ocean-structure-seabed interaction o · pdf fileocean-structure-seabed interaction: o-tube...

Australian Geomechanics Vol 49 No 4 December 2014 157

OCEAN-STRUCTURE-SEABED INTERACTION: O-TUBE MODELLING OF PIPELINE STABILITY

D.J. White, S. Draper, L. Cheng, H. An, H. Mohr and S. LeckieARC Centre of Excellence for Geotechnical Science and Engineering, The University of Western Australia, Crawley, WA, Australia

ABSTRACT A key facility used by researchers in the Centre for Geotechnical Science and Engineering is the set of O-tube flumes established at UWA. These flumes are a unique concept that have been developed at UWA to allow simulation of ocean-structure-seabed interactions using realistic metocean and geotechnical conditions. The large, small and mini O-tube flumes allow seabed flows to be simulated at a range of scales, including full scale modelling of small subsea pipelines. Interactions between mobile sediments and infrastructure can be monitored. This paper describes the O-tube facilities and uses example results to illustrate the range of problems that can be tackled. A key outcome from the O-tube research program has been a new methodology to assess the stability of pipelines on mobile seabeds, which is a common design requirement offshore Australia. This methodology is allowing more efficient and cost-effective design of the pipelines that are the vital arteries of Australia’s offshore oil and gas infrastructure.

1 INTRODUCTIONThe interaction of environmental flows in the ocean with the seabed and subsea structures is an important area of offshore engineering, attracting significant attention in the design of pipelines (Sumer & Fredsøe, 2002; Bonjean et al.,2008) and foundations (Whitehouse et al., 2011; Nielsen et al., 2011).

This paper summarises a facility and research program underway at UWA which is focused on better understanding and quantifying ocean-structure-seabed interaction behaviour. The aim is to improve the design of seabed structures, through a multi-disciplinary approach that gives a balanced emphasis between the hydrodynamic, structural and geotechnical aspects of the behaviour.

When a structure is placed on the seabed it is subjected to hydrodynamic forces induced by current flows, such as tidal and circulation currents, and oscillatory flows induced by internal and surface waves. The structure also disturbs these flows and obstructs existing seabed processes such as sediment transport. Under certain conditions this can lead to significant modifications to the topography of the seabed around the structure, potentially affecting the hydrodynamic forces on the structure, the geotechnical capacity and therefore the stability.

Because of the interaction between the structure and the environmental flows, fluid-structure-seabed interaction is inherently non-linear. It cannot, for instance, be investigated through linear superposition of individual interaction processes such as fluid-structure, fluid-seabed and structure-seabed. A proper engineering design of the structure must consider the full interaction between environmental flow, seabed and the structure, allowing for changes in the seabed topography induced by the installation of the structure.

Demand for pipelines and subsea infrastructure has increased in recent years offshore Australia, as new oil and gas projects enter the design and construction phases. Meanwhile, Australia’s earliest oil and gas developments such as the North West Shelf Venture are entering life extension phases, in which the long term effects of ocean-structure-seabed interaction are requiring careful assessment. Ocean-structure-seabed interaction is particular important in the Australian North West Shelf (NWS) environment due to the onerous hydrodynamic conditions due to cyclones and solitons.

Conventional engineering design methods to assess pipeline on-bottom stability, for example, still assume that the pipeline is resting on an immobile seabed. Damgaard and Palmer (2001) showed that this assumption is fundamentally flawed and suggested that a more realistic approach is to consider the pipeline to be resting on an unstable seabed, in which the seabed can move due to sediment transport and/or wave-induced liquefaction. Otherwise, ignoring sediment mobility can typically lead to overly conservative and costly pipeline stability design. A radical new approach to pipeline stability design has been one of the key aspects of the O-tube research program, linking also with research using numerical modelling and interpretation of field observations.

2 SEABED CONDITIONS Near bed velocities in the offshore environment mobilise sediment and present a destabilising force to pipelines and subsea structures. In the field they may originate from tidal currents, wind driven currents, waves and wave-induced currents, with each component varying with water depth and location.


On the NWS of Australia the geometry of the shelf acts to amplify the semidiurnal tidal constituents, leading to median peak near-bed tidal currents ranging from 0.1 to 0.3 m/s at a depth as large as 100 m across the shelf (Holloway, 1983). In addition to tidal currents, strong tidal forcing on the NWS, combined with density stratification through the water column, leads to significant internal wave activity along the edge of the inner/middle shelf. This can lead to soliton currents orientated inline with the shelf slope, which reach near bed velocities in excess of approximately 0.5 to 1.0 m/s (over 5 – 10 year return periods), and are most significant in the summer (Van Gastel et al., 2009).

Seasonal winds and climate conditions on the NWS also influence the local sea waves, whilst swell waves originate typically from the Southern Ocean swells, travelling from the southwest throughout the year. Seastates due to Southern swells and local seas tend to be more severe during winter when wave heights can reach up to 2 m, with periods of the order of 12 to 18 seconds. However, they are dwarfed by the seastates observed during the intense tropical cyclones that form on the NWS from November to April and have significant wave heights that can exceed 15 m close the centre of the cyclone. Reproducing the resulting near-bed velocities which peak at 1-2 m/s in the laboratory is impossible without special flume designs, such as the O-tube.

Seabed sediments are traditionally classified according to their particle size, particle size distribution (or grading), angularity and origin. Each of these aspects influences the fundamental transport properties of the sediment, including threshold shear stress and volumetric transport rate. Most existing laboratory work on fluid-seabed-structure interaction has tended to focus on uniform silica sands having erosion properties that correlate well with median particle size (as indicated by the well-known Shields relationship (Shields, 1936). In contrast, the surficial sediment on the NWS consists mostly calcareous deposits, comprised of shell fragments and residue from marine fauna. Sediment size can range from gravel to silt (Baker et al., 2008) with high local variability reflecting the relic features from sea level lowstands (Hengesh et al., 2011). Because of these differences NWS calcareous and carbonate deposits can exhibit very different erosion properties to traditional silica sands. Mohr et al. (2013) has recently demonstrated this using the mini O-tube flume at UWA to conduct erosion testing of samples recovered from the NWS in addition to silica sands.

When a subsea structure (or a pipeline) is placed on the seabed, ambient current and wave velocities are modified around the structure which leads to a local increase in seabed shear stress. If the local sediment is mobile under the applied shear stress, net sediment transport or scour will occur.

Scour can have important effects on the stability of offshore structures. Undermining of foundations can lead to uneven settlement and instability of infrastructure. Alternatively, in the case of pipelines, scour can promote self-burial (see Fredsøe et al., 1988) and an associated increase in stability through soil resistance. This is particularly beneficial on the NWS of Australia, where stabilisation requirements can be significant to ensure stability in cyclonic conditions. However, self-burial can be unwanted for pipelines that are designed to buckle laterally under thermal loading (Borges Rodriguez et al., 2013). These contrasting requirements mean that more accurate predictions of self-burial are desirable.

3 O-TUBE CONCEPT Many different physical modelling facilities have been developed to study offshore structures at small scale, including (i) open channel flumes (with pumped current, wave paddles or driven trolleys), (ii) closed U-tube flumes, allowing oscillatory flow at the resonant frequency and (iii) oscillating water tunnels driven by a piston. Open channel flumes are limited to wave velocities below which wave breaking occurs. A driven trolley allows higher velocities to be achieved but is impractical for large regions of mobile bed. U-tubes allow higher velocities to be achieved, but with limited flexibility (due to the requirement to operate at or near resonance). Piston driven water tunnels offer more flexible control but are limited by the stroke of the piston.

These limitations prevent realistic near-seabed conditions representative of Australia’s NWS from being reproduced in any conventional flume. An alternative flume configuration, known as an O-tube, has been developed at the University of Western Australia. This flume comprises a horizontal fully enclosed circulating water channel, with a rectangular test section and an impeller-type pump driven by a motor (Figure 1). This arrangement has the relative advantages that (i) currents can be introduced easily, and (ii) wave velocities are limited only by the pump characteristics and not by wave breaking, resonance of the water mass or the stroke of a piston.

Three O-tubes have been constructed at the University of Western Australia. The key dimensions and performance characteristics of these flumes are given in Table 1 and the facilities are shown in Figures 1 – 3. The mini O-tube (MOT) was constructed first to prove the concept, followed quickly by the large O-tube (LOT). The small O-tube (SOT) was constructed recently using an improved design compared to the MOT. The different scales of O-tube are suited to different purposes. The LOT is capable of modelling small pipelines at full scale, with negligible blockage effects and 1:1 scale flow conditions. The MOT and SOT require less sediment to fill and nourish the working section compared with the LOT. This allows small scale tests and sediment-specific erosion testing to be undertaken using prototype sediments gathered from the field.

OCEAN-STRUCTURE-SEABED INTERACTION: O-TUBE MODELLING OF PIPELINE STABILITY WHITE ET AL.


Table 1: Key characteristics of UWA’s O-tube flumes.

Large O-tube (LOT) Small O-tube (SOT) Mini O-tube (MOT)

Year commissioned 2010 2014 2008

Working section dimensions

Length, L 17 m 3.0 m 2.0 m

Width, W 1.0 m 0.3 m 0.2 m

Height, H 1.4 m 0.45 m 0.3 m

Maximum steady current 3 m/s 4.5 m/s 1.5 m/s

Typical maximum oscillatory flow

Velocity, v 1 – 2.5 m/s 2 m/s 0.5 m/s

Period, T 5 – 13 s 6 s 6 s

(a) General arrangement (Draper et al., 2014a) (b) View from above (image courtesy of Joan Costa)

Figure 1: General arrangement of the large O-tube (LOT) at UWA’s Shenton Park campus.

4 O-TUBE MEASUREMENT AND CONTROL HARDWARE The LOT, SOT and MOT all use similar drive and control systems, varying only in scale and power. Flow is forced around the LOT with an impeller driven by a brushless 580 kW AC motor. The rotational speed of motor is controlled by a Variable Frequency Drive (VFD) and can be controlled from a desktop computer via a signal that is transmitted digitally over a local wireless network. The internal control software on the VFD includes safety interlocks that limit motor acceleration and rotation speed. The SOT and MOT use smaller drives operating on the same principles.

To reduce turbulence and turbulent length scales in the test section, a flow laminator is installed at each end. For the LOT, the laminator is composed of stainless steel tubes with diameter 50 mm, length 300 mm and wall thickness of 2 mm. Finer laminators are used on the MOT and SOT. Flow conditions including steady current, accelerating (decelerating) current, sinusoidal oscillatory flow, random oscillatory flow and any combination of above conditions can be generated. These flows are achieved by controlling the propeller to run at a steady, oscillation or irregular rotational speed. Detailed study of motor speed-flow speed relationships, flow asymmetry across the working section, turbulence intensity and secondary flows has been undertaken in all of the O-tubes (Mohr et al., 2014; An et al., 2013; Luo et al., 2012). Flow measurements within the O-tubes are made using triaxial acoustic Doppler velocimeters (ADVs).

An important component of the LOT facility is a 200 mm instrumented model pipe mounted on an actuator system to record the applied horizontal and vertical forces. The pipe is also equipped with a network of surface pressure cells to record the hydrodynamic load around the pipe circumference. The actuator system prevents model pipe movements in unrealistic degrees of freedom such as roll and yaw. The feedback system can provide neutral horizontal and vertical control, allowing the pipe free lateral movement in response to the natural balance between hydrodynamic loading and soil resistance, and free vertical movement as if acting under self-weight.

Data acquisition throughout the LOT laboratory is performed using the DIGIDAQ system developed at UWA (Gaudin et al., 2009). This system comprises of up to 8 independent 8-channel acquisition units that are connected to a base station in the Control Room via wired or wireless Ethernet. Each DIGIDAQ unit provides excitation, signal conditioning, analogue-to-digital conversion, data storage and real-time transmission of up to 8 instrument channels at a sampling rate of up to 10 kHz.


In addition to monitoring forces on a pipeline or structure, the O-tube flumes are equipped with various devices to monitor the seabed profile during experiments. Three different devices have been developed to achieve this: a laser scanner, a binocular infrared scanner, and a sonar device. An example seabed profile from testing of a pile group in steady current is shown in Figure 2.

(a) 200 mm diameter model pipe (b) Seabed scanning capability

Figure 2: O-tube hardware.

5 EXAMPLE OBSERVATIONS OF PIPELINE (IN)STABILITY Significant research efforts have been devoted to pipeline on-bottom stability and seabed mobility at UWA over the past five years, supported by the ARC and industry partners Woodside and Chevron through the STABLEpipe Joint Industry Project. In conventional design practice, pipeline stability assessments neglect the mobility of the seabed. The analysis instead focusses on simulating fluid-pipe and pipe-seabed interaction without including the fluid-seabed interaction that leads to scour and erosion processes (e.g. Zeitoun et al., 2009).

Model tests in the large O-tube have highlighted the strong influence of seabed mobility on pipeline stability. The processes of scour and self-burial have been observed and quantified in controlled conditions, leading to new calculations methods for design purposes. The mechanisms of scour and self-burial have some surprising consequences that go against the trends predicted by conventional ‘immobile seabed’ design methods. For example, on mobile sandy soils the general response is for the seabed to begin scouring before the storm is sufficiently intense to dislodge a pipe resting on the seabed surface. If the pipe is shallowly embedded, a scour tunnel opens beneath the pipe, widening into a scour hole that grows longitudinally. As the pipe sinks into this hole it becomes more stable because of the shielding from the flow and the increased soil resistance that results from the trench (Draper et al., 2014a).

However, if the pipe is initially too deeply embedded for a scour tunnel to open up, then this process is delayed or may not occur. Consequently, a pipeline that is installed at a deeper embedment can be less stable than a pipeline that is initially resting on the seabed at zero embedment. Two tests, detailed in Table 2, are used to demonstrate this effect. The only difference between the two tests was the initial embedment. The model pipe had the same specific gravity (SG) and was subjected to the same flow condition, which represented a growing cyclone with irregular wave cycles and a superimposed current.

Table 2: Example O-tube tests

Test Flow Simulated specific gravity, SG

Initial pipe embedment, e0/D

Stability outcome

1 Combined irregular flow 1.35 0.01 Pipe sank into scour hole

2 Combined irregular flow 1.35 0.15 Pipe was broken out

In Test 1, tunnel scour was initiated at 1915 s at a flow velocity of 0.48 m/s. After 800 seconds the model pipe started to sink into the scour hole. By the end of the test (after 4800 s), the model pipe had sunk to 0.6D embedment and survived the peak velocity of 1.2 m/s. In Test 2, the onset of scour was later, after 2640 s (at u = 0.6 m/s), due to the larger initial embedment. After 3113 s (u = 0.81 m/s), the model pipe broke out from the scour hole during a large wave cycle.



The onset of scour in Test 2 happened later than in Test 1 due to the larger initial embedment. For this reason, the development of the scour hole was postponed. When the growing storm reached a velocity of 0.8 m/s the model pipe broke out in Test 2. In contrast, the earlier onset of scour in Test 1 caused the self-burial process to begin sooner, allowing the pipe to attain sufficient hydrodynamic shielding and enhanced soil resistance to prevent breakout. Further information on these tests is presented by Cheng et al. (2014b).

Figure 3. The response of the model pipe and the seabed model during example Test 2.

6 APPLICATION TO PIPELINE DESIGNA wide range of parametric studies have been performed in the O-tube flumes to quantify the various mechanisms illustrated in Figure 3. Much of this research remains confidential to the industry sponsors, although some elements have been published and the general philosophy of the design approach can be described.

To distill the complex interactions shown in Figure 3 into a practical design process, the ocean-pipeline-seabed interaction processes can be split into a series of 2D and 3D mechanisms, which each form the building blocks of a design assessment. The two-dimensional seabed mobility behaviour can be divided into (i) onset of scour, (ii) rate of scour development, (iii) equilibrium scour depth (and cross-sectional profile) and (iv) backfilling process. O-tube results have been used to refine and calibrate assessment methods for these different aspects, supported by numerical simulations. A key feature of the work has been parametric studies which build on previous literature by focusing on (i) testing at large geometric scale, (ii) testing on silica and calcareous sediments and (iii) testing in non-stationary metocean conditions including, for example, the ‘build-up’ phase of a cyclonic storm.

Figure 4: Illustration of pipeline self-burial processes by (a) sagging and (b) sinking (Draper et al., 2014a).

The self-burial process is three-dimensional, as scour must propagate longitudinally along the pipeline from an initiation point, as well as vertically downwards. Also, the movement of the pipe at any location is dependent on the support from, and interaction with, adjacent lengths of pipe. An element of pipe will not drop directly into a scour hole as it opens. Instead, the pipe can self-bury through processes of sagging (Figure 4a) and sinking (Figure 4b). The sagging process is influenced by the rate of longitudinal scour progression and the bending stiffness of the pipe. Leckie


et al. (2014) have observed both sagging and sinking on the NWS using video survey data. Assessment methods for longitudinal scour have been derived from physical modelling in wider flumes (Cheng et al., 2009, 2014a; Wu and Chiew, 2012), although O-tube testing also provides information on 3D scour rates. The sinking process is dependent on the bearing capacity of the seabed at the shoulders of the pipe span over the scour hole.

One of the first programs of research undertaken in the O-tube flumes provided support for the life extension process of the North West Shelf Venture’s First Trunkline (1TL) running from the North Rankin A platform to the Burrup Peninsula (Jas et al., 2012). During a study performed to assess the condition of 1TL it was identified that within the limitations of existing design codes it was not possible to demonstrate that the pipeline satisfied on-bottom stability requirements. This was partly due to changes in the design storm since the pipeline was designed, and partly because the level of embedment was different to the installed condition. However, it was recognised that the limitations in the design codes overlooked potentially beneficial effects from seabed mobility. To allow such effects to be incorporated in the design assessment, it was decided that physical model testing would be performed to provide additional information specific to the conditions relevant to this pipeline.

A programme of O-tube tests was performed to identify the ocean-pipeline-seabed interactions under several realisations of the design storm. The model pipe was initially set up at various levels of embedment, including asymmetric profiles, to represent the range of conditions known to exist around 1TL in situ. These tests allowed the changing pipe embedment during the storm to be quantified. The resulting temporal variation in embedment and therefore hydrodynamic exposure and lateral pipe-soil resistance was used as input for 3D FE analyses of 1TL to assess the stability of the pipe under the current design storm. This method was effective in assessing the stability of 1TL, accounting for the mobile seabed environment (Jas et al., 2012).

The 1TL example was the first application of data from the O-tube to a field design situation. Since then, results from O-tube studies have been used to validate and refine the design of various pipelines and other subsea structures across the NWS. These include rock berms (An et al., 2014), concrete mattresses, erosion testing of project specific sediments (Mohr et al., 2013), secondary stabilisation (Zhao et al., 2014), pipelines on shallow sediment veneers (Draper et al., 2014b) and decommissioned pipelines and new umbilicals and flowlines.

Work has also been performed to create more generic guidelines to allow seabed mobility to be considered in design without recourse to project-specific physical modelling in the O-tube. This work has been performed within the STABLEpipe Joint Industry Project which is supported by Woodside and Chevron. This JIP has led to the development by UWA and Wood Group Kenny of a design guideline that is currently confidential to the JIP Sponsors. This Guideline has been verified by the Norwegian Verification Agency, DNV, and re-released in a revised form under a DNV cover.

The general philosophy devised for pipeline stability design, accounting for seabed mobility, is to perform a two stage assessment: seabed mobility then pipeline stability. Firstly, the effect of seabed mobility is calculated first, to determine the spatial variation in embedment along the pipeline accounting for scour. Secondly, a conventional stability assessment is performed. based on the as-scoured embedment profile, taking account of the load-sharing around pipeline spans at locations where scour has occurred.

To verify that the calculation methods devised from laboratory observations are applicable in field conditions, an additional stream of research activity has analysed observations of existing pipelines across the NWS. Survey data gathered at intervals over the life of the NWS pipeline network has been back-analysed, using the appropriate metocean and geotechnical data. This back-analysis work has involved the development of image analysis tools that allow reconstruction of the three-dimensional seabed topography from historic sonar profiler data. An example of the time-varying embedment of a pipeline on the NWS is shown in Figure 5, illustrating significant self-lowering over 4 years.

(a) Soon after laying (b) Four years later

Figure 5: Temporal changes in the seabed topography around a pipeline on the NWS (Leckie et al., 2014).



7 CONCLUSIONS Engineering problems involving ocean-structure-seabed interactions are sufficiently complex that physical model testing provides an invaluable research approach. The O-tubes developed at UWA have been motivated by a need to better design subsea pipelines on sandy seabeds, accounting for changes to pipeline embedment resulting from scour. Parametric studies in the O-tubes focussing on the mechanisms of scour have enabled pipeline lowering to be rationalised to a level that allows it to be incorporated into on-bottom stability design. In addition to this outcome, the O-tubes have also proven useful in model scale testing of rock berms, concrete mattresses, erosion testing of project specific soil samples and stability analysis of decommissioned pipelines and new umbilicals and flowlines. This work has provided valuable support to ongoing offshore engineering in Australia. Looking to the future, the combination of model scale laboratory experiments conducting in the O-tubes, together with numerical modelling and evidence fromfield data, is expected to provide a powerful approach to unlock conservatism in subsea design on mobile seabeds and to improve reliability and safety offshore.

8 ACKNOWLEDGEMENTS Development of the O-tube flumes at UWA were funded by Woodside, Chevron, the Australian Research Council and UWA Strategic Funds. We acknowledge this financial assistance, as well as the technical support provided by Tuarn Brown, Alex Duff, Mike McCarthy and John Breen. The work presented here forms part of the activities of the Centre for Offshore Foundation Systems (COFS), supported as a node of the ARC Centre of Excellence for Geotechnical Science and Engineering (CGSE). The first author is supported by Shell Australia. The second, fifth and sixth authors kindly acknowledge the support of the Lloyd’s Register Foundation. Lloyd’s Register Foundation helps to protect life and property by supporting engineering-related education, public engagement and the application of research.

9 REFERENCES An, H., Luo, C., Cheng, L., & White, D. (2013). ‘A new facility for studying ocean-structure-seabed interactions: The

O-tube’. Coastal Engineering, 82, 88-101. An, H., Cheng, L., Luo, C., Draper, S., White, D.J. and Jas, E. (2014) ‘Sediment Transport Behaviour in the Voids of a

Rock Berm’, 7th International Conference on Scour and Erosion, Perth, Australia. Baker C., Potter A., Tran M. and Heap A. D. (2008). 'Geomorphology and Sedimentology of the Northwest Marine

Region of Australia', Geoscience Australia, Canberra, Australia, pp. 220. Bonjean, D., Erbrich, C. T., & Zhang, J. (2008). ‘Pipeline flotation in liquefiable soil’. In Proc. Offshore Technology

Conference, Houston, Paper OTC (Vol. 19668). Borges Rodriguez, A., Bransby, M.F., Finnie, I.M.S., Low, H.E., White, D.J., (2013) ‘Changes in Pipeline Embedment

due to Sediment Mobility: Observations and Implications for Design’, Offshore Marine and Arctic Engineering Con-ference 2013, OMAE2013-11425.

Cheng, L., Yeow, K., Zhang, Z. and Teng, B. (2009), ‘Three-dimensional scour below offshore pipelines in steady currents’, Coastal Engineering, vol. 56, 577-590.

Cheng, L., Yeow, K., Zhang, Z. and Teng, B. (2014a), ‘3D scour below pipelines under waves and combined waves and currents’, Coastal Engineering, vol. 83, 137-149.

Cheng, L., An, H., Draper, S., Luo, H., Brown, T. and White, D. (2014b) ‘Invited Keynote: UWA’s O-tube facilities: physical modelling of fluid-structure-seabed interactions’. International Conference on Physical Modelling in Geotechnics, Perth, Australia, 2014.

Damgaard, J. S., & Palmer, A. C. (2001). ‘Pipeline stability on a mobile and liquefied seabed: A discussion of magnitudes and engineering implications’. In 20th Int. Conf. on Off-shore Mechanics and Artic Engineering.

Draper, S., An, H., Cheng, L, White, D.J., Luo, C. and Griffiths, T. (2014a) ‘Stability of subsea pipelines during large storms’ Philosophical Transactions of the Royal Society, Part A (In Press).

Draper, S., Cheng, L., Sun, W., An, H., White, D., Bransby, F., Borges-Rodriguez, A. and Griffiths, T. (2014b) ‘Pipeline scour and self-burial on a thin veneer of sand overlying rock due to steady currents’, 7th International Conference on Scour and Erosion, Perth, Australia.

Fredsøe, J., Hansen, E.A., Mao, Y. & Sumer, B.M., (1988). ‘Three-dimensional scour below pipelines’. Journal of Off-shore Mechanics and Arctic Engineering, Vol. 110, 373-379.

Gaudin C., White D.J., Boylan N., Breen J., Brown T. & De Catania S. 2009. ‘Development of a miniature high speed wireless data acquisition system for geotechnical centrifuges’. Proc. Int. Conf. on Physical Modelling in Geotechnics. Zurich, Switzerland, 229-234.

Hengesh, J., B. Whitney, and A. Royere, ‘A Tectonic Influence on Seafloor Stability along Australia’s North West Shelf’. (2011), International Offshore and Polar Engineering Conference: USA. p. 596-604.

Holloway, P. E. (1983). ‘Tides on the Australian North-west shelf’. Marine and Freshwater Research, 34(1), 213-230. Jas, E, O'Brien, D, Fricke, D, Gillen, A, Cheng, L., White, D., Palmer, A (2012), ‘Pipeline stability revisited’. Journal

of Pipeline Engineering, 4th Quarter, pp. 259-268.


Leckie, S.H., Draper, S., White, D.J. and Cheng, L. (2014) ‘Lifelong embedment and spanning of a pipeline on a mobile seabed’, Coastal Engineering (http://dx.doi.org/10.1016/j.coastaleng.2014.10.003).

Luo, C., An, H., Cheng, L., White, D. (2012), 'Calibration of UWA’s O-tube flume facility', 31st International Conference on Ocean, Offshore and Arctic Engineering, USA, CD, pp. 11pp

Mohr. H., Draper, S., & White, D. (2013) ‘Free field sediment mobility on Australia’s North West Shelf’ 32nd International Conference on Ocean, Offshore and Arctic Engineering, Nantes, France, OMAE2013-11490.

Mohr. H., Draper, S., Cheng, L., An, H., Zhang, Q. and White, D. (2014) ‘Hydrodynamics of a recirculating (O-tube) flume’ Under review.

Nielsen, A. W., Sumer, B. M., Fredsøe, J., & Christensen, E. D. (2011). ‘Sinking of armour layer around a cylinder exposed to a current’. Proceedings of the ICE-Maritime Engineering, 164(4), 159-172.

Shields, A. (1936) ‘Application of Similarity Principles and Turbulence Research to Bedload Movement (Englishtranslation of the original German manuscript). Publication no. 167. Hydrodynamics Laboratory, California Institute of Technology, 36 pp.

Sumer, M. & Fredsøe, J., (2002). ‘The mechanics of scour in the marine environment’, World Scientific. Van Gastel, P., Ivey, G. N., Meuleners, M. J., Antenucci, J. P., & Fringer, O. (2009). ‘The variability of the large-

amplitude internal wave field on the Australian’ North West Shelf. Continental Shelf Research, 29(11), 1373-1383.

Whitehouse, R. J., Harris, J. M., Sutherland, J., & Rees, J. (2011). ‘The nature of scour development and scour protection at offshore wind farm foundations’. Marine Pollution Bulletin, 62(1), 73-88.

Wu, Y., & Chiew, Y. M. (2013). ‘Mechanics of Three-Dimensional Pipeline Scour in Unidirectional Steady Current’. Journal of Pipeline Systems Engineering and Practice, vol. 4(1), 3-10.

Zeitoun, H. O., Tørnes, K., Li, J., Wong, S., Brevet, R., & Willcocks, J. (2009). ‘Advanced dynamic pipeline stability analysis’. In ASME 2009 28th International Conference on Ocean, Offshore and Arctic Engineering (pp. 661-673). American Society of Mechanical Engineers.

Zhao, F., Draper, S. and Cheng, L. (2014) ‘Experimental and Numerical Scour Investigation of a Backfilled Trench’, 7th International Conference on Scour and Erosion, Perth, Australia.

ocean-structure-seabed interaction o · pdf fileocean-structure-seabed interaction: o-tube...

Documents