DSM) has been proven as a viable ground improvement foundation method used to enhance in situ soils by increasing stiffness and positively altering in situ soil characteristics. However, current industry practice for confirming the effectiveness of the DSM method involves specific laboratory and absolute soil test methods associated with the mixed column element itself. Currently, the response of the soil around the columns to DSM installation is poorly understood. This research aims to understand and quantify the effects of DSM columns on near surface alluvial soils between the DSM columns though the implementation of standardised empirical soil test methods. These soil strength properties and ground improvement changes have been investigated using shear wave velocity (Vs), soil behaviour and density response methods. The results of the three different empirical tests indicated a consistent improvement within the ground around the DSM columns in sandier soils. By contrast, cohesive silty soils portrayed less of a consistent response to DSM, although still recorded increases. Generally, within the tests completed 50 mm from the column edge, the soil response indicated a deterioration to DSM. This is likely to be a result of the destruction of the soil fabric as the stress and strain of DSM is applied to the un‐mixed in situ soils. The results suggest that during the installation of DSM columns, a positive ground effect occurs in a similar way to other methods of ground improvement. However, further research, including additional testing following this empirical method, laboratory testing and finite 2D and 3D modelling, would be useful to quantify, in detail, how in situ soils respond and how practitioners should consider these test results in their designs. This thesis begins to evaluate how alluvial soils tend to respond to DSM. Conducting more testing on the research site, on other sites in Christchurch, and around the world, would provide a more complete data set to confirm the results of this research and enable further evaluation. Completing this additional research could help geotechnical DSM practitioners to use standardised empirical test methods to measure and confirm ground improvement rather than using existing test methods in future DSM projects. Further, demonstrating the effectiveness of empirical test methods in a DSM context is likely to enable more cost effective and efficient testing of DSM columns in future geotechnical projects.
Capacity design and hierarchy of strength philosophies at the base of modern seismic codes allow inelastic response in case of severe earthquakes and thus, in most traditional systems, damage develops at well-defined locations of reinforced concrete (RC) structures, known as plastic hinges. The 2010 and 2011 Christchurch earthquakes have demonstrated that this philosophy worked as expected. Plastic hinges formed in beams, in coupling beams and at the base of columns and walls. Structures were damaged permanently, but did not collapse. The 2010 and 2011 Christchurch earthquakes also highlighted a critical issue: the reparability of damaged buildings. No methodologies or techniques were available to estimate the level of subsequent earthquakes that RC buildings could still sustain before collapse. No repair techniques capable of restoring the initial condition of buildings were known. Finally, the cost-effectiveness of an eventual repair intervention, when compared with a new building, was unknown. These aspects, added to nuances of New Zealand building owners’ insurance coverage, encouraged the demolition of many buildings. Moreover, there was a perceived strong demand from government and industry to develop techniques for assessing damage to steel reinforcement bars embedded in cracked structural concrete elements. The most common questions were: “Have the steel bars been damaged in correspondence to the concrete cracks?”, “How much plastic deformation have the steel bars undergone?”, and “What is the residual strain capacity of the damaged bars?” Minimally invasive techniques capable of quantifying the level and extent of plastic deformation and residual strain capacity are not yet available. Although some studies had been recently conducted, a validated method is yet to be widely accepted. In this thesis, a least-invasive method for the damage-assessment of steel reinforcement is developed. Based on the information obtained from hardness testing and a single tensile test, it is possible to estimate the mechanical properties of earthquake-damaged rebars. The reduction in the low-cycle fatigue life due to strain ageing is also quantified. The proposed damage assessment methodology is based on empirical relationships between hardness and strain and residual strain capacity. If damage is suspected from in situ measurements, visual inspection or computer analysis, a bar may be removed and more accurate hardness measurements can be obtained using the lab-based Vickers hardness methodology. The Vickers hardness profile of damaged bars is then compared with calibration curves (Vickers hardness versus strain and residual strain capacity) previously developed for similar steel reinforcement bars extracted from undamaged locations. Experimental tests demonstrated that the time- and temperature-dependent strain-ageing phenomenon causes changes in the mechanical properties of plastically deformed steels. In particular, yield strength and hardness increases, whereas ductility decreases. The changes in mechanical properties are quantified and their implications on the hardness method are highlighted. Low-cycle fatigue (LCF) failures of steel reinforcing bars have been observed in laboratory testing and post-earthquake damage inspections. Often, failure might not occur during a first seismic event. However, damage is accumulated and the remaining fatigue life is reduced. Failure might therefore occur in a subsequent seismic event. Although numerous studies exist on the LCF behaviour of steel rebars, no studies had been conducted on the strain-ageing effects on the remaining fatigue life. In this thesis, the reduction in fatigue life due to this phenomenon is determined through a number of experimental tests.
