Sullivan, Australia
Sullivan, Australia

Time filter

Source Type

Green D.K.E.,University of New South Wales | Douglas K.,University of New South Wales | Mostyn G.,Pells Sullivan Meynink
Computers and Geotechnics | Year: 2015

This paper presents a new methodology to discretise random fields for reliability assessment by the Random Finite Element Method. The methodology maps a random field to an arbitrary triangular finite element mesh by local averaging. Derivations for the variance reduction and covariance between local averages of triangular areas using numeric integration are presented. The new methodology is compared against a published reliability assessment of a drained slope. The results matched expectations that not accounting for spatial variability will, in the case analysed, significantly underestimate reliability. A method to generate random fields using a form of Cholesky decomposition appropriate even for singular covariance matrices is presented and analysed. Finally, the derivations for the discretisation of random fields onto triangular meshes are presented for three dimensional tetrahedral elements. © 2015 Elsevier Ltd.


Bertuzzi R.,Pells Sullivan Meynink | Douglas K.,University of New South Wales | Mostyn G.,Pells Sullivan Meynink
International Journal of Rock Mechanics and Mining Sciences | Year: 2016

The aim of this paper is to expand the database linking the GSI and the Hoek-Brown criterion parameters mb, s and a. The database includes in situ stress failures of rock masses and of laboratory testing where rock mass quality has been interpreted. Assessment of the database showed that whilst the current relationships provide reasonable fits to the data, improved fits could be found by using equations developed by the authors. The new equations are similar to the existing ones. The substantial change is the one that is proposed between GSI and a, which allows the parameter a to vary between 0.5 for GSI =100 and 1.0 for GSI≤10. © 2016 Elsevier Ltd


Rogan A.,Pells Sullivan Meynink | Meynink B.,Mine Water Specialists International
The Art and Science of Water - 36th Hydrology and Water Resources Symposium, HWRS 2015 | Year: 2015

Accurate estimation of evaporation from a spilling 4 km2 shallow lake (Lake Namuk) on the equator in Kalimantan, Indonesia, was required to allow assessment of 1,000 year design drought water levels. Eight square kilometres of perennial contributing peripheral catchments prevented direct measurement of lake evaporation in the 3,860 mm/a rainfall environment. Few examples of climate parameter based estimates of evaporation could be found in literature generally for equatorial areas or for Kalimantan in particular. The climate at the lake was measured by an automatic weather station with: pyranometer (short wave solar radiation), wind speed (2 and 10 m), dew point temperature, air temperature, rainfall and water temperature near surface and at one metre depth. Lake Namuk has a surface water temperature consistently higher (∼3.2°C) than the ambient temperature. Several combination method algorithms and the equilibrium temperature method of Edinger et al (1974) were applied to estimate evaporation with observed water surface (equilibrium) temperatures used firstly for testing the algorithms and then wind function calibration. The minimal variations in seasonal temperatures allowed surface temperatures to be used as a proxy for equilibrium temperatures. This process demonstrated the need for back radiation estimates to be based on the surface rather than ambient temperature. Methods based on ambient temperature gave a 0.5 mm/day over-estimation of evaporation. A wind function that considered both convective and turbulent mixing processes was developed for the low wind environment. Once incorporated all algorithms produced similar evaporation estimates, in turn consistent with the limited available regional pan data. © 2015, Engineers Australia. All rights reserved.


Toh J.C.W.,Pells Sullivan Meynink | Pender M.J.,University of Auckland
Soil-Foundation-Structure Interaction - Selected Papers from the International Workshop on Soil-Foundation-Structure Interaction, SFSI 09 | Year: 2010

Current practice for earthquake-resistant shallow foundation design does not allow the foundation to fail in bearing, and does not consider the interaction of non-linear responses of the soil, foundation, and superstructure. Design tools exist that allow this interaction to be considered, and with the use of these tools, an alternative design approach that allows shallow foundations to yield during an earthquake becomes available. This paper uses examples to demonstrate the benefits of this alternative design approach. The approach enables the performance of the foundation to be balanced against that of the superstructure. Foundation and superstructure actions may be reduced significantly, whilst incurring only modest permanent foundation displacements. Broad suggestions of the type of criteria that might be required for the yielding foundation design approach are made. It is concluded that it may be preferable to develop criteria that describe performance aspects and system responses that should be considered, rather than prescribe firm quantitative criteria. © 2010 Taylor & Francis Group.


Strang D.R.,Pells Sullivan Meynink | Strang D.R.,University of Canterbury
Geological Society Engineering Geology Special Publication | Year: 2016

The Northern Wedge Failure (NWF) is a complex failure mass within Whitehall Quarry, Cambridge, New Zealand. Initiation of failure of the c. 500 000 m3 mass occurred during spring of 1988. Complete failure was expected following displacement recordings of 2.3-5.6 m a–1; however, this did not eventuate. By the mid 1990s, the failure mass was deemed stable. In 2010, an engineering geological investigation was conducted to assess the stability of the NWF at that time. The rock mass that failed comprises moderately dipping interbedded feldspathic sandstone and argillaceous mudstone below a thin alluvial cover. It is bound by two fault planes, both exhibiting up to 100 mm of gravelly clay fault gouge. X-ray diffraction analysis of the gouge indicated the presence of kaolinite and montmorillonite, and Atterberg Limit tests showed the plastic and liquid limits to be 31% and 51%, respectively. Electronic distance monitoring (EDM) was carried out over 12 months. Recorded velocities increased by 17% above the 21 mm/month baseline over the wetter months. Limit equilibrium analysis carried out using historic and recent field and laboratory observations input to the computer software SWedge estimated a Factor of Safety (FOS) of 0.97. Complete failure of the NWF could be initiated at any time by an extended period of heavy rainfall and/or unloading of the toe or the stripping of overburden. © 2016 The Author(s). Published by The Geological Society of London. All rights reserved.


Duran A.,Pells Sullivan Meynink
Australian Geomechanics Journal | Year: 2012

A study has been undertaken to confirm appropriate analysis methods applicable to design of granular waste materials and specifically that of "spoil piles" for open pit coal mining. The authors' experience with spoil pile stability is that the critical failure mechanism is a two wedge mechanism in accord with that highlighted in the literature. Analysis requires careful consideration of the combination of spoil shear strengths and the angle of the "rear" scarp adopted. Moreover, the literature indicates analyses need to take account of the inclined "inter-slice" within the mechanism. To highlight these considerations an example case was assessed utilising five limit equilibrium stability packages, two "numerical based" packages and a spreadsheet developed by the author. Several analyses are based on vertical slices to pose the question: what is the magnitude of error not using an inclined "inter-slice"? The results are presented as overall spoil pile angle indicated by the use of different software, stability method and assumptions on inclination of "rear" scarp and inter-slice. The example case highlights that, provided a rigorous stability method is utilised, for a given rear scarp assumption, the results of the analyses are insensitive to the software utilised. Therefore with appropriate judgement assessment of spoil pile stability should not be predicated by use of one particular software or stability method.


Baxter H.,Pells Sullivan Meynink
Australian Geomechanics Journal | Year: 2013

This paper presents two potential geohazards associated with Cenozoic detrital sequences in Western Australia's Pilbara region. The first considers carbonate rich calcrete layers, often reflecting the position of historic and current water tables. Observations of geochemistry of the calcrete layers compared with textural features in the resulting rock suggests a link between CaO, MgO and Loss on Ignition (LOI) abundance and the likelihood of cavities existing. Geohazards in the form of sinkhole or doline formation may result from changes to groundwater by dewatering for mining or town-water extraction. Contributing factors that increase this likelihood are high water flow, presence of dispersive soils in the blanketing layer, a geochemical signature of >20% CaO in the calcrete and commensurate thicknesses of the calcrete layer (and potential void space) and the blanketing layer. Relic rock slides have been recognised in several detrital valleys in the southern Pilbara and represent the second potential geohazard. The slides appear to represent a specific marker horizon in the detrital stratigraphy, attributed to a high rainfall, global warming climatic event in the Miocene. The slides consist of large rafted slabs/blocks of Archean bedrock with lesser cobbles and clasts of high strength rock. The voids between blocks are infilled with high plasticity, firm to hard kaolinitic clay, thought to be derived from subsequent lacustrine deposition. The preserved unit thickness varies from 10 to 80 m and can be buried by over 100 m of younger detritals. The slides present a unique geohazard to mining operations, not simply due to the variability in rock mass strength which can impact slope design. A high variability in void and matrix size and distribution is noted, though size and distribution of these zones is typically too small to be "mapped" by infill drilling. The size is however sufficiently large to cause trafficability issues on haul roads.


Smith A.,Pells Sullivan Meynink | Habte M.,Pells Sullivan Meynink
14th Australasian Tunnelling Conference 2011: Development of Underground Space, Proceedings | Year: 2011

Rock masses usually exhibit scale dependent mechanical properties primarily due to presence of discontinuities and their inhomogeneous nature. One of the essential preconditions for selection of appropriate strength and deformation parameters of a rock mass in relation to tunnel design is assessment of the scale of the joint network relative to the scale of excavation. In tunnel design practice, scale effects are estimated indirectly using small-scale unconfined compressive strength (UCS) tests and rock mass classification based empirical correlations. Large-scale UCS tests have the benefit of direct assessment of anisotropic and scale dependent properties of a rock mass. This paper presents results of a large-scale UCS test on a jointed siltstone. A 450 mm diameter sample with inclined bedding planes was cored and tested in uniaxial compression until failure with load and displacement measurements. The aim of the test is to determine strength and deformation parameters and correlate them with results of small-scale UCS tests of the same sample. The relationships between the small-scale and large-scale rock mass parameters were validated against published correlations from the literature.


Weir F.M.,Pells Sullivan Meynink
International Journal of Geotechnical Engineering | Year: 2015

This paper presents a review of both the advantages and the limitations associated with the collection of structural data using borehole imaging techniques. Case studies from excavations in hard rock in Australia and South East Asia are provided to compare with traditional methods of collecting structural data. Borehole imaging is a geophysical logging technique that produces a 360u-oriented image of the borehole wall. Initially developed by the petroleum industry, it is rapidly becoming a standard investigation tool for geological, hydrogeological, and geotechnical investigations. Borehole imaging is widely used to obtain reliable structural data, detect bedding orientation, for breakout analysis, for casing inspection, and for high-resolution caliper measurements. In excavated slopes, the orientation and characteristics of structures play a crucial role in slope stability. A significant advantage of borehole imaging is the ability to obtain reliable structural data in zones of core loss or in poor rock masses where traditional orientation techniques are difficult. While borehole imaging provides many advantages compared to traditional core orientation techniques, it is not a ''silver bullet'' and like any method, in the hands of the inexperienced may be misleading. © 2015 W. S. Maney & Son Ltd.


Cammack R.,Pells Sullivan Meynink
Geological Society Engineering Geology Special Publication | Year: 2016

Successful open-pit slope design depends on the formulation of a representative geotechnical model. This paper presents the development of an engineering geological rock-mass model as part of a feasibility geotechnical study for the Tampakan copper/gold open-pit mine in the Philippines. This porphyry deposit has been subjected to several cycles of volcanism, chemical alteration and tectonic disturbance resulting in a complex, highly fractured and brecciated rock mass. The model requirements for design were identified early in the project and a unique geotechnical logging system was developed to effectively capture the relevant rock-mass characteristics. © 2016 The Author(s). Published by The Geological Society of London. All rights reserved.

Loading Pells Sullivan Meynink collaborators
Loading Pells Sullivan Meynink collaborators