Researcher's name | Title of the project | Supervisor(s) | Software/Tools | Constitutive model | University/Organisation involved | Funding source |
|---|---|---|---|---|---|---|
Arman Kamalzadeh | Further Numerical Modelling of the Seismic Response of Retaining Structures | Michael Pender | OpenSees, DEEPSOIL, MATLAB, GiD | Manzari and Dafalias | University of Auckland/ NeSI | The University of Canterbury Quake Centre (UCQC) |
| Abstract: Infrastructure facilities are the major poles of development. In the case of geological and environmental hazards, such as earthquakes and landslides, retaining structures can protect against these threats and shield roadways, rail facilities, commercial buildings and residential properties from ground movement and landslide debris. Furthermore, basements are mandatory for city-centre high-rise buildings to provide parking facilities. Designers acknowledge that the pressures exerted on basement walls during earthquake loading need better definition. The current design approach for retaining structures subject to earthquake can be traced back to observations from 1g shaking table tests performed in Japan in the 1920s. Recent research has shed light on scaling problems associated with 1g tests and centrifuge testing has been used to clarify these issues. Some centrifuge tests conclude that present design approaches for retaining walls may result in over-design and in rare cases under-design. New Zealand is prone to several natural hazards, which can put the resilience of its infrastructures at stake causing serious economic losses and possible casualties. These hazards range from earthquake-induced slope instability, liquefaction, and heavy precipitation leading to loss of strength in soils. In addition, with more than 75 percent of our imports and exports being shipped by means of the ocean, port facilities containing retaining structures play an important role in the New Zealand economy. Therefore, a thorough investigation of the retaining structures of New Zealand contributes to making our infrastructure more resilient. Furthermore, current experimental research worldwide especially at The University of Berkeley by Sitar’s research group (Mikola, Candia, & Sitar, 2014, 2016; Sitar & Al Atik, 2008; Sitar, Mikola, & Candia, 2012; Wagner, Candia, Mikola, & Sitar, 2017; Wagner & Sitar, 2016) and recent numerical study by Chin et al. (2016) in New Zealand have shown current standard design method may result in significant overestimation of equivalent static seismic force. This over-conservative approach can lead to an economic burden as severe as future catastrophic failures. By localizing the study for New Zealand conditions (typical retaining structures, soil type and seismic zones) this research seeks to investigate on the integrity of the previous works with the hope to find and embrace the probable factors of this possible overestimation and the extent of it. In this research, I am using the OpenSees (McKenna, Mazzoni, & Fenves, 2013) finite element analyses (FEM) software. The appeal of finite element studies is that they can be validated using results from centrifuge models and then used to perform parametric studies on retaining structure types and configurations used in New Zealand. This study intends to look at the design of more frequent and practical retaining walls, which has been controversial among geotechnical professionals. These walls are the ones constructed in smaller scales for residential and commercial buildings such as stepped, terraced and embedded pole retaining walls. It is hoped to present a guideline or encourage further research to do so, for designing these types of walls. The scope of this research will be of interest of the government, civil engineering firms and research institutes, the intended outcome could attract funds for experimental studies and for further investigations with the ultimate aim of modifying standards and guidelines leading to more resilient and more cost-effective infrastructures and buildings around New Zealand. | ||||||
Caleb Gasston | Structural Controls on Landscape Evolution: Surface Fault Rupture and Landslides, M7.8 Kaikoura Earthquake 2016 | Martin Brook, Julie Rowland, Chris Massey | Python, RS2, ArcGIS | - | University of Auckland | MBIE |
| Abstract: The 14 November 2016 M w 7.8 Kaikoura earthquake on New Zealand’s South Island had an along-strike length of 165 km of ground and seafloor surface rupture. Deformation involved a large number of faults with differing styles (including strike-slip and oblique-slip faults with both dextral- and sinistral-slip senses), fault lengths, and slip magnitudes. This generated more than 10,000 sub aerial landslides over a total area of c. 10,000 sq km , most of which were concentrated within a zone of about 3600 sq km. This research investigates the mechanisms of failure of 4 of the largest landslides which were generated by the earthquake: the Seafront, Leader, Stanton and Hapuku Landslides. | ||||||
Zaid Rana | Dynamic Site Characterization of Hauraki Plains | Liam Whotherspoon, Jennifer Eccles | MATLAB, Python, Geopsy | - | University of Auckland | - |
| Abstract: Site Characterization of Hauraki Plains using passive seismic (HVSR), deriving site periods and collating with existing gravity and active seismic data to better understand basement structure and improve understanding of the ground shaking during earthquakes in the Hauraki region, both in past and future events. | ||||||
Sayed Hessam Bahmani | Discrete Element Modelling Of Crushable Pumice Sands | Rolando Orense, Michael Pender | Python | - | University of Auckland | - |
| Abstract: Pumice deposits are found in many areas in the central part of North Island. Previous studies have shown that pumice particles are highly crushable, compressible and lightweight as a result of their vesicular nature. Some researchers showed that conventional relationships between relative density, and resistance from cone penetration tests (CPT) were not valid for pumice due to breakage of particles during penetration. To address the challenges in characterizing crushable pumice sands only by means of extended empirical observations and laboratory tests are possible, but they are slow and costly. Discrete Element Method (DEM) can be a promising solution to provide insight into the behaviour of this granular material at both micro- and macro-level. While there are limited DEM results available for other crushable soils, they are not accurate for pumice deposits. In this research, attempts will be made to model the 3D shape of sand particles to obtain micromechanical insight into the behaviour of the crushable pumice particles. Using the results, DEM simulations of monotonic triaxial tests will be performed to investigate the mechanical behaviour of pumice particles. This will be supplemented by constructing a 3D DEM model that can reproduce physical calibration chamber tests of CPT tests on pumiceous soils. | ||||||
Yin-Chin Wang | Water retention properties and unsaturated permeability of soil mixed with natural fibres | Ryan Yan | GeoStudio (Seep/W), MathCAD | - | University of Auckland | - |
| Abstract: The hydraulic behaviour of soil-fibre mixture, Auckland residual soil mixed with flax fibres, were studied by experimental and numerical methods. The aims of experiments are to measure and estimate the soil-water characteristic curve (SWCC) and unsaturated permeability. Then, they can be fitted in Fredlund and Xing’s SWCC equation and Leong and Rahardjo’s permeability equation for numerical modelling, which is a one-dimensional numerical modelling in Seep/W. This Auckland residual soil is silty and clayey, and it is classified as ML in the unified soil classification system. Shrinkage or swelling is relatively small. The saturated permeability is 2.2x10^-6 m/s, which is close to the permeability of silts. In terms of SWCC experiments, pressure plate extractor was used for measuring matric suction less than 1500 kPa. Combined with the experimental results of relative humidity method, that can measure total suction up to 10^5 kPa, the SWCC can almost be tested in the entire range. It was found that this soil-fibre mixture has bimodal SWCC, which has two sharp changes in SWCC corresponding to macro and micro pore sizes of soils. Unsaturated permeability was estimated by soil column experiment for the suction less than 100 kPa. It was found that the unsaturated permeability has a significant reduction corresponding to first entry value, from 2.2x10^-6 m/s at its fully saturated state to 1.3x10^-11 m/s at the suction of 100 kPa. This large reduction in permeability leads to that topsoils became dry but relatively impermeable. This feature was also identified by a numerical model for validation. The features of the soil-fibre mixture were found that the mixture has bimodal SWCC and a significant reduction in permeability corresponding to first air entry value. The hydraulic behaviours, such as slowly increased suction, a large difference in volumetric water content between two depths, etc., are mainly governed by these features. In addition, both of experimental and numerical results are consistent with each other. It provides an evidence that the estimated SWCC and permeability from experiments are reasonable and the numerical model is capable to identify the hydraulic behaviour of this soil-fibre mixture. | ||||||
Farbod Yarmohammadi | Numerical Modelling of Structure-soil-structure Interaction Involving Liquefaction | Connor Hayden, Liam Whotherspoon, | FLAC, MATLAB | PM4SAND | University of Auckland | Earthquake Commission (EQC) |
| Abstract: Soil-structure interaction (SSI) is integral to the seismic analysis and design of the important structures. But, interaction between multiple adjacent structures supported on the same soil domain through structure-soil-structure interaction (SSSI) is generally ignored. This project will first involve validating numerical models against centrifuge tests involving liquefaction. Next analysing a wide range of parameters for adjacent structures, input ground motions, and soil profiles will lead to more comprehensive and robust conclusions on liquefaction-induced SSSI effects. Since This level of numerical modeling is unlikely to be performed in industry at present, the study will culminate in developing simplified methods for use in more routine engineering practice, enabling this research to have rapid effects on improving earthquake resilience. | ||||||
Tahoura Khansari | Ground Motion Selection for Geotechnical Systems | Connor Hayden, Liam Whotherspoon, | FLAC, DEEPSOIL | PM4SAND | University of Auckland | University of Auckland Scholarship |
| Abstract: An important part of the design of the structures and geotechnical systems is to select an appropriate input ground motion. 1-D site response analysis of some number of sites in Christchurch could help to understand more about the ground motions. The deconvolved motions applied to the soil model, to verify that against the pore pressure transducers installed there during the Canterbury Earthquake Sequences 2011. The pore pressure transducers provided a unique situation so that the liquefaction evaluation in these different sites can be analysed. Also, one of the important characteristics of the ground motions especially in the near-fault regions is pulse-like motions. The spectral matching should be done in a manner that the input ground motion preserves the pulse motions. Furthermore, there are different methods to predict future earthquakes. One of the state-of-the-art of these methods is physics-based ground motion simulation. This research is going to compare the physics-based ground motion simulations to the real recorded motions to find out if there is any systematic difference between them and the way that they can affect the geotechnical systems. | ||||||
Majid Zakerinia | New Zealand Tools and Procedures for Seismic Effective Stress Analysis | Connor Hayden, Christopher McGann | OpenSees, FLAC | PM4SAND, Stress-Density Model | University of Auckland and QuakeCoRE | QuakeCoRE |
| Abstract: To create computational and visualisation tools in which seismic effective stress analysis for the assessment of full system response is as simple for the user to run and evaluate as comparable simplified analytical tools (e.g. equivalent linear site response analysis, liquefaction potential/consequence evaluation). The end result will incorporate state-of-the-art soil constitutive models capable of 1D/2D/3D analysis, and consideration for the interface elements, boundary conditions, loading conditions, and other numerical features necessary for analysis of the seismic response of earth structures and soil-structure interaction problems with nonlinear super- and sub-structure elements. The project will utilise existing QuakeCoRE expertise in advanced simulation tools such as OpenSees and FLAC, and will be integrated with HPC resources to maximise computational efficiency. The cross-flagship nature of this project will be embraced such that the final product will allow for increased understanding in critical QuakeCoRE research areas such as the use of simulated ground motions in engineering practice, the assessment of existing/heritage structures, low-damage technologies, and distributed infrastructure, and the effects of damage to these systems on society. The proposed tool will be one of the principal state-of-the-art tools to be utilized in earthquake scenario studies for NZ. | ||||||
Ribu Dhakal | Liquefaction Assessment of Reclaimed Soils | Misko Cubrinovski, Jonathan D. Bray | MATLAB, DIANA J | Stress-Density Model | University of Canterbury | University of Canterbury + QuakeCoRE |
| Abstract: The 14 November 2016 Mw7.8 Kaikoura Earthquake caused widespread liquefaction in CentrePort, Wellington, resulting in substantial lateral ground movement, global and differential settlements, and large volumes of soil ejecta. Pile-supported wharves and buildings on shallow and deep foundations were damaged severely. Following the earthquake, a subsurface exploration program was executed consisting of CPTs and surface wave testing to characterize the thick end-dumped gravelly fills and hydraulically-placed dredged reclamations. This study uses the extensive subsurface data to produce detailed subsurface soil profiles for the Port of Wellington, and then assess liquefaction susceptibility and triggering of the reclamations. The complex subsurface soil composition for the gravelly fills contain mixtures of gravels, sands, and some non-plastic fines. The hydraulic fills were constructed recently by slurry deposition of seabed materials with little compaction. CPT data from 47 sites and grain-size compositions of the fills are used to produce detailed characterization of the subsurface conditions. The geotechnical data provide insights on the role of sand and silt-sized fractions in the matrix of well-graded gravelly fills, as well as their influence on the cone penetration resistance and dynamic response of the soil. The geotechnical investigations also suggest this combination of soil composition, minimal compaction effort and little cementation effects due to its relatively young age played a key role in the poor liquefaction performance of the hydraulic fills. Ranges for the penetration resistance and soil behavior type index values are provided for the characteristic soil layers at the port. The spatial distribution of the end-dumped gravelly fills and sandy hydraulic fills at the port are presented. The liquefaction susceptibility and triggering potential of the different soil deposits within the port are assessed and compared to the observed field performance. State of the practice simplified procedures are used to evaluate liquefaction triggering and ground damage indices. Nonlinear seismic effective stress analyses (ESA) using a state-concept based constitutive model calibrated to the simplified procedures are also conducted to investigate the ground response of the port reclamations in detailed time history analyses. In these analyses, excess pore water pressure generation and dissipation, and water flow effects are modeled. The results from the ESA and simplified procedures are compared to detailed observations of liquefaction manifestations and induced land and building damage. Emphasis is placed on the investigation of the robustness of conventional procedures in their ability to capture the liquefaction resistance and cyclic response of the reclaimed soil. | ||||||
Polly Guan | Computational Modelling of Aviemore Embankment Dam | Christopher McGann, Mark Foster | FLAC, PLAXIS | Drucker-Prager Model, Mohr-Coulomb Model, and Bilinear Strain-hardening/softening | University of Canterbury | - |
| Abstract: The objective of this research's dissertation is to evaluate the soil behavior, understand the embankment dam deformation and embody computational modelling in static analysis of an existing, operational embankment dam. Secondary intention is to adopt the currently available numerical modelling packages and the sophistication of the constitutive models on which they are based, making them useful tolls for modelling embankment dam behavior and developing algorithms for detailed geometric models to understand and model the complex soil embankment behaviour. | ||||||
Kevin Foster | A Vs30 map for New Zealand | Brendon Bradley, Christopher McGann, Liam Wotherspoon | R, GDAL | - | University of Canterbury | Marsden Fund grant, QuakeCoRE |
| Abstract: A time-averaged 30-meter depth shear wave velocity (Vs30) map is developed for New Zealand as a weighted combination of a geology-based and a terrain-based model. A Bayesian updating process allows local Vs30 measurements to control model estimates where data exist, and uses model estimates developed for other parts of the world where local data are sparse or nonexistent. Geostatistical interpolation is performed on the geology- and terrain-based models using local Vs30 measurements to constrain the model in the vicinity of data. Conventional regression kriging is compared with a flexible multivariate normal (MVN) approach that allows for arbitrary assumptions regarding measurement uncertainty at each data location. A modification to the covariance structure in the MVN application allows for more realistic estimates by reducing undesirable extrapolation across geologic boundaries. The results of kriging and MVN approaches are compared. The geology- and terrain-based MVN models are combined to produce a final model suitable for engineering applications. 100m resolution map outputs are publicly available. | ||||||
Ananth Balachandra | Validation of numerical simulations of SSI for buildings on liquefiable deposits using centrifuge experiments | Connor Hayden, Christopher McGann, Liam Wotherspoon | OpenSees, FLAC | PM4Sand and PDMY02 | University of Auckland, University of Canterbury | QuakeCoRE Flagship 2 |
| Abstract: The T4.6-40 centrifuge experiment undertaken as part of the NEESR Seismic Performance Assessment in Dense Urban Environments project has been used to validate 1D numerical simulation of free field response and 2D numerical simulation of SSI response of isolated buildings on liquefiable deposits. The 1D numerical simulations were developed using FLAC and the PM4Sand constitutive soil model and OpenSEES and the PDMY02 constitutive soil model. The 2D numerical simulations were developed using FLAC and PM4Sand. | ||||||
Yu Wang | Numerical study on the seismic behavior of inclined piles in liquefiable soils | Rolando Orense | OpenSees, MATLAB, GiD | PDMY, Manzari-Dafalias | University of Auckland | |
| Abstract: With extensive damage to pile foundations observed during previous earthquakes, vertical piles have been proven to be weak in the horizontal direction and susceptible to failures especially in the liquefied ground. Theoretically speaking, inclined piles, also called raked or batter piles, are believed to have larger lateral resistance than vertical piles by transferring part of the lateral force through axial compression. There are also numerous case studies, experimental investigations and numerical simulations announcing the benefits of inclined piles. However, failures of wharves and bridge foundations resulted in the discouragement of inclined piles in the engineering design. The beneficial or detrimental effects of inclined piles in the liquefiable ground are still not well established. This research investigates the seismic performance of inclined piles in liquefiable soils through numerical approaches. Firstly, a comprehensive 3D Finite Element (FE) model of the soil-pile-superstructure (SPSI) system will be established with special concern about the liquefaction behavior of soil and the slippage and separation mechanisms at the soil-pile interface. Detailed validations will be carried out based on static and dynamic laboratory experiments and field studies in the literature. Secondly, the wide-spread Beam-on-Winkler-Foundation (BNWF) method will be modified to account for the effects of pile inclination and soil liquefaction based on the validated FE model. Finally, an approximate design procedure for inclined piles will be proposed based on the modified BNWF method and case studies will be conducted within the probabilistic framework of the Performance-Based Earthquake Engineering methodology. | ||||||
Overview
Content Tools
Activity